diff --git a/cmake/cmake.define b/cmake/cmake.define index f55a9bdabc..bb0bd6500b 100644 --- a/cmake/cmake.define +++ b/cmake/cmake.define @@ -120,8 +120,14 @@ ELSE () SET(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -Wno-literal-suffix -Werror=return-type -fPIC -gdwarf-2 -fsanitize=address -fsanitize=undefined -fsanitize-recover=all -fsanitize=float-divide-by-zero -fsanitize=float-cast-overflow -fno-sanitize=shift-base -fno-sanitize=alignment -g3 -Wformat=0") MESSAGE(STATUS "Compile with Address Sanitizer!") ELSE () - SET(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -Werror -Werror=return-type -fPIC -gdwarf-2 -g3 -Wformat=2 -Wno-format-nonliteral -Wno-format-truncation -Wno-format-y2k") - SET(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -Wno-reserved-user-defined-literal -Wno-literal-suffix -Werror=return-type -fPIC -gdwarf-2 -g3 -Wformat=2 -Wno-format-nonliteral -Wno-format-truncation -Wno-format-y2k") + MESSAGE(STATUS "XXXXXXXXXXXXXX Clang/AppleClang" ${TD_DARWIN}) + IF (${TD_DARWIN}) + SET(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -Werror -Werror=return-type -fPIC -gdwarf-2 -g3 -Wformat=2 -Wno-format-nonliteral -Wno-format-y2k") + SET(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -Werror -Wno-reserved-user-defined-literal -Werror=return-type -fPIC -gdwarf-2 -g3 -Wformat=2 -Wno-format-nonliteral -Wno-format-y2k") + ELSE () + SET(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -Werror -Werror=return-type -fPIC -gdwarf-2 -g3 -Wformat=2 -Wno-format-nonliteral -Wno-format-truncation -Wno-format-y2k") + SET(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -Wno-reserved-user-defined-literal -Wno-literal-suffix -Werror=return-type -fPIC -gdwarf-2 -g3 -Wformat=2 -Wno-format-nonliteral -Wno-format-truncation -Wno-format-y2k") + ENDIF () ENDIF () # disable all assert diff --git a/contrib/CMakeLists.txt b/contrib/CMakeLists.txt index 185ebae51e..c1255d15bb 100644 --- a/contrib/CMakeLists.txt +++ b/contrib/CMakeLists.txt @@ -259,13 +259,26 @@ if(${BUILD_WITH_ROCKSDB}) option(WITH_BENCHMARK_TOOLS "" OFF) option(WITH_TOOLS "" OFF) option(WITH_LIBURING "" OFF) - + option(WITH_IOSTATS_CONTEXT "" OFF) + option(WITH_PERF_CONTEXT "" OFF) + option(FAIL_ON_WARNINGS "" OFF) + #option(WITH_JEMALLOC "" ON) option(ROCKSDB_BUILD_SHARED "Build shared versions of the RocksDB libraries" OFF) + IF (${TD_WINDOWS}) + option(WITH_MD_LIBRARY "build with MD" OFF) + set(SYSTEM_LIBS ${SYSTEM_LIBS} shlwapi.lib rpcrt4.lib) + endif(${TD_WINDOWS}) add_subdirectory(rocksdb EXCLUDE_FROM_ALL) target_include_directories( rocksdb PUBLIC $ ) + IF (${TD_DARWIN}) + target_compile_options( + rocksdb + PRIVATE -Wno-unused-private-field + ) + endif(${TD_DARWIN}) endif(${BUILD_WITH_ROCKSDB}) # lucene diff --git a/contrib/test/rocksdb/CMakeLists.txt b/contrib/test/rocksdb/CMakeLists.txt index b500e0a661..5ae1b0395b 100644 --- a/contrib/test/rocksdb/CMakeLists.txt +++ b/contrib/test/rocksdb/CMakeLists.txt @@ -1,6 +1,8 @@ +message("contrib test/rocksdb:" ${BUILD_DEPENDENCY_TESTS}) + add_executable(rocksdbTest "") target_sources(rocksdbTest PRIVATE "${CMAKE_CURRENT_SOURCE_DIR}/main.c" ) -target_link_libraries(rocksdbTest rocksdb) \ No newline at end of file +target_link_libraries(rocksdbTest rocksdb) diff --git a/contrib/test/rocksdb/main.c b/contrib/test/rocksdb/main.c index 33652386d7..bda717f953 100644 --- a/contrib/test/rocksdb/main.c +++ b/contrib/test/rocksdb/main.c @@ -116,9 +116,20 @@ int main(int argc, char const *argv[]) { rocksdb_options_set_create_if_missing(opt, 1); rocksdb_options_set_create_missing_column_families(opt, 1); - const char *cfName[] = {"default", "cf1"}; - int len = sizeof(cfName) / sizeof(cfName[0]); + // Read + rocksdb_readoptions_t *readoptions = rocksdb_readoptions_create(); + // rocksdb_readoptions_set_snapshot(readoptions, rocksdb_create_snapshot(db)); + int len = 1; + char buf[256] = {0}; + size_t vallen = 0; + char *val = rocksdb_get(db, readoptions, "key", 3, &vallen, &err); + snprintf(buf, vallen + 5, "val:%s", val); + printf("%ld %ld %s\n", strlen(val), vallen, buf); + char **cfName = calloc(len, sizeof(char *)); + for (int i = 0; i < len; i++) { + cfName[i] = "test"; + } const rocksdb_options_t **cfOpt = malloc(len * sizeof(rocksdb_options_t *)); for (int i = 0; i < len; i++) { cfOpt[i] = rocksdb_options_create_copy(opt); @@ -129,7 +140,7 @@ int main(int argc, char const *argv[]) { } rocksdb_column_family_handle_t **cfHandle = malloc(len * sizeof(rocksdb_column_family_handle_t *)); - db = rocksdb_open_column_families(opt, path, len, cfName, cfOpt, cfHandle, &err); + db = rocksdb_open_column_families(opt, path, len, (const char *const *)cfName, cfOpt, cfHandle, &err); { rocksdb_readoptions_t *rOpt = rocksdb_readoptions_create(); @@ -302,4 +313,4 @@ int main(int argc, char const *argv[]) { // rocksdb_close(db); return 0; -} \ No newline at end of file +} diff --git a/docs/en/05-get-started/03-package.md b/docs/en/05-get-started/03-package.md index 3282f600ac..159760ed21 100644 --- a/docs/en/05-get-started/03-package.md +++ b/docs/en/05-get-started/03-package.md @@ -208,6 +208,8 @@ The following `launchctl` commands can help you manage TDengine service: - Check TDengine Server status: `sudo launchctl list | grep taosd` +- Check TDengine Server status details: `launchctl print system/com.tdengine.taosd` + :::info - Please use `sudo` to run `launchctl` to manage _com.tdengine.taosd_ with administrator privileges. - The administrator privilege is required for service management to enhance security. diff --git a/docs/en/07-develop/03-insert-data/_rust_schemaless.mdx b/docs/en/07-develop/03-insert-data/_rust_schemaless.mdx new file mode 100644 index 0000000000..a26613bd15 --- /dev/null +++ b/docs/en/07-develop/03-insert-data/_rust_schemaless.mdx @@ -0,0 +1,3 @@ +```rust +{{#include docs/examples/rust/nativeexample/examples/schemaless_insert_line.rs}} +``` diff --git a/docs/en/07-develop/07-tmq.mdx b/docs/en/07-develop/07-tmq.mdx index 7465cc0a12..739eabd1cf 100644 --- a/docs/en/07-develop/07-tmq.mdx +++ b/docs/en/07-develop/07-tmq.mdx @@ -222,7 +222,7 @@ A database including one supertable and two subtables is created as follows: ```sql DROP DATABASE IF EXISTS tmqdb; -CREATE DATABASE tmqdb; +CREATE DATABASE tmqdb WAL_RETENTION_PERIOD 3600; CREATE TABLE tmqdb.stb (ts TIMESTAMP, c1 INT, c2 FLOAT, c3 VARCHAR(16)) TAGS(t1 INT, t3 VARCHAR(16)); CREATE TABLE tmqdb.ctb0 USING tmqdb.stb TAGS(0, "subtable0"); CREATE TABLE tmqdb.ctb1 USING tmqdb.stb TAGS(1, "subtable1"); diff --git a/docs/en/07-develop/09-udf.md b/docs/en/07-develop/09-udf.md index dc42743b51..a7753647e3 100644 --- a/docs/en/07-develop/09-udf.md +++ b/docs/en/07-develop/09-udf.md @@ -6,10 +6,12 @@ description: This document describes how to create user-defined functions (UDF), The built-in functions of TDengine may not be sufficient for the use cases of every application. In this case, you can define custom functions for use in TDengine queries. These are known as user-defined functions (UDF). A user-defined function takes one column of data or the result of a subquery as its input. -TDengine supports user-defined functions written in C or C++. This document describes the usage of user-defined functions. - User-defined functions can be scalar functions or aggregate functions. Scalar functions, such as `abs`, `sin`, and `concat`, output a value for every row of data. Aggregate functions, such as `avg` and `max` output one value for multiple rows of data. +TDengine supports user-defined functions written in C or Python. This document describes the usage of user-defined functions. + +## Implement a UDF in C + When you create a user-defined function, you must implement standard interface functions: - For scalar functions, implement the `scalarfn` interface function. - For aggregate functions, implement the `aggfn_start`, `aggfn`, and `aggfn_finish` interface functions. @@ -17,7 +19,7 @@ When you create a user-defined function, you must implement standard interface f There are strict naming conventions for these interface functions. The names of the start, finish, init, and destroy interfaces must be _start, _finish, _init, and _destroy, respectively. Replace `scalarfn`, `aggfn`, and `udf` with the name of your user-defined function. -## Implementing a Scalar Function +### Implementing a Scalar Function in C The implementation of a scalar function is described as follows: ```c #include "taos.h" @@ -49,7 +51,7 @@ int32_t scalarfn_destroy() { ``` Replace `scalarfn` with the name of your function. -## Implementing an Aggregate Function +### Implementing an Aggregate Function in C The implementation of an aggregate function is described as follows: ```c @@ -100,7 +102,7 @@ int32_t aggfn_destroy() { ``` Replace `aggfn` with the name of your function. -## Interface Functions +### UDF Interface Definition in C There are strict naming conventions for interface functions. The names of the start, finish, init, and destroy interfaces must be _start, _finish, _init, and _destroy, respectively. Replace `scalarfn`, `aggfn`, and `udf` with the name of your user-defined function. @@ -108,8 +110,7 @@ Interface functions return a value that indicates whether the operation was succ For information about the parameters for interface functions, see Data Model -### Interfaces for Scalar Functions - +#### Scalar Interface `int32_t scalarfn(SUdfDataBlock* inputDataBlock, SUdfColumn *resultColumn)` Replace `scalarfn` with the name of your function. This function performs scalar calculations on data blocks. You can configure a value through the parameters in the `resultColumn` structure. @@ -118,7 +119,7 @@ The parameters in the function are defined as follows: - inputDataBlock: The data block to input. - resultColumn: The column to output. The column to output. -### Interfaces for Aggregate Functions +#### Aggregate Interface `int32_t aggfn_start(SUdfInterBuf *interBuf)` @@ -126,7 +127,7 @@ The parameters in the function are defined as follows: `int32_t aggfn_finish(SUdfInterBuf* interBuf, SUdfInterBuf *result)` -Replace `aggfn` with the name of your function. In the function, aggfn_start is called to generate a result buffer. Data is then divided between multiple blocks, and aggfn is called on each block to update the result. Finally, aggfn_finish is called to generate final results from the intermediate results. The final result contains only one or zero data points. +Replace `aggfn` with the name of your function. In the function, aggfn_start is called to generate a result buffer. Data is then divided between multiple blocks, and the `aggfn` function is called on each block to update the result. Finally, aggfn_finish is called to generate the final results from the intermediate results. The final result contains only one or zero data points. The parameters in the function are defined as follows: - interBuf: The intermediate result buffer. @@ -135,15 +136,15 @@ The parameters in the function are defined as follows: - result: The final result. -### Initializing and Terminating User-Defined Functions +#### Initialization and Cleanup Interface `int32_t udf_init()` `int32_t udf_destroy()` -Replace `udf`with the name of your function. udf_init initializes the function. udf_destroy terminates the function. If it is not necessary to initialize your function, udf_init is not required. If it is not necessary to terminate your function, udf_destroy is not required. +Replace `udf` with the name of your function. udf_init initializes the function. udf_destroy terminates the function. If it is not necessary to initialize your function, udf_init is not required. If it is not necessary to terminate your function, udf_destroy is not required. -## Data Structure of User-Defined Functions +### Data Structures for UDF in C ```c typedef struct SUdfColumnMeta { int16_t type; @@ -193,7 +194,7 @@ typedef struct SUdfInterBuf { ``` The data structure is described as follows: -- The SUdfDataBlock block includes the number of rows (numOfRows) and number of columns (numCols). udfCols[i] (0 <= i <= numCols-1) indicates that each column is of type SUdfColumn. +- The SUdfDataBlock block includes the number of rows (numOfRows) and the number of columns (numCols). udfCols[i] (0 <= i <= numCols-1) indicates that each column is of type SUdfColumn. - SUdfColumn includes the definition of the data type of the column (colMeta) and the data in the column (colData). - The member definitions of SUdfColumnMeta are the same as the data type definitions in `taos.h`. - The data in SUdfColumnData can become longer. varLenCol indicates variable-length data, and fixLenCol indicates fixed-length data. @@ -201,9 +202,9 @@ The data structure is described as follows: Additional functions are defined in `taosudf.h` to make it easier to work with these structures. -## Compile UDF +### Compiling C UDF -To use your user-defined function in TDengine, first compile it to a dynamically linked library (DLL). +To use your user-defined function in TDengine, first, compile it to a shared library. For example, the sample UDF `bit_and.c` can be compiled into a DLL as follows: @@ -213,12 +214,9 @@ gcc -g -O0 -fPIC -shared bit_and.c -o libbitand.so The generated DLL file `libbitand.so` can now be used to implement your function. Note: GCC 7.5 or later is required. -## Manage and Use User-Defined Functions -After compiling your function into a DLL, you add it to TDengine. For more information, see [User-Defined Functions](../12-taos-sql/26-udf.md). +### UDF Sample Code in C -## Sample Code - -### Sample scalar function: [bit_and](https://github.com/taosdata/TDengine/blob/3.0/tests/script/sh/bit_and.c) +#### Scalar function: [bit_and](https://github.com/taosdata/TDengine/blob/3.0/tests/script/sh/bit_and.c) The bit_and function implements bitwise addition for multiple columns. If there is only one column, the column is returned. The bit_and function ignores null values. @@ -231,7 +229,7 @@ The bit_and function implements bitwise addition for multiple columns. If there -### Sample aggregate function: [l2norm](https://github.com/taosdata/TDengine/blob/3.0/tests/script/sh/l2norm.c) +#### Aggregate function 1: [l2norm](https://github.com/taosdata/TDengine/blob/3.0/tests/script/sh/l2norm.c) The l2norm function finds the second-order norm for all data in the input column. This squares the values, takes a cumulative sum, and finds the square root. @@ -243,3 +241,151 @@ The l2norm function finds the second-order norm for all data in the input column ``` + +#### Aggregate function 2: [max_vol](https://github.com/taosdata/TDengine/blob/3.0/tests/script/sh/max_vol.c) + +The max_vol function returns a string concatenating the deviceId column, the row number and column number of the maximum voltage and the maximum voltage given several voltage columns as input. + +Create Table: +```bash +create table battery(ts timestamp, vol1 float, vol2 float, vol3 float, deviceId varchar(16)); +``` +Create the UDF: +```bash +create aggregate function max_vol as '/root/udf/libmaxvol.so' outputtype binary(64) bufsize 10240 language 'C'; +``` +Use the UDF in the query： +```bash +select max_vol(vol1,vol2,vol3,deviceid) from battery; +``` + +
+max_vol.c + +```c +{{#include tests/script/sh/max_vol.c}} +``` + +
+ +## Implement a UDF in Python + +Implement the specified interface functions when implementing a UDF in Python. +- implement `process` function for the scalar UDF。 +- implement `start`, `reduce`, `finish` for the aggregate UDF。 +- implement `init` for initialization and `destroy` for termination。 + +### Implement a Scalar UDF in Python + +The implementation of a scalar UDF is described as follows: + +```Python +def init(): + # initialization +def destroy(): + # destroy +def process(input: datablock) -> tuple[output_type]: + # process input datablock, + # datablock.data(row, col) is to access the python object in location(row,col) + # return tuple object consisted of object of type outputtype +``` + +### Implement an Aggregate UDF in Python + +The implementation of an aggregate function is described as follows: + +```Python +def init(): + #initialization +def destroy(): + #destroy +def start() -> bytes: + #return serialize(init_state) +def reduce(inputs: datablock, buf: bytes) -> bytes + # deserialize buf to state + # reduce the inputs and state into new_state. + # use inputs.data(i,j) to access python ojbect of location(i,j) + # serialize new_state into new_state_bytes + return new_state_bytes +def finish(buf: bytes) -> output_type: + #return obj of type outputtype +``` + +### Python UDF Interface Definition + +#### Scalar interface +```Python +def process(input: datablock) -> tuple[output_type]: +``` +- `input` is a data block two-dimension matrix-like object, of which method `data(row, col)` returns the Python object located at location (`row`, `col`) +- return a Python tuple object, of which each item is a Python object of type `output_type` + +#### Aggregate Interface +```Python +def start() -> bytes: +def reduce(input: datablock, buf: bytes) -> bytes +def finish(buf: bytes) -> output_type: +``` + +- first `start()` is called to return the initial result in type `bytes` +- then the input data are divided into multiple data blocks and for each block `input`, `reduce` is called with the data block `input` and the current result `buf` bytes and generates a new intermediate result buffer. +- finally, the `finish` function is called on the intermediate result `buf` and outputs 0 or 1 data of type `output_type` + + +#### Initialization and Cleanup Interface +```Python +def init() +def destroy() +``` +Implement `init` for initialization and `destroy` for termination. + +### Data Mapping between TDengine SQL and Python UDF + +The following table describes the mapping between TDengine SQL data type and Python UDF Data Type. The `NULL` value of all TDengine SQL types is mapped to the `None` value in Python. + +| **TDengine SQL Data Type** | **Python Data Type** | +| :-----------------------: | ------------ | +|TINYINT / SMALLINT / INT / BIGINT | int | +|TINYINT UNSIGNED / SMALLINT UNSIGNED / INT UNSIGNED / BIGINT UNSIGNED | int | +|FLOAT / DOUBLE | float | +|BOOL | bool | +|BINARY / VARCHAR / NCHAR | bytes| +|TIMESTAMP | int | +|JSON and other types | Not Supported | + +### Installing Python UDF +1. Install Python package `taospyudf` that executes Python UDF +```bash +sudo pip install taospyudf +ldconfig +``` +2. If PYTHONPATH is needed to find Python packages when the Python UDF executes, include the PYTHONPATH contents into the udfdLdLibPath variable of the taos.cfg configuration file + +### Python UDF Sample Code +#### Scalar Function [pybitand](https://github.com/taosdata/TDengine/blob/3.0/tests/script/sh/pybitand.py) + +The `pybitand` function implements bitwise addition for multiple columns. If there is only one column, the column is returned. The `pybitand` function ignores null values. + +
+pybitand.py + +```Python +{{#include tests/script/sh/pybitand.py}} +``` + +
+ +#### Aggregate Function [pyl2norm](https://github.com/taosdata/TDengine/blob/3.0/tests/script/sh/pyl2norm.py) + +The `pyl2norm` function finds the second-order norm for all data in the input column. This squares the values, takes a cumulative sum, and finds the square root. +
+pyl2norm.py + +```c +{{#include tests/script/sh/pyl2norm.py}} +``` + +
+ +## Manage and Use UDF +You need to add UDF to TDengine before using it in SQL queries. For more information about how to manage UDF and how to invoke UDF, please see [Manage and Use UDF](../12-taos-sql/26-udf.md). diff --git a/docs/en/12-taos-sql/02-database.md b/docs/en/12-taos-sql/02-database.md index ec007d6830..8c579622c9 100644 --- a/docs/en/12-taos-sql/02-database.md +++ b/docs/en/12-taos-sql/02-database.md @@ -72,8 +72,8 @@ database_option: { - 0: The database can contain multiple supertables. - 1: The database can contain only one supertable. - STT_TRIGGER: specifies the number of file merges triggered by flushed files. The default is 8, ranging from 1 to 16. For high-frequency scenarios with few tables, it is recommended to use the default configuration or a smaller value for this parameter; For multi-table low-frequency scenarios, it is recommended to configure this parameter with a larger value. -- TABLE_PREFIX：The prefix length in the table name that is ignored when distributing table to vnode based on table name. -- TABLE_SUFFIX：The suffix length in the table name that is ignored when distributing table to vnode based on table name. +- TABLE_PREFIX： The prefix in the table name that is ignored when distributing a table to a vgroup when it's a positive number, or only the prefix is used when distributing a table to a vgroup, the default value is 0; For example, if the table name v30001, then "0001" is used if TSDB_PREFIX is set to 2 but "v3" is used if TSDB_PREFIX is set to -2; It can help you to control the distribution of tables. +- TABLE_SUFFIX：The suffix in the table name that is ignored when distributing a table to a vgroup when it's a positive number, or only the suffix is used when distributing a table to a vgroup, the default value is 0; For example, if the table name v30001, then "v300" is used if TSDB_SUFFIX is set to 2 but "01" is used if TSDB_SUFFIX is set to -2; It can help you to control the distribution of tables. - TSDB_PAGESIZE: The page size of the data storage engine in a vnode. The unit is KB. The default is 4 KB. The range is 1 to 16384, that is, 1 KB to 16 MB. - WAL_RETENTION_PERIOD: specifies the maximum time of which WAL files are to be kept for consumption. This parameter is used for data subscription. Enter a time in seconds. The default value 0. A value of 0 indicates that WAL files are not required to keep for consumption. Alter it with a proper value at first to create topics. - WAL_RETENTION_SIZE: specifies the maximum total size of which WAL files are to be kept for consumption. This parameter is used for data subscription. Enter a size in KB. The default value is 0. A value of 0 indicates that the total size of WAL files to keep for consumption has no upper limit. diff --git a/docs/en/12-taos-sql/07-tag-index.md b/docs/en/12-taos-sql/07-tag-index.md index cb2a61d3e8..af1d0a352e 100644 --- a/docs/en/12-taos-sql/07-tag-index.md +++ b/docs/en/12-taos-sql/07-tag-index.md @@ -6,7 +6,7 @@ description: Use Tag Index to Improve Query Performance ## Introduction -Prior to TDengine 3.0.3.0 (excluded)，only one index is created by default on the first tag of each super talbe, but it's not allowed to dynamically create index on any other tags. From version 3.0.30, you can dynamically create index on any tag of any type. The index created automatically by TDengine is still valid. Query performance can benefit from indexes if you use properly. +Prior to TDengine 3.0.3.0 (excluded)，only one index is created by default on the first tag of each super table, but it's not allowed to dynamically create index on any other tags. From version 3.0.30, you can dynamically create index on any tag of any type. The index created automatically by TDengine is still valid. Query performance can benefit from indexes if you use properly. ## Syntax @@ -48,4 +48,4 @@ You can also add filter conditions to limit the results. 6. You can' create index on a normal table or a child table. -7. If the unique values of a tag column are too few, it's better not to create index on such tag columns, the benefit would be very small. \ No newline at end of file +7. If the unique values of a tag column are too few, it's better not to create index on such tag columns, the benefit would be very small. diff --git a/docs/en/12-taos-sql/10-function.md b/docs/en/12-taos-sql/10-function.md index 5c1a833e05..0b5f9103c7 100644 --- a/docs/en/12-taos-sql/10-function.md +++ b/docs/en/12-taos-sql/10-function.md @@ -5,9 +5,9 @@ description: This document describes the standard SQL functions available in TDe toc_max_heading_level: 4 --- -## Single Row Functions +## Scalar Functions -Single row functions return a result for each row. +Scalar functions return one result for each row. ### Mathematical Functions diff --git a/docs/en/12-taos-sql/14-stream.md b/docs/en/12-taos-sql/14-stream.md index b8f6c3a163..43c49c03cd 100644 --- a/docs/en/12-taos-sql/14-stream.md +++ b/docs/en/12-taos-sql/14-stream.md @@ -13,8 +13,11 @@ Because stream processing is built in to TDengine, you are no longer reliant on ```sql CREATE STREAM [IF NOT EXISTS] stream_name [stream_options] INTO stb_name SUBTABLE(expression) AS subquery stream_options: { - TRIGGER [AT_ONCE | WINDOW_CLOSE | MAX_DELAY time] - WATERMARK time + TRIGGER [AT_ONCE | WINDOW_CLOSE | MAX_DELAY time] + WATERMARK time + IGNORE EXPIRED [0|1] + DELETE_MARK time + FILL_HISTORY [0|1] } ``` @@ -141,3 +144,27 @@ The data in expired windows is tagged as expired. TDengine stream processing pro 2. Recalculate the data. In this method, all data in the window is reobtained from the database and recalculated. The latest results are then returned. In both of these methods, configuring the watermark is essential for obtaining accurate results (if expired data is dropped) and avoiding repeated triggers that affect system performance (if expired data is recalculated). + +## Supported functions + +All [scalar functions](../function/#scalar-functions) are available in stream processing. All [Aggregate functions](../function/#aggregate-functions) and [Selection functions](../function/#selection-functions) are available in stream processing, except the followings: + - [leastsquares](../function/#leastsquares) + - [percentile](../function/#percentile) + - [top](../function/#top) + - [bottom](../function/#bottom) + - [elapsed](../function/#elapsed) + - [interp](../function/#interp) + - [derivative](../function/#derivative) + - [irate](../function/#irate) + - [twa](../function/#twa) + - [histogram](../function/#histogram) + - [diff](../function/#diff) + - [statecount](../function/#statecount) + - [stateduration](../function/#stateduration) + - [csum](../function/#csum) + - [mavg](../function/#mavg) + - [sample](../function/#sample) + - [tail](../function/#tail) + - [unique](../function/#unique) + - [mode](../function/#mode) + diff --git a/docs/en/12-taos-sql/22-meta.md b/docs/en/12-taos-sql/22-meta.md index 81284aeaed..e63f682761 100644 --- a/docs/en/12-taos-sql/22-meta.md +++ b/docs/en/12-taos-sql/22-meta.md @@ -120,6 +120,9 @@ Provides information about user-defined functions. | 5 | create_time | TIMESTAMP | Creation time | | 6 | code_len | INT | Length of the source code | | 7 | bufsize | INT | Buffer size | +| 8 | func_language | BINARY(31) | UDF programming language | +| 9 | func_body | BINARY(16384) | UDF function body | +| 10 | func_version | INT | UDF function version. starting from 0. Increasing by 1 each time it is updated| ## INS_INDEXES diff --git a/docs/en/12-taos-sql/26-udf.md b/docs/en/12-taos-sql/26-udf.md index cb64873705..c4d6d4fca4 100644 --- a/docs/en/12-taos-sql/26-udf.md +++ b/docs/en/12-taos-sql/26-udf.md @@ -7,17 +7,18 @@ description: This document describes the SQL statements related to user-defined You can create user-defined functions and import them into TDengine. ## Create UDF -SQL command can be executed on the host where the generated UDF DLL resides to load the UDF DLL into TDengine. This operation cannot be done through REST interface or web console. Once created, any client of the current TDengine can use these UDF functions in their SQL commands. UDF are stored in the management node of TDengine. The UDFs loaded in TDengine would be still available after TDengine is restarted. +SQL command can be executed on the host where the generated UDF DLL resides to load the UDF DLL into TDengine. This operation cannot be done through REST interface or web console. Once created, any client of the current TDengine can use these UDF functions in their SQL commands. UDF is stored in the management node of TDengine. The UDFs loaded in TDengine would be still available after TDengine is restarted. When creating UDF, the type of UDF, i.e. a scalar function or aggregate function must be specified. If the specified type is wrong, the SQL statements using the function would fail with errors. The input data type and output data type must be consistent with the UDF definition. - Create Scalar Function ```sql -CREATE FUNCTION function_name AS library_path OUTPUTTYPE output_type; +CREATE [OR REPLACE] FUNCTION function_name AS library_path OUTPUTTYPE output_type [LANGUAGE 'C|Python']; ``` - - - function_name: The scalar function name to be used in SQL statement which must be consistent with the UDF name and is also the name of the compiled DLL (.so file). - - library_path: The absolute path of the DLL file including the name of the shared object file (.so). The path must be quoted with single or double quotes. + - OR REPLACE: if the UDF exists, the UDF properties are modified + - function_name: The scalar function name to be used in the SQL statement + - LANGUAGE 'C|Python': the programming language of UDF. Now C or Python is supported. If this clause is omitted, C is assumed as the programming language. + - library_path: For C programming language, The absolute path of the DLL file including the name of the shared object file (.so). For Python programming language, the absolute path of the Python UDF script. The path must be quoted with single or double quotes. - output_type: The data type of the results of the UDF. For example, the following SQL statement can be used to create a UDF from `libbitand.so`. @@ -25,14 +26,20 @@ CREATE FUNCTION function_name AS library_path OUTPUTTYPE output_type; ```sql CREATE FUNCTION bit_and AS "/home/taos/udf_example/libbitand.so" OUTPUTTYPE INT; ``` + For Example, the following SQL statement can be used to modify the existing function `bit_and`. The OUTPUT type is changed to BIGINT and the programming language is changed to Python. + + ```sql + CREATE OR REPLACE FUNCTION bit_and AS "/home/taos/udf_example/bit_and.py" OUTPUTTYPE BIGINT LANGUAGE 'Python'; + ``` - Create Aggregate Function ```sql CREATE AGGREGATE FUNCTION function_name AS library_path OUTPUTTYPE output_type [ BUFSIZE buffer_size ]; ``` - - - function_name: The aggregate function name to be used in SQL statement which must be consistent with the udfNormalFunc name and is also the name of the compiled DLL (.so file). - - library_path: The absolute path of the DLL file including the name of the shared object file (.so). The path must be quoted with single or double quotes. + - OR REPLACE: if the UDF exists, the UDF properties are modified + - function_name: The aggregate function name to be used in the SQL statement + - LANGUAGE 'C|Python': the programming language of the UDF. Now C or Python is supported. If this clause is omitted, C is assumed as the programming language. + - library_path: For C programming language, The absolute path of the DLL file including the name of the shared object file (.so). For Python programming language, the absolute path of the Python UDF script. The path must be quoted with single or double quotes. - output_type: The output data type, the value is the literal string of the supported TDengine data type. - buffer_size: The size of the intermediate buffer in bytes. This parameter is optional. @@ -41,6 +48,11 @@ CREATE AGGREGATE FUNCTION function_name AS library_path OUTPUTTYPE output_type [ ```sql CREATE AGGREGATE FUNCTION l2norm AS "/home/taos/udf_example/libl2norm.so" OUTPUTTYPE DOUBLE bufsize 8; ``` + For example, the following SQL statement modifies the buffer size of existing UDF `l2norm` to 64 + ```sql + CREATE AGGREGATE FUNCTION l2norm AS "/home/taos/udf_example/libl2norm.so" OUTPUTTYPE DOUBLE bufsize 64; + ``` + For more information about user-defined functions, see [User-Defined Functions](/develop/udf). ## Manage UDF @@ -61,9 +73,9 @@ SHOW FUNCTIONS; ## Call UDF -The function name specified when creating UDF can be used directly in SQL statements, just like builtin functions. For example: +The function name specified when creating UDF can be used directly in SQL statements, just like built-in functions. For example: ```sql SELECT bit_and(c1,c2) FROM table; ``` -The above SQL statement invokes function X for column c1 and c2 on table. You can use query keywords like WHERE with user-defined functions. +The above SQL statement invokes function X for columns c1 and c2 on the table. You can use query keywords like WHERE with user-defined functions. diff --git a/docs/en/12-taos-sql/29-changes.md b/docs/en/12-taos-sql/29-changes.md index f4606f263f..086aee59fe 100644 --- a/docs/en/12-taos-sql/29-changes.md +++ b/docs/en/12-taos-sql/29-changes.md @@ -27,7 +27,7 @@ The following data types can be used in the schema for standard tables. | - | :------- | :-------- | :------- | | 1 | ALTER ACCOUNT | Deprecated| This Enterprise Edition-only statement has been removed. It returns the error "This statement is no longer supported." | 2 | ALTER ALL DNODES | Added | Modifies the configuration of all dnodes. -| 3 | ALTER DATABASE | Modified | Deprecated
  • QUORUM: Specified the required number of confirmations. TDengine 3.0 provides strict consistency by default and doesn't allow to change to weak consitency.
  • BLOCKS: Specified the memory blocks used by each vnode. BUFFER is now used to specify the size of the write cache pool for each vnode.
  • UPDATE: Specified whether update operations were supported. All databases now support updating data in certain columns.
  • CACHELAST: Specified how to cache the newest row of data. CACHEMODEL now replaces CACHELAST.
  • COMP: Cannot be modified.
    Added
  • CACHEMODEL: Specifies whether to cache the latest subtable data.
  • CACHESIZE: Specifies the size of the cache for the newest subtable data.
  • WAL_FSYNC_PERIOD: Replaces the FSYNC parameter.
  • WAL_LEVEL: Replaces the WAL parameter.
  • WAL_RETENTION_PERIOD: specifies the time after which WAL files are deleted. This parameter is used for data subscription.
  • WAL_RETENTION_SIZE: specifies the size at which WAL files are deleted. This parameter is used for data subscription.
    Modified
  • REPLICA: Cannot be modified.
  • KEEP: Now supports units.
+| 3 | ALTER DATABASE | Modified | Deprecated
  • QUORUM: Specified the required number of confirmations. TDengine 3.0 provides strict consistency by default and doesn't allow to change to weak consistency.
  • BLOCKS: Specified the memory blocks used by each vnode. BUFFER is now used to specify the size of the write cache pool for each vnode.
  • UPDATE: Specified whether update operations were supported. All databases now support updating data in certain columns.
  • CACHELAST: Specified how to cache the newest row of data. CACHEMODEL now replaces CACHELAST.
  • COMP: Cannot be modified.
    Added
  • CACHEMODEL: Specifies whether to cache the latest subtable data.
  • CACHESIZE: Specifies the size of the cache for the newest subtable data.
  • WAL_FSYNC_PERIOD: Replaces the FSYNC parameter.
  • WAL_LEVEL: Replaces the WAL parameter.
  • WAL_RETENTION_PERIOD: specifies the time after which WAL files are deleted. This parameter is used for data subscription.
  • WAL_RETENTION_SIZE: specifies the size at which WAL files are deleted. This parameter is used for data subscription.
    Modified
  • REPLICA: Cannot be modified.
  • KEEP: Now supports units.
| 4 | ALTER STABLE | Modified | Deprecated
  • CHANGE TAG: Modified the name of a tag. Replaced by RENAME TAG.
    Added
  • RENAME TAG: Replaces CHANGE TAG.
  • COMMENT: Specifies comments for a supertable.
| 5 | ALTER TABLE | Modified | Deprecated
  • CHANGE TAG: Modified the name of a tag. Replaced by RENAME TAG.
    Added
  • RENAME TAG: Replaces CHANGE TAG.
  • COMMENT: Specifies comments for a standard table.
  • TTL: Specifies the time-to-live for a standard table.
| 6 | ALTER USER | Modified | Deprecated
  • PRIVILEGE: Specified user permissions. Replaced by GRANT and REVOKE.
    Added
  • ENABLE: Enables or disables a user.
  • SYSINFO: Specifies whether a user can query system information.
diff --git a/docs/en/13-operation/10-monitor.md b/docs/en/13-operation/10-monitor.md index 346b874059..19107240bf 100644 --- a/docs/en/13-operation/10-monitor.md +++ b/docs/en/13-operation/10-monitor.md @@ -42,3 +42,304 @@ An existing Grafana Notification Channel can be specified with parameter `-E`, t Launch `TDinsight.sh` with the command above and restart Grafana, then open Dashboard `http://localhost:3000/d/tdinsight`. For more use cases and restrictions please refer to [TDinsight](/reference/tdinsight/). + +## log database + +The data of tdinsight dashboard is stored in `log` database (default. You can change it in taoskeeper's config file. For more infrmation, please reference to [taoskeeper document](/reference/taosKeeper)). The taoskeeper will create log database on taoskeeper startup. + +### cluster\_info table + +`cluster_info` table contains cluster information records. + +|field|type|is\_tag|comment| +|:----|:---|:-----|:------| +|ts|TIMESTAMP||timestamp| +|first\_ep|VARCHAR||first ep of cluster| +|first\_ep\_dnode\_id|INT||dnode id or first\_ep| +|version|VARCHAR||tdengine version. such as: 3.0.4.0| +|master\_uptime|FLOAT||days of master's uptime| +|monitor\_interval|INT||monitor interval in second| +|dbs\_total|INT||total number of databases in cluster| +|tbs\_total|BIGINT||total number of tables in cluster| +|stbs\_total|INT||total number of stables in cluster| +|dnodes\_total|INT||total number of dnodes in cluster| +|dnodes\_alive|INT||total number of dnodes in ready state| +|mnodes\_total|INT||total number of mnodes in cluster| +|mnodes\_alive|INT||total number of mnodes in ready state| +|vgroups\_total|INT||total number of vgroups in cluster| +|vgroups\_alive|INT||total number of vgroups in ready state| +|vnodes\_total|INT||total number of vnode in cluster| +|vnodes\_alive|INT||total number of vnode in ready state| +|connections\_total|INT||total number of connections to cluster| +|topics\_total|INT||total number of topics in cluster| +|streams\_total|INT||total number of streams in cluster| +|protocol|INT||protocol version| +|cluster\_id|NCHAR|TAG|cluster id| + +### d\_info table + +`d_info` table contains dnodes information records. + +|field|type|is\_tag|comment| +|:----|:---|:-----|:------| +|ts|TIMESTAMP||timestamp| +|status|VARCHAR||dnode status| +|dnode\_ep|NCHAR|TAG|dnode endpoint| +|cluster\_id|NCHAR|TAG|cluster id| + +### m\_info table + +`m_info` table contains mnode information records. + +|field|type|is\_tag|comment| +|:----|:---|:-----|:------| +|ts|TIMESTAMP||timestamp| +|role|VARCHAR||the role of mnode. leader or follower| +|mnode\_id|INT|TAG|master node id| +|mnode\_ep|NCHAR|TAG|master node endpoint| +|cluster\_id|NCHAR|TAG|cluster id| + +### dnodes\_info table + +`dnodes_info` table contains dnodes information records. + +|field|type|is\_tag|comment| +|:----|:---|:-----|:------| +|ts|TIMESTAMP||timestamp| +|uptime|FLOAT||dnode uptime| +|cpu\_engine|FLOAT||cpu usage of tdengine. read from `/proc//stat`| +|cpu\_system|FLOAT||cpu usage of server. read from `/proc/stat`| +|cpu\_cores|FLOAT||cpu cores of server| +|mem\_engine|INT||memory usage of tdengine. read from `/proc//status`| +|mem\_system|INT||available memory on the server| +|mem\_total|INT||total memory of server in `KB`| +|disk\_engine|INT||| +|disk\_used|BIGINT||usage of data dir in `bytes`| +|disk\_total|BIGINT||the capacity of data dir in `bytes`| +|net\_in|FLOAT||network throughput rate in kb/s. read from `/proc/net/dev`| +|net\_out|FLOAT||network throughput rate in kb/s. read from `/proc/net/dev`| +|io\_read|FLOAT||io throughput rate in kb/s. read from `/proc//io`| +|io\_write|FLOAT||io throughput rate in kb/s. read from `/proc//io`| +|io\_read\_disk|FLOAT||io throughput rate of disk in kb/s. read from `/proc//io`| +|io\_write\_disk|FLOAT||io throughput rate of disk in kb/s. read from `/proc//io`| +|req\_select|INT||number of select queries received per dnode| +|req\_select\_rate|FLOAT||number of select queries received per dnode divided by monitor interval.| +|req\_insert|INT||number of insert queries received per dnode| +|req\_insert\_success|INT||number of successfully insert queries received per dnode| +|req\_insert\_rate|FLOAT||number of insert queries received per dnode divided by monitor interval| +|req\_insert\_batch|INT||number of batch insertions| +|req\_insert\_batch\_success|INT||number of successful batch insertions| +|req\_insert\_batch\_rate|FLOAT||number of batch insertions divided by monitor interval| +|errors|INT||dnode errors| +|vnodes\_num|INT||number of vnodes per dnode| +|masters|INT||number of master vnodes| +|has\_mnode|INT||if the dnode has mnode| +|has\_qnode|INT||if the dnode has qnode| +|has\_snode|INT||if the dnode has snode| +|has\_bnode|INT||if the dnode has bnode| +|dnode\_id|INT|TAG|dnode id| +|dnode\_ep|NCHAR|TAG|dnode endpoint| +|cluster\_id|NCHAR|TAG|cluster id| + +### data\_dir table + +`data_dir` table contains data directory information records. + +|field|type|is\_tag|comment| +|:----|:---|:-----|:------| +|ts|TIMESTAMP||timestamp| +|name|NCHAR||data directory. default is `/var/lib/taos`| +|level|INT||level for multi-level storage| +|avail|BIGINT||available space for data directory| +|used|BIGINT||used space for data directory| +|total|BIGINT||total space for data directory| +|dnode\_id|INT|TAG|dnode id| +|dnode\_ep|NCHAR|TAG|dnode endpoint| +|cluster\_id|NCHAR|TAG|cluster id| + +### log\_dir table + +`log_dir` table contains log directory information records. + +|field|type|is\_tag|comment| +|:----|:---|:-----|:------| +|ts|TIMESTAMP||timestamp| +|name|NCHAR||log directory. default is `/var/log/taos/`| +|avail|BIGINT||available space for log directory| +|used|BIGINT||used space for data directory| +|total|BIGINT||total space for data directory| +|dnode\_id|INT|TAG|dnode id| +|dnode\_ep|NCHAR|TAG|dnode endpoint| +|cluster\_id|NCHAR|TAG|cluster id| + +### temp\_dir table + +`temp_dir` table contains temp dir information records. + +|field|type|is\_tag|comment| +|:----|:---|:-----|:------| +|ts|TIMESTAMP||timestamp| +|name|NCHAR||temp directory. default is `/tmp/`| +|avail|BIGINT||available space for temp directory| +|used|BIGINT||used space for temp directory| +|total|BIGINT||total space for temp directory| +|dnode\_id|INT|TAG|dnode id| +|dnode\_ep|NCHAR|TAG|dnode endpoint| +|cluster\_id|NCHAR|TAG|cluster id| + +### vgroups\_info table + +`vgroups_info` table contains vgroups information records. + +|field|type|is\_tag|comment| +|:----|:---|:-----|:------| +|ts|TIMESTAMP||timestamp| +|vgroup\_id|INT||vgroup id| +|database\_name|VARCHAR||database for the vgroup| +|tables\_num|BIGINT||number of tables per vgroup| +|status|VARCHAR||status| +|dnode\_id|INT|TAG|dnode id| +|dnode\_ep|NCHAR|TAG|dnode endpoint| +|cluster\_id|NCHAR|TAG|cluster id| + +### vnodes\_role table + +`vnodes_role` table contains vnode role information records. + +|field|type|is\_tag|comment| +|:----|:---|:-----|:------| +|ts|TIMESTAMP||timestamp| +|vnode\_role|VARCHAR||role. leader or follower| +|dnode\_id|INT|TAG|dnode id| +|dnode\_ep|NCHAR|TAG|dnode endpoint| +|cluster\_id|NCHAR|TAG|cluster id| + +### logs table + +`logs` table contains login information records. + +|field|type|is\_tag|comment| +|:----|:---|:-----|:------| +|ts|TIMESTAMP||timestamp| +|level|VARCHAR||log level| +|content|NCHAR||log content| +|dnode\_id|INT|TAG|dnode id| +|dnode\_ep|NCHAR|TAG|dnode endpoint| +|cluster\_id|NCHAR|TAG|cluster id| + +### log\_summary table + +`log_summary` table contains log summary information records. + +|field|type|is\_tag|comment| +|:----|:---|:-----|:------| +|ts|TIMESTAMP||timestamp| +|error|INT||error count| +|info|INT||info count| +|debug|INT||debug count| +|trace|INT||trace count| +|dnode\_id|INT|TAG|dnode id| +|dnode\_ep|NCHAR|TAG|dnode endpoint| +|cluster\_id|NCHAR|TAG|cluster id| + +### grants\_info table + +`grants_info` table contains grants information records. + +|field|type|is\_tag|comment| +|:----|:---|:-----|:------| +|ts|TIMESTAMP||timestamp| +|expire\_time|BIGINT||time until grants expire in seconds| +|timeseries\_used|BIGINT||timeseries used| +|timeseries\_total|BIGINT||total timeseries| +|dnode\_id|INT|TAG|dnode id| +|dnode\_ep|NCHAR|TAG|dnode endpoint| +|cluster\_id|NCHAR|TAG|cluster id| + +### keeper\_monitor table + +`keeper_monitor` table contains keeper monitor information records. + +|field|type|is\_tag|comment| +|:----|:---|:-----|:------| +|ts|TIMESTAMP||timestamp| +|cpu|FLOAT||cpu usage| +|mem|FLOAT||memory usage| +|identify|NCHAR|TAG|| + +### taosadapter\_restful\_http\_request\_total table + +`taosadapter_restful_http_request_total` table contains taosadapter rest request information record. The timestamp column of this table is `_ts`. + +|field|type|is\_tag|comment| +|:----|:---|:-----|:------| +|\_ts|TIMESTAMP||timestamp| +|guage|DOUBLE||metric value| +|client\_ip|NCHAR|TAG|client ip| +|endpoint|NCHAR|TAG|taosadpater endpoint| +|request\_method|NCHAR|TAG|request method| +|request\_uri|NCHAR|TAG|request uri| +|status\_code|NCHAR|TAG|status code| + +### taosadapter\_restful\_http\_request\_fail table + +`taosadapter_restful_http_request_fail` table contains taosadapter failed rest request information record. The timestamp column of this table is `_ts`. + +|field|type|is\_tag|comment| +|:----|:---|:-----|:------| +|\_ts|TIMESTAMP||timestamp| +|guage|DOUBLE||metric value| +|client\_ip|NCHAR|TAG|client ip| +|endpoint|NCHAR|TAG|taosadpater endpoint| +|request\_method|NCHAR|TAG|request method| +|request\_uri|NCHAR|TAG|request uri| +|status\_code|NCHAR|TAG|status code| + +### taosadapter\_restful\_http\_request\_in\_flight table + +`taosadapter_restful_http_request_in_flight` table contains taosadapter rest request information record in real time. The timestamp column of this table is `_ts`. + +|field|type|is\_tag|comment| +|:----|:---|:-----|:------| +|\_ts|TIMESTAMP||timestamp| +|guage|DOUBLE||metric value| +|endpoint|NCHAR|TAG|taosadpater endpoint| + +### taosadapter\_restful\_http\_request\_summary\_milliseconds table + +`taosadapter_restful_http_request_summary_milliseconds` table contains the summary or rest information record. The timestamp column of this table is `_ts`. + +|field|type|is\_tag|comment| +|:----|:---|:-----|:------| +|\_ts|TIMESTAMP||timestamp| +|count|DOUBLE||| +|sum|DOUBLE||| +|0.5|DOUBLE||| +|0.9|DOUBLE||| +|0.99|DOUBLE||| +|0.1|DOUBLE||| +|0.2|DOUBLE||| +|endpoint|NCHAR|TAG|taosadpater endpoint| +|request\_method|NCHAR|TAG|request method| +|request\_uri|NCHAR|TAG|request uri| + +### taosadapter\_system\_mem\_percent table + +`taosadapter_system_mem_percent` table contains taosadapter memory usage information. The timestamp of this table is `_ts`. + +|field|type|is\_tag|comment| +|:----|:---|:-----|:------| +|\_ts|TIMESTAMP||timestamp| +|guage|DOUBLE||metric value| +|endpoint|NCHAR|TAG|taosadpater endpoint| + +### taosadapter\_system\_cpu\_percent table + +`taosadapter_system_cpu_percent` table contains taosadapter cup usage information. The timestamp of this table is `_ts`. + +|field|type|is\_tag|comment| +|:----|:---|:-----|:------| +|\_ts|TIMESTAMP||timestamp| +|guage|DOUBLE||mertic value| +|endpoint|NCHAR|TAG|taosadpater endpoint| + diff --git a/docs/en/14-reference/03-connector/03-cpp.mdx b/docs/en/14-reference/03-connector/03-cpp.mdx index b543879b3c..014109b77e 100644 --- a/docs/en/14-reference/03-connector/03-cpp.mdx +++ b/docs/en/14-reference/03-connector/03-cpp.mdx @@ -423,6 +423,6 @@ In addition to writing data using the SQL method or the parameter binding API, w **Description** - The above seven interfaces are extension interfaces, which are mainly used to pass ttl and reqid parameters, and can be used as needed. - - Withing _raw interfaces represent data through the passed parameters lines and len. In order to solve the problem that the original interface data contains '\0' and is truncated. The totalRows pointer returns the number of parsed data rows. - - Withing _ttl interfaces can pass the ttl parameter to control the ttl expiration time of the table. - - Withing _reqid interfaces can track the entire call chain by passing the reqid parameter. + - Within _raw interfaces represent data through the passed parameters lines and len. In order to solve the problem that the original interface data contains '\0' and is truncated. The totalRows pointer returns the number of parsed data rows. + - Within _ttl interfaces can pass the ttl parameter to control the ttl expiration time of the table. + - Within _reqid interfaces can track the entire call chain by passing the reqid parameter. diff --git a/docs/en/14-reference/03-connector/04-java.mdx b/docs/en/14-reference/03-connector/04-java.mdx index fd4b4641d7..2847bf20f8 100644 --- a/docs/en/14-reference/03-connector/04-java.mdx +++ b/docs/en/14-reference/03-connector/04-java.mdx @@ -40,19 +40,19 @@ Please refer to [version support list](/reference/connector#version-support) TDengine currently supports timestamp, number, character, Boolean type, and the corresponding type conversion with Java is as follows: -| TDengine DataType | JDBCType | -| ----------------- | ---------------------------------- | -| TIMESTAMP | java.sql.Timestamp | -| INT | java.lang.Integer | -| BIGINT | java.lang.Long | -| FLOAT | java.lang.Float | -| DOUBLE | java.lang.Double | -| SMALLINT | java.lang.Short | -| TINYINT | java.lang.Byte | -| BOOL | java.lang.Boolean | -| BINARY | byte array | -| NCHAR | java.lang.String | -| JSON | java.lang.String | +| TDengine DataType | JDBCType | +| ----------------- | ------------------ | +| TIMESTAMP | java.sql.Timestamp | +| INT | java.lang.Integer | +| BIGINT | java.lang.Long | +| FLOAT | java.lang.Float | +| DOUBLE | java.lang.Double | +| SMALLINT | java.lang.Short | +| TINYINT | java.lang.Byte | +| BOOL | java.lang.Boolean | +| BINARY | byte array | +| NCHAR | java.lang.String | +| JSON | java.lang.String | **Note**: Only TAG supports JSON types @@ -82,7 +82,7 @@ Add following dependency in the `pom.xml` file of your Maven project: com.taosdata.jdbc taos-jdbcdriver - 3.0.0 + 3.1.0 ``` @@ -227,7 +227,7 @@ In addition to getting the connection from the specified URL, you can use Proper Note: - The client parameter set in the application is process-level. If you want to update the parameters of the client, you need to restart the application. This is because the client parameter is a global parameter that takes effect only the first time the application is set. -- The following sample code is based on taos-jdbcdriver-3.0.0. +- The following sample code is based on taos-jdbcdriver-3.1.0. ```java public Connection getConn() throws Exception{ @@ -350,7 +350,12 @@ try (Statement statement = connection.createStatement()) { } ``` -There are three types of error codes that the JDBC connector can report: - Error code of the JDBC driver itself (error code between 0x2301 and 0x2350), - Error code of the native connection method (error code between 0x2351 and 0x2400), and - Error code of other TDengine function modules. +There are four types of error codes that the JDBC connector can report: + +- Error code of the JDBC driver itself (error code between 0x2301 and 0x2350), +- Error code of the native connection method (error code between 0x2351 and 0x2360) +- Error code of the consumer method (error code between 0x2371 and 0x2380) +- Error code of other TDengine function modules. For specific error codes, please refer to. @@ -364,7 +369,7 @@ TDengine has significantly improved the bind APIs to support data writing (INSER **Note:** - JDBC REST connections do not currently support bind interface -- The following sample code is based on taos-jdbcdriver-3.0.0 +- The following sample code is based on taos-jdbcdriver-3.1.0 - The setString method should be called for binary type data, and the setNString method should be called for nchar type data - both setString and setNString require the user to declare the width of the corresponding column in the size parameter of the table definition @@ -632,7 +637,7 @@ TDengine supports schemaless writing. It is compatible with InfluxDB's Line Prot Note: - JDBC REST connections do not currently support schemaless writes -- The following sample code is based on taos-jdbcdriver-3.0.0 +- The following sample code is based on taos-jdbcdriver-3.1.0 ```java public class SchemalessInsertTest { @@ -965,17 +970,17 @@ The source code of the sample application is under `TDengine/examples/JDBC`: ## Recent update logs -| taos-jdbcdriver version | major changes | -| :---------------------: | :--------------------------------------------: | -| 3.1.0 | JDBC REST connection supports subscription over WebSocket | -| 3.0.1 - 3.0.4 | fix the resultSet data is parsed incorrectly sometimes. 3.0.1 is compiled on JDK 11, you are advised to use other version in the JDK 8 environment | -| 3.0.0 | Support for TDengine 3.0 | -| 2.0.42 | fix wasNull interface return value in WebSocket connection | -| 2.0.41 | fix decode method of username and password in REST connection | -| 2.0.39 - 2.0.40 | Add REST connection/request timeout parameters | -| 2.0.38 | JDBC REST connections add bulk pull function | -| 2.0.37 | Support json tags | -| 2.0.36 | Support schemaless writing | +| taos-jdbcdriver version | major changes | +| :---------------------: | :------------------------------------------------------------------------------------------------------------------------------------------------: | +| 3.1.0 | JDBC REST connection supports subscription over WebSocket | +| 3.0.1 - 3.0.4 | fix the resultSet data is parsed incorrectly sometimes. 3.0.1 is compiled on JDK 11, you are advised to use other version in the JDK 8 environment | +| 3.0.0 | Support for TDengine 3.0 | +| 2.0.42 | fix wasNull interface return value in WebSocket connection | +| 2.0.41 | fix decode method of username and password in REST connection | +| 2.0.39 - 2.0.40 | Add REST connection/request timeout parameters | +| 2.0.38 | JDBC REST connections add bulk pull function | +| 2.0.37 | Support json tags | +| 2.0.36 | Support schemaless writing | ## Frequently Asked Questions @@ -999,9 +1004,9 @@ The source code of the sample application is under `TDengine/examples/JDBC`: 4. java.lang.NoSuchMethodError: setByteArray - **Cause**: taos-jbdcdriver 3.* only supports TDengine 3.0 and later. + **Cause**: taos-jbdcdriver 3.\* only supports TDengine 3.0 and later. - **Solution**: Use taos-jdbcdriver 2.* with your TDengine 2.* deployment. + **Solution**: Use taos-jdbcdriver 2.\* with your TDengine 2.\* deployment. 5. java.lang.NoSuchMethodError: java.nio.ByteBuffer.position(I)Ljava/nio/ByteBuffer; ... taos-jdbcdriver-3.0.1.jar diff --git a/docs/en/14-reference/03-connector/06-rust.mdx b/docs/en/14-reference/03-connector/06-rust.mdx index ad522e9d2b..597a9b0f52 100644 --- a/docs/en/14-reference/03-connector/06-rust.mdx +++ b/docs/en/14-reference/03-connector/06-rust.mdx @@ -11,6 +11,7 @@ import TabItem from '@theme/TabItem'; import Preparition from "./_preparation.mdx" import RustInsert from "../../07-develop/03-insert-data/_rust_sql.mdx" import RustBind from "../../07-develop/03-insert-data/_rust_stmt.mdx" +import RustSml from "../../07-develop/03-insert-data/_rust_schemaless.mdx" import RustQuery from "../../07-develop/04-query-data/_rust.mdx" [![Crates.io](https://img.shields.io/crates/v/taos)](https://crates.io/crates/taos) ![Crates.io](https://img.shields.io/crates/d/taos) [![docs.rs](https://img.shields.io/docsrs/taos)](https://docs.rs/taos) @@ -232,6 +233,10 @@ There are two ways to query data: Using built-in types or the [serde](https://se +#### Schemaless Write + + + ### Query data diff --git a/docs/en/14-reference/03-connector/index.mdx b/docs/en/14-reference/03-connector/index.mdx index a35d5bc2d1..28b7b83b58 100644 --- a/docs/en/14-reference/03-connector/index.mdx +++ b/docs/en/14-reference/03-connector/index.mdx @@ -62,7 +62,7 @@ The different database framework specifications for various programming language | **Regular Query** | Support | Support | Support | Support | Support | Support | | **Parameter Binding** | Not Supported | Not Supported | Support | Support | Not Supported | Support | | **Subscription (TMQ) ** | Supported | Support | Support | Not Supported | Not Supported | Support | -| **Schemaless** | Not Supported | Not Supported | Not Supported | Not Supported | Not Supported | Not Supported | +| **Schemaless** | Supported | Not Supported | Not Supported | Not Supported | Not Supported | Not Supported | | **Bulk Pulling (based on WebSocket) ** | Support | Support | Support | Support | Support | Support | | **DataFrame** | Not Supported | Support | Not Supported | Not Supported | Not Supported | Not Supported | diff --git a/docs/en/14-reference/06-taosdump.md b/docs/en/14-reference/06-taosdump.md index dcfc068a3d..7348add4bd 100644 --- a/docs/en/14-reference/06-taosdump.md +++ b/docs/en/14-reference/06-taosdump.md @@ -76,6 +76,7 @@ Usage: taosdump [OPTION...] dbname [tbname ...] -A, --all-databases Dump all databases. -D, --databases=DATABASES Dump listed databases. Use comma to separate database names. + -e, --escape-character Use escaped character for database name -N, --without-property Dump database without its properties. -s, --schemaonly Only dump table schemas. -y, --answer-yes Input yes for prompt. It will skip data file diff --git a/docs/en/14-reference/12-config/index.md b/docs/en/14-reference/12-config/index.md index 2e6f5ec1e2..715a68696b 100644 --- a/docs/en/14-reference/12-config/index.md +++ b/docs/en/14-reference/12-config/index.md @@ -358,6 +358,17 @@ The charset that takes effect is UTF-8. | Value Range | 0-4096 | | Default Value | 2x the CPU cores | +## Performance Tuning + +### numOfCommitThreads + +| Attribute | Description | +| ------------- | ----------------------------------- | +| Applicable | Server Only | +| Meaning | Maximum number of threads to commit | +| Value Range | 0-1024 | +| Default Value | | + ## Log Parameters ### logDir diff --git a/docs/en/14-reference/14-taosKeeper.md b/docs/en/14-reference/14-taosKeeper.md index 895bd82e19..cfc2554387 100644 --- a/docs/en/14-reference/14-taosKeeper.md +++ b/docs/en/14-reference/14-taosKeeper.md @@ -108,7 +108,7 @@ The following `launchctl` commands can help you manage taoskeeper service: #### Launch With Configuration File -You can quickly launch taosKeeper with the following commands. If you do not specify a configuration file, `/etc/taos/keeper.toml` is used by default. If this file does not specify configurations, the default values are used. +You can quickly launch taosKeeper with the following commands. If you do not specify a configuration file, `/etc/taos/taoskeeper.toml` is used by default. If this file does not specify configurations, the default values are used. ```shell $ taoskeeper -c @@ -153,6 +153,10 @@ database = "log" # standard tables to monitor tables = ["normal_table"] + +# database options for db storing metrics data +[metrics.databaseoptions] +cachemodel = "none" ``` ### Obtain Monitoring Metrics @@ -203,7 +207,7 @@ taos_cluster_info_dnodes_total{cluster_id="5981392874047724755"} 1 taos_cluster_info_first_ep{cluster_id="5981392874047724755",value="hlb:6030"} 1 ``` -### check_health +### check\_health ``` $ curl -i http://127.0.0.1:6043/check_health @@ -219,3 +223,29 @@ Content-Length: 19 {"version":"1.0.0"} ``` + +### taoskeeper with Prometheus + +There is `/metrics` api in taoskeeper provide TDengine metric data for Prometheus. + +#### scrape config + +Scrape config in Prometheus specifies a set of targets and parameters describing how to scrape metric data from endpoint. For more information, please reference to [Prometheus documents](https://prometheus.io/docs/prometheus/latest/configuration/configuration/#scrape_config). + +``` +# A scrape configuration containing exactly one endpoint to scrape: +# Here it's Prometheus itself. +scrape_configs: + - job_name: "taoskeeper" + # metrics_path defaults to '/metrics' + # scheme defaults to 'http'. + static_configs: + - targets: ["localhost:6043"] +``` + +#### Dashboard + +There is a dashboard named `TaosKeeper Prometheus Dashboard for 3.x`, which provides a monitoring dashboard similar to TInsight. + +In Grafana, click the Dashboard menu and click `import`, enter the dashboard ID `18587` and click the `Load` button. Then finished importing `TaosKeeper Prometheus Dashboard for 3.x` dashboard. + diff --git a/docs/en/28-releases/01-tdengine.md b/docs/en/28-releases/01-tdengine.md index acfbf6a0ba..fb31ad67c0 100644 --- a/docs/en/28-releases/01-tdengine.md +++ b/docs/en/28-releases/01-tdengine.md @@ -10,6 +10,10 @@ For TDengine 2.x installation packages by version, please visit [here](https://w import Release from "/components/ReleaseV3"; +## 3.0.4.1 + + + ## 3.0.4.0 diff --git a/docs/en/28-releases/02-tools.md b/docs/en/28-releases/02-tools.md index e8f7a54566..9f8dbfee7e 100644 --- a/docs/en/28-releases/02-tools.md +++ b/docs/en/28-releases/02-tools.md @@ -10,6 +10,10 @@ For other historical version installers, please visit [here](https://www.taosdat import Release from "/components/ReleaseV3"; +## 2.5.0 + + + ## 2.4.12 diff --git a/docs/examples/c/async_query_example.c b/docs/examples/c/async_query_example.c index b370420b12..3807c4bfd7 100644 --- a/docs/examples/c/async_query_example.c +++ b/docs/examples/c/async_query_example.c @@ -8,7 +8,7 @@ #include #include -typedef int16_t VarDataLenT; +typedef uint16_t VarDataLenT; #define TSDB_NCHAR_SIZE sizeof(int32_t) #define VARSTR_HEADER_SIZE sizeof(VarDataLenT) diff --git a/docs/examples/c/query_example.c b/docs/examples/c/query_example.c index fcae95bcd4..c7d52115b5 100644 --- a/docs/examples/c/query_example.c +++ b/docs/examples/c/query_example.c @@ -6,7 +6,7 @@ #include #include -typedef int16_t VarDataLenT; +typedef uint16_t VarDataLenT; #define TSDB_NCHAR_SIZE sizeof(int32_t) #define VARSTR_HEADER_SIZE sizeof(VarDataLenT) diff --git a/docs/examples/rust/nativeexample/examples/schemaless_insert_json.rs b/docs/examples/rust/nativeexample/examples/schemaless_insert_json.rs new file mode 100644 index 0000000000..d5aa45016e --- /dev/null +++ b/docs/examples/rust/nativeexample/examples/schemaless_insert_json.rs @@ -0,0 +1,74 @@ +use taos_query::common::SchemalessPrecision; +use taos_query::common::SchemalessProtocol; +use taos_query::common::SmlDataBuilder; + +use crate::AsyncQueryable; +use crate::AsyncTBuilder; +use crate::TaosBuilder; + +async fn put_json() -> anyhow::Result<()> { + // std::env::set_var("RUST_LOG", "taos=trace"); + std::env::set_var("RUST_LOG", "taos=debug"); + pretty_env_logger::init(); + let dsn = + std::env::var("TDENGINE_ClOUD_DSN").unwrap_or("http://localhost:6041".to_string()); + log::debug!("dsn: {:?}", &dsn); + + let client = TaosBuilder::from_dsn(dsn)?.build().await?; + + let db = "demo_schemaless_ws"; + + client.exec(format!("drop database if exists {db}")).await?; + + client + .exec(format!("create database if not exists {db}")) + .await?; + + // should specify database before insert + client.exec(format!("use {db}")).await?; + + // SchemalessProtocol::Json + let data = [ + r#"[{"metric": "meters.current", "timestamp": 1681345954000, "value": 10.3, "tags": {"location": "California.SanFrancisco", "groupid": 2}}, {"metric": "meters.voltage", "timestamp": 1648432611249, "value": 219, "tags": {"location": "California.LosAngeles", "groupid": 1}}, {"metric": "meters.current", "timestamp": 1648432611250, "value": 12.6, "tags": {"location": "California.SanFrancisco", "groupid": 2}}, {"metric": "meters.voltage", "timestamp": 1648432611250, "value": 221, "tags": {"location": "California.LosAngeles", "groupid": 1}}]"# + ] + .map(String::from) + .to_vec(); + + // demo with all fields + let sml_data = SmlDataBuilder::default() + .protocol(SchemalessProtocol::Json) + .precision(SchemalessPrecision::Millisecond) + .data(data.clone()) + .ttl(1000) + .req_id(300u64) + .build()?; + assert_eq!(client.put(&sml_data).await?, ()); + + // demo with default precision + let sml_data = SmlDataBuilder::default() + .protocol(SchemalessProtocol::Json) + .data(data.clone()) + .ttl(1000) + .req_id(301u64) + .build()?; + assert_eq!(client.put(&sml_data).await?, ()); + + // demo with default ttl + let sml_data = SmlDataBuilder::default() + .protocol(SchemalessProtocol::Json) + .data(data.clone()) + .req_id(302u64) + .build()?; + assert_eq!(client.put(&sml_data).await?, ()); + + // demo with default req_id + let sml_data = SmlDataBuilder::default() + .protocol(SchemalessProtocol::Json) + .data(data.clone()) + .build()?; + assert_eq!(client.put(&sml_data).await?, ()); + + client.exec(format!("drop database if exists {db}")).await?; + + Ok(()) +} diff --git a/docs/examples/rust/nativeexample/examples/schemaless_insert_line.rs b/docs/examples/rust/nativeexample/examples/schemaless_insert_line.rs new file mode 100644 index 0000000000..3cf4d969ce --- /dev/null +++ b/docs/examples/rust/nativeexample/examples/schemaless_insert_line.rs @@ -0,0 +1,78 @@ +use taos_query::common::SchemalessPrecision; +use taos_query::common::SchemalessProtocol; +use taos_query::common::SmlDataBuilder; + +use crate::AsyncQueryable; +use crate::AsyncTBuilder; +use crate::TaosBuilder; + +async fn put_line() -> anyhow::Result<()> { + // std::env::set_var("RUST_LOG", "taos=trace"); + std::env::set_var("RUST_LOG", "taos=debug"); + pretty_env_logger::init(); + + let dsn = + std::env::var("TDENGINE_ClOUD_DSN").unwrap_or("http://localhost:6041".to_string()); + log::debug!("dsn: {:?}", &dsn); + + let client = TaosBuilder::from_dsn(dsn)?.build().await?; + + let db = "demo_schemaless_ws"; + + client.exec(format!("drop database if exists {db}")).await?; + + client + .exec(format!("create database if not exists {db}")) + .await?; + + // should specify database before insert + client.exec(format!("use {db}")).await?; + + let data = [ + "measurement,host=host1 field1=2i,field2=2.0 1577837300000", + "measurement,host=host1 field1=2i,field2=2.0 1577837400000", + "measurement,host=host1 field1=2i,field2=2.0 1577837500000", + "measurement,host=host1 field1=2i,field2=2.0 1577837600000", + ] + .map(String::from) + .to_vec(); + + // demo with all fields + let sml_data = SmlDataBuilder::default() + .protocol(SchemalessProtocol::Line) + .precision(SchemalessPrecision::Millisecond) + .data(data.clone()) + .ttl(1000) + .req_id(100u64) + .build()?; + assert_eq!(client.put(&sml_data).await?, ()); + + // demo with default ttl + let sml_data = SmlDataBuilder::default() + .protocol(SchemalessProtocol::Line) + .precision(SchemalessPrecision::Millisecond) + .data(data.clone()) + .req_id(101u64) + .build()?; + assert_eq!(client.put(&sml_data).await?, ()); + + // demo with default ttl and req_id + let sml_data = SmlDataBuilder::default() + .protocol(SchemalessProtocol::Line) + .precision(SchemalessPrecision::Millisecond) + .data(data.clone()) + .build()?; + assert_eq!(client.put(&sml_data).await?, ()); + + // demo with default precision + let sml_data = SmlDataBuilder::default() + .protocol(SchemalessProtocol::Line) + .data(data) + .req_id(103u64) + .build()?; + assert_eq!(client.put(&sml_data).await?, ()); + + client.exec(format!("drop database if exists {db}")).await?; + + Ok(()) +} diff --git a/docs/examples/rust/nativeexample/examples/schemaless_insert_telnet.rs b/docs/examples/rust/nativeexample/examples/schemaless_insert_telnet.rs new file mode 100644 index 0000000000..d2bbc69261 --- /dev/null +++ b/docs/examples/rust/nativeexample/examples/schemaless_insert_telnet.rs @@ -0,0 +1,80 @@ +use taos_query::common::SchemalessPrecision; +use taos_query::common::SchemalessProtocol; +use taos_query::common::SmlDataBuilder; + +use crate::AsyncQueryable; +use crate::AsyncTBuilder; +use crate::TaosBuilder; + +async fn put_telnet() -> anyhow::Result<()> { + // std::env::set_var("RUST_LOG", "taos=trace"); + std::env::set_var("RUST_LOG", "taos=debug"); + pretty_env_logger::init(); + let dsn = + std::env::var("TDENGINE_ClOUD_DSN").unwrap_or("http://localhost:6041".to_string()); + log::debug!("dsn: {:?}", &dsn); + + let client = TaosBuilder::from_dsn(dsn)?.build().await?; + + let db = "demo_schemaless_ws"; + + client.exec(format!("drop database if exists {db}")).await?; + + client + .exec(format!("create database if not exists {db}")) + .await?; + + // should specify database before insert + client.exec(format!("use {db}")).await?; + + let data = [ + "meters.current 1648432611249 10.3 location=California.SanFrancisco group=2", + "meters.current 1648432611250 12.6 location=California.SanFrancisco group=2", + "meters.current 1648432611249 10.8 location=California.LosAngeles group=3", + "meters.current 1648432611250 11.3 location=California.LosAngeles group=3", + "meters.voltage 1648432611249 219 location=California.SanFrancisco group=2", + "meters.voltage 1648432611250 218 location=California.SanFrancisco group=2", + "meters.voltage 1648432611249 221 location=California.LosAngeles group=3", + "meters.voltage 1648432611250 217 location=California.LosAngeles group=3", + ] + .map(String::from) + .to_vec(); + + // demo with all fields + let sml_data = SmlDataBuilder::default() + .protocol(SchemalessProtocol::Telnet) + .precision(SchemalessPrecision::Millisecond) + .data(data.clone()) + .ttl(1000) + .req_id(200u64) + .build()?; + assert_eq!(client.put(&sml_data).await?, ()); + + // demo with default precision + let sml_data = SmlDataBuilder::default() + .protocol(SchemalessProtocol::Telnet) + .data(data.clone()) + .ttl(1000) + .req_id(201u64) + .build()?; + assert_eq!(client.put(&sml_data).await?, ()); + + // demo with default ttl + let sml_data = SmlDataBuilder::default() + .protocol(SchemalessProtocol::Telnet) + .data(data.clone()) + .req_id(202u64) + .build()?; + assert_eq!(client.put(&sml_data).await?, ()); + + // demo with default req_id + let sml_data = SmlDataBuilder::default() + .protocol(SchemalessProtocol::Telnet) + .data(data.clone()) + .build()?; + assert_eq!(client.put(&sml_data).await?, ()); + + client.exec(format!("drop database if exists {db}")).await?; + + Ok(()) +} diff --git a/docs/zh/05-get-started/03-package.md b/docs/zh/05-get-started/03-package.md index bdf6a730d0..1cd0076ba5 100644 --- a/docs/zh/05-get-started/03-package.md +++ b/docs/zh/05-get-started/03-package.md @@ -100,7 +100,8 @@ sudo apt-get install tdengine :::tip apt-get 方式只适用于 Debian 或 Ubuntu 系统。 -:::: +::: + @@ -206,6 +207,8 @@ Active: inactive (dead) - 查看服务状态：`sudo launchctl list | grep taosd` +- 查看服务详细信息：`launchctl print system/com.tdengine.taosd` + :::info - `launchctl` 命令管理`com.tdengine.taosd`需要管理员权限，务必在前面加 `sudo` 来增强安全性。 diff --git a/docs/zh/07-develop/03-insert-data/30-influxdb-line.mdx b/docs/zh/07-develop/03-insert-data/30-influxdb-line.mdx index 876f123fe1..258adaaeed 100644 --- a/docs/zh/07-develop/03-insert-data/30-influxdb-line.mdx +++ b/docs/zh/07-develop/03-insert-data/30-influxdb-line.mdx @@ -38,7 +38,7 @@ meters,location=California.LosAngeles,groupid=2 current=13.4,voltage=223,phase=0 - field_set 中的每个数据项都需要对自身的数据类型进行描述, 比如 1.2f32 代表 FLOAT 类型的数值 1.2, 如果不带类型后缀会被当作 DOUBLE 处理; - timestamp 支持多种时间精度。写入数据的时候需要用参数指定时间精度，支持从小时到纳秒的 6 种时间精度。 - 为了提高写入的效率，默认假设同一个超级表中 field_set 的顺序是一样的（第一条数据包含所有的 field，后面的数据按照这个顺序），如果顺序不一样，需要配置参数 smlDataFormat 为 false，否则，数据写入按照相同顺序写入，库中数据会异常。（3.0.1.3 之后的版本 smlDataFormat 默认为 false，从3.0.3.0开始，该配置废弃） [TDengine 无模式写入参考指南](/reference/schemaless/#无模式写入行协议) -- 默认产生的子表名是根据规则生成的唯一 ID 值。用户也可以通过在 taos.cfg 里配置 smlChildTableName 参数来指定某个标签值作为子表名。该标签值应该具有全局唯一性。举例如下：假设有个标签名为tname, 配置 smlChildTableName=tname, 插入数据为 st,tname=cpu1,t1=4 c1=3 1626006833639000000 则创建的子表名为 cpu1。注意如果多行数据 tname 相同，但是后面的 tag_set 不同，则使用第一行自动建表时指定的 tag_set，其他的行会忽略）。[TDengine 无模式写入参考指南](/reference/schemaless/#无模式写入行协议) +- 默认产生的子表名是根据规则生成的唯一 ID 值。用户也可以通过在client端的 taos.cfg 里配置 smlChildTableName 参数来指定某个标签值作为子表名。该标签值应该具有全局唯一性。举例如下：假设有个标签名为tname, 配置 smlChildTableName=tname, 插入数据为 st,tname=cpu1,t1=4 c1=3 1626006833639000000 则创建的子表名为 cpu1。注意如果多行数据 tname 相同，但是后面的 tag_set 不同，则使用第一行自动建表时指定的 tag_set，其他的行会忽略）。[TDengine 无模式写入参考指南](/reference/schemaless/#无模式写入行协议) ::: 要了解更多可参考：[InfluxDB Line 协议官方文档](https://docs.influxdata.com/influxdb/v2.0/reference/syntax/line-protocol/) 和 [TDengine 无模式写入参考指南](/reference/schemaless/#无模式写入行协议) diff --git a/docs/zh/07-develop/03-insert-data/40-opentsdb-telnet.mdx b/docs/zh/07-develop/03-insert-data/40-opentsdb-telnet.mdx index 3b2148ef4a..9da767a691 100644 --- a/docs/zh/07-develop/03-insert-data/40-opentsdb-telnet.mdx +++ b/docs/zh/07-develop/03-insert-data/40-opentsdb-telnet.mdx @@ -32,7 +32,7 @@ OpenTSDB 行协议同样采用一行字符串来表示一行数据。OpenTSDB meters.current 1648432611250 11.3 location=California.LosAngeles groupid=3 ``` -- 默认生产的子表名是根据规则生成的唯一 ID 值。用户也可以通过在 taos.cfg 里配置 smlChildTableName 参数来指定某个标签值作为子表名。该标签值应该具有全局唯一性。举例如下：假设有个标签名为tname, 配置 smlChildTableName=tname, 插入数据为 meters.current 1648432611250 11.3 tname=cpu1 location=California.LosAngeles groupid=3 则创建的表名为 cpu1，注意如果多行数据 tname 相同，但是后面的 tag_set 不同，则使用第一行自动建表时指定的 tag_set，其他的行会忽略）。 +- 默认生产的子表名是根据规则生成的唯一 ID 值。用户也可以通过在client端的 taos.cfg 里配置 smlChildTableName 参数来指定某个标签值作为子表名。该标签值应该具有全局唯一性。举例如下：假设有个标签名为tname, 配置 smlChildTableName=tname, 插入数据为 meters.current 1648432611250 11.3 tname=cpu1 location=California.LosAngeles groupid=3 则创建的表名为 cpu1，注意如果多行数据 tname 相同，但是后面的 tag_set 不同，则使用第一行自动建表时指定的 tag_set，其他的行会忽略）。 参考 [OpenTSDB Telnet API 文档](http://opentsdb.net/docs/build/html/api_telnet/put.html)。 ## 示例代码 diff --git a/docs/zh/07-develop/03-insert-data/50-opentsdb-json.mdx b/docs/zh/07-develop/03-insert-data/50-opentsdb-json.mdx index e1fd3dacc8..f2cfbc857b 100644 --- a/docs/zh/07-develop/03-insert-data/50-opentsdb-json.mdx +++ b/docs/zh/07-develop/03-insert-data/50-opentsdb-json.mdx @@ -47,7 +47,7 @@ OpenTSDB JSON 格式协议采用一个 JSON 字符串表示一行或多行数据 :::note - 对于 JSON 格式协议，TDengine 并不会自动把所有标签转成 NCHAR 类型， 字符串将将转为 NCHAR 类型， 数值将同样转换为 DOUBLE 类型。 -- 默认生成的子表名是根据规则生成的唯一 ID 值。用户也可以通过在 taos.cfg 里配置 smlChildTableName 参数来指定某个标签值作为子表名。该标签值应该具有全局唯一性。举例如下：假设有个标签名为tname, 配置 smlChildTableName=tname, 插入数据为 `"tags": { "host": "web02","dc": "lga","tname":"cpu1"}` 则创建的子表名为 cpu1。注意如果多行数据 tname 相同，但是后面的 tag_set 不同，则使用第一行自动建表时指定的 tag_set，其他的行会忽略）。 +- 默认生成的子表名是根据规则生成的唯一 ID 值。用户也可以通过在client端的 taos.cfg 里配置 smlChildTableName 参数来指定某个标签值作为子表名。该标签值应该具有全局唯一性。举例如下：假设有个标签名为tname, 配置 smlChildTableName=tname, 插入数据为 `"tags": { "host": "web02","dc": "lga","tname":"cpu1"}` 则创建的子表名为 cpu1。注意如果多行数据 tname 相同，但是后面的 tag_set 不同，则使用第一行自动建表时指定的 tag_set，其他的行会忽略）。 ::: ## 示例代码 diff --git a/docs/zh/07-develop/03-insert-data/_rust_schemaless.mdx b/docs/zh/07-develop/03-insert-data/_rust_schemaless.mdx new file mode 100644 index 0000000000..a26613bd15 --- /dev/null +++ b/docs/zh/07-develop/03-insert-data/_rust_schemaless.mdx @@ -0,0 +1,3 @@ +```rust +{{#include docs/examples/rust/nativeexample/examples/schemaless_insert_line.rs}} +``` diff --git a/docs/zh/07-develop/07-tmq.mdx b/docs/zh/07-develop/07-tmq.mdx index 036c9aa86e..a3833438a2 100644 --- a/docs/zh/07-develop/07-tmq.mdx +++ b/docs/zh/07-develop/07-tmq.mdx @@ -221,7 +221,7 @@ void Close() ```sql DROP DATABASE IF EXISTS tmqdb; -CREATE DATABASE tmqdb; +CREATE DATABASE tmqdb WAL_RETENTION_PERIOD 3600; CREATE TABLE tmqdb.stb (ts TIMESTAMP, c1 INT, c2 FLOAT, c3 VARCHAR(16)) TAGS(t1 INT, t3 VARCHAR(16)); CREATE TABLE tmqdb.ctb0 USING tmqdb.stb TAGS(0, "subtable0"); CREATE TABLE tmqdb.ctb1 USING tmqdb.stb TAGS(1, "subtable1"); diff --git a/docs/zh/07-develop/09-udf.md b/docs/zh/07-develop/09-udf.md index 8a8ef82009..99ecd903b4 100644 --- a/docs/zh/07-develop/09-udf.md +++ b/docs/zh/07-develop/09-udf.md @@ -6,18 +6,20 @@ description: "支持用户编码的聚合函数和标量函数，在查询中嵌 在有些应用场景中，应用逻辑需要的查询无法直接使用系统内置的函数来表示。利用 UDF(User Defined Function) 功能，TDengine 可以插入用户编写的处理代码并在查询中使用它们，就能够很方便地解决特殊应用场景中的使用需求。 UDF 通常以数据表中的一列数据做为输入，同时支持以嵌套子查询的结果作为输入。 -TDengine 支持通过 C/C++ 语言进行 UDF 定义。接下来结合示例讲解 UDF 的使用方法。 - 用户可以通过 UDF 实现两类函数：标量函数和聚合函数。标量函数对每行数据输出一个值，如求绝对值 abs，正弦函数 sin，字符串拼接函数 concat 等。聚合函数对多行数据进行输出一个值，如求平均数 avg，最大值 max 等。 -实现 UDF 时，需要实现规定的接口函数 +TDengine 支持通过 C/Python 语言进行 UDF 定义。接下来结合示例讲解 UDF 的使用方法。 + +## 用 C 语言实现 UDF + +使用 C 语言实现 UDF 时，需要实现规定的接口函数 - 标量函数需要实现标量接口函数 scalarfn 。 - 聚合函数需要实现聚合接口函数 aggfn_start ， aggfn ， aggfn_finish。 - 如果需要初始化，实现 udf_init；如果需要清理工作，实现udf_destroy。 接口函数的名称是 UDF 名称，或者是 UDF 名称和特定后缀（_start, _finish, _init, _destroy)的连接。列表中的scalarfn，aggfn, udf需要替换成udf函数名。 -## 实现标量函数 +### 用 C 语言实现标量函数 标量函数实现模板如下 ```c #include "taos.h" @@ -49,7 +51,7 @@ int32_t scalarfn_destroy() { ``` scalarfn 为函数名的占位符，需要替换成函数名，如bit_and。 -## 实现聚合函数 +### 用 C 语言实现聚合函数 聚合函数的实现模板如下 ```c @@ -100,7 +102,7 @@ int32_t aggfn_destroy() { ``` aggfn为函数名的占位符，需要修改为自己的函数名，如l2norm。 -## 接口函数定义 +### C 语言 UDF 接口函数定义 接口函数的名称是 udf 名称，或者是 udf 名称和特定后缀（_start, _finish, _init, _destroy)的连接。以下描述中函数名称中的 scalarfn，aggfn, udf 需要替换成udf函数名。 @@ -108,7 +110,7 @@ aggfn为函数名的占位符，需要修改为自己的函数名，如l2norm。 接口函数参数类型见数据结构定义。 -### 标量接口函数 +#### 标量函数接口 `int32_t scalarfn(SUdfDataBlock* inputDataBlock, SUdfColumn *resultColumn)` @@ -118,7 +120,7 @@ aggfn为函数名的占位符，需要修改为自己的函数名，如l2norm。 - inputDataBlock: 输入的数据块 - resultColumn: 输出列 -### 聚合接口函数 +#### 聚合函数接口 `int32_t aggfn_start(SUdfInterBuf *interBuf)` @@ -135,7 +137,7 @@ aggfn为函数名的占位符，需要修改为自己的函数名，如l2norm。 - result：最终结果。 -### UDF 初始化和销毁 +#### 初始化和销毁接口 `int32_t udf_init()` `int32_t udf_destroy()` @@ -143,7 +145,7 @@ aggfn为函数名的占位符，需要修改为自己的函数名，如l2norm。 其中 udf 是函数名的占位符。udf_init 完成初始化工作。 udf_destroy 完成清理工作。如果没有初始化工作，无需定义udf_init函数。如果没有清理工作，无需定义udf_destroy函数。 -## UDF 数据结构 +### C 语言 UDF 数据结构 ```c typedef struct SUdfColumnMeta { int16_t type; @@ -201,7 +203,7 @@ typedef struct SUdfInterBuf { 为了更好的操作以上数据结构，提供了一些便利函数，定义在 taosudf.h。 -## 编译 UDF +### 编译 C UDF 用户定义函数的 C 语言源代码无法直接被 TDengine 系统使用，而是需要先编译为 动态链接库，之后才能载入 TDengine 系统。 @@ -213,12 +215,9 @@ gcc -g -O0 -fPIC -shared bit_and.c -o libbitand.so 这样就准备好了动态链接库 libbitand.so 文件，可以供后文创建 UDF 时使用了。为了保证可靠的系统运行，编译器 GCC 推荐使用 7.5 及以上版本。 -## 管理和使用UDF -编译好的UDF，还需要将其加入到系统才能被正常的SQL调用。关于如何管理和使用UDF，参见[UDF使用说明](../12-taos-sql/26-udf.md) +### C UDF 示例代码 -## 示例代码 - -### 标量函数示例 [bit_and](https://github.com/taosdata/TDengine/blob/develop/tests/script/sh/bit_and.c) +#### 标量函数示例 [bit_and](https://github.com/taosdata/TDengine/blob/3.0/tests/script/sh/bit_and.c) bit_add 实现多列的按位与功能。如果只有一列，返回这一列。bit_add 忽略空值。 @@ -231,7 +230,7 @@ bit_add 实现多列的按位与功能。如果只有一列，返回这一列。 -### 聚合函数示例1 返回值为数值类型 [l2norm](https://github.com/taosdata/TDengine/blob/develop/tests/script/sh/l2norm.c) +#### 聚合函数示例1 返回值为数值类型 [l2norm](https://github.com/taosdata/TDengine/blob/3.0/tests/script/sh/l2norm.c) l2norm 实现了输入列的所有数据的二阶范数，即对每个数据先平方，再累加求和，最后开方。 @@ -244,7 +243,7 @@ l2norm 实现了输入列的所有数据的二阶范数，即对每个数据先 -### 聚合函数示例2 返回值为字符串类型 [max_vol](https://github.com/taosdata/TDengine/blob/develop/tests/script/sh/max_vol.c) +#### 聚合函数示例2 返回值为字符串类型 [max_vol](https://github.com/taosdata/TDengine/blob/3.0/tests/script/sh/max_vol.c) max_vol 实现了从多个输入的电压列中找到最大电压，返回由设备ID + 最大电压所在（行，列）+ 最大电压值 组成的组合字符串值 @@ -268,4 +267,125 @@ select max_vol(vol1,vol2,vol3,deviceid) from battery; {{#include tests/script/sh/max_vol.c}} ``` - \ No newline at end of file + + +## 用 Python 语言实现 UDF + +使用 Python 语言实现 UDF 时，需要实现规定的接口函数 +- 标量函数需要实现标量接口函数 process 。 +- 聚合函数需要实现聚合接口函数 start ，reduce ，finish。 +- 如果需要初始化，实现 init；如果需要清理工作，实现 destroy。 + +### 用 Python 实现标量函数 + +标量函数实现模版如下 +```Python +def init(): + # initialization +def destroy(): + # destroy +def process(input: datablock) -> tuple[output_type]: + # process input datablock, + # datablock.data(row, col) is to access the python object in location(row,col) + # return tuple object consisted of object of type outputtype +``` + +### 用 Python 实现聚合函数 + +聚合函数实现模版如下 +```Python +def init(): + #initialization +def destroy(): + #destroy +def start() -> bytes: + #return serialize(init_state) +def reduce(inputs: datablock, buf: bytes) -> bytes + # deserialize buf to state + # reduce the inputs and state into new_state. + # use inputs.data(i,j) to access python ojbect of location(i,j) + # serialize new_state into new_state_bytes + return new_state_bytes +def finish(buf: bytes) -> output_type: + #return obj of type outputtype +``` + +### Python UDF 接口函数定义 + +#### 标量函数接口 +```Python +def process(input: datablock) -> tuple[output_type]: +``` +- input:datablock 类似二维矩阵，通过成员方法 data(row,col)返回位于 row 行，col 列的 python 对象 +- 返回值是一个 Python 对象元组，每个元素类型为输出类型。 + +#### 聚合函数接口 +```Python +def start() -> bytes: +def reduce(inputs: datablock, buf: bytes) -> bytes +def finish(buf: bytes) -> output_type: +``` + +首先调用 start 生成最初结果 buffer，然后输入数据会被分为多个行数据块，对每个数据块 inputs 和当前中间结果 buf 调用 reduce，得到新的中间结果，最后再调用 finish 从中间结果 buf 产生最终输出，最终输出只能含 0 或 1 条数据。 + + +#### 初始化和销毁接口 +```Python +def init() +def destroy() +``` + +其中 init 完成初始化工作。 destroy 完成清理工作。如果没有初始化工作，无需定义 init 函数。如果没有清理工作，无需定义 destroy 函数。 + +### Python 和 TDengine之间的数据类型映射 + +下表描述了TDengine SQL数据类型和Python数据类型的映射。任何类型的NULL值都映射成Python的None值。 + +| **TDengine SQL数据类型** | **Python数据类型** | +| :-----------------------: | ------------ | +|TINYINT / SMALLINT / INT / BIGINT | int | +|TINYINT UNSIGNED / SMALLINT UNSIGNED / INT UNSIGNED / BIGINT UNSIGNED | int | +|FLOAT / DOUBLE | float | +|BOOL | bool | +|BINARY / VARCHAR / NCHAR | bytes| +|TIMESTAMP | int | +|JSON and other types | 不支持 | + +### Python UDF 环境的安装 +1. 安装 taospyudf 包。此包执行Python UDF程序。 +```bash +sudo pip install taospyudf +ldconfig +``` +2. 如果 Python UDF 程序执行时，通过 PYTHONPATH 引用其它的包，可以设置 taos.cfg 的 UdfdLdLibPath 变量为PYTHONPATH的内容 + +### Python UDF 示例代码 +#### 标量函数示例 [pybitand](https://github.com/taosdata/TDengine/blob/3.0/tests/script/sh/pybitand.py) + +pybitand 实现多列的按位与功能。如果只有一列，返回这一列。pybitand 忽略空值。 + +
+pybitand.py + +```Python +{{#include tests/script/sh/pybitand.py}} +``` + +
+ +#### 聚合函数示例 [pyl2norm](https://github.com/taosdata/TDengine/blob/3.0/tests/script/sh/pyl2norm.py) + +pyl2norm 实现了输入列的所有数据的二阶范数，即对每个数据先平方，再累加求和，最后开方。 + +
+pyl2norm.py + +```c +{{#include tests/script/sh/pyl2norm.py}} +``` + +
+ +## 管理和使用 UDF +在使用 UDF 之前需要先将其加入到 TDengine 系统中。关于如何管理和使用 UDF，请参考[管理和使用 UDF](../12-taos-sql/26-udf.md) + diff --git a/docs/zh/08-connector/14-java.mdx b/docs/zh/08-connector/14-java.mdx index d1c1258365..bf89b1df83 100644 --- a/docs/zh/08-connector/14-java.mdx +++ b/docs/zh/08-connector/14-java.mdx @@ -40,19 +40,19 @@ REST 连接支持所有能运行 Java 的平台。 TDengine 目前支持时间戳、数字、字符、布尔类型，与 Java 对应类型转换如下： -| TDengine DataType | JDBCType | -| ----------------- | ---------------------------------- | -| TIMESTAMP | java.sql.Timestamp | -| INT | java.lang.Integer | -| BIGINT | java.lang.Long | -| FLOAT | java.lang.Float | -| DOUBLE | java.lang.Double | -| SMALLINT | java.lang.Short | -| TINYINT | java.lang.Byte | -| BOOL | java.lang.Boolean | -| BINARY | byte array | -| NCHAR | java.lang.String | -| JSON | java.lang.String | +| TDengine DataType | JDBCType | +| ----------------- | ------------------ | +| TIMESTAMP | java.sql.Timestamp | +| INT | java.lang.Integer | +| BIGINT | java.lang.Long | +| FLOAT | java.lang.Float | +| DOUBLE | java.lang.Double | +| SMALLINT | java.lang.Short | +| TINYINT | java.lang.Byte | +| BOOL | java.lang.Boolean | +| BINARY | byte array | +| NCHAR | java.lang.String | +| JSON | java.lang.String | **注意**：JSON 类型仅在 tag 中支持。 @@ -82,7 +82,7 @@ Maven 项目中，在 pom.xml 中添加以下依赖： com.taosdata.jdbc taos-jdbcdriver - 3.0.0 + 3.1.0 ``` @@ -97,7 +97,7 @@ cd taos-connector-jdbc mvn clean install -Dmaven.test.skip=true ``` -编译后，在 target 目录下会产生 taos-jdbcdriver-3.0.*-dist.jar 的 jar 包，并自动将编译的 jar 文件放在本地的 Maven 仓库中。 +编译后，在 target 目录下会产生 taos-jdbcdriver-3.0.\*-dist.jar 的 jar 包，并自动将编译的 jar 文件放在本地的 Maven 仓库中。 @@ -230,7 +230,7 @@ INSERT INTO test.t1 USING test.weather (ts, temperature) TAGS('California.SanFra **注意**： - 应用中设置的 client parameter 为进程级别的，即如果要更新 client 的参数，需要重启应用。这是因为 client parameter 是全局参数，仅在应用程序的第一次设置生效。 -- 以下示例代码基于 taos-jdbcdriver-3.0.0。 +- 以下示例代码基于 taos-jdbcdriver-3.1.0。 ```java public Connection getConn() throws Exception{ @@ -272,7 +272,7 @@ properties 中的配置参数如下： - TSDBDriver.HTTP_SOCKET_TIMEOUT: socket 超时时间，单位 ms，默认值为 5000。仅在 REST 连接且 batchfetch 设置为 false 时生效。 - TSDBDriver.PROPERTY_KEY_MESSAGE_WAIT_TIMEOUT: 消息超时时间, 单位 ms， 默认值为 3000。 仅在 REST 连接且 batchfetch 设置为 true 时生效。 - TSDBDriver.PROPERTY_KEY_USE_SSL: 连接中是否使用 SSL。仅在 REST 连接时生效。 -此外对 JDBC 原生连接，通过指定 URL 和 Properties 还可以指定其他参数，比如日志级别、SQL 长度等。更多详细配置请参考[客户端配置](/reference/config/#仅客户端适用)。 + 此外对 JDBC 原生连接，通过指定 URL 和 Properties 还可以指定其他参数，比如日志级别、SQL 长度等。更多详细配置请参考[客户端配置](/reference/config/#仅客户端适用)。 ### 配置参数的优先级 @@ -353,7 +353,12 @@ try (Statement statement = connection.createStatement()) { } ``` -JDBC 连接器可能报错的错误码包括 3 种：JDBC driver 本身的报错（错误码在 0x2301 到 0x2350 之间），原生连接方法的报错（错误码在 0x2351 到 0x2400 之间），TDengine 其他功能模块的报错。 +JDBC 连接器可能报错的错误码包括 4 种： + +- JDBC driver 本身的报错（错误码在 0x2301 到 0x2350 之间） +- 原生连接方法的报错（错误码在 0x2351 到 0x2360 之间） +- 数据订阅的报错（错误码在 0x2371 到 0x2380 之间） +- TDengine 其他功能模块的报错。 具体的错误码请参考： @@ -367,7 +372,7 @@ TDengine 的 JDBC 原生连接实现大幅改进了参数绑定方式对数据 **注意**： - JDBC REST 连接目前不支持参数绑定 -- 以下示例代码基于 taos-jdbcdriver-3.0.0 +- 以下示例代码基于 taos-jdbcdriver-3.1.0 - binary 类型数据需要调用 setString 方法，nchar 类型数据需要调用 setNString 方法 - setString 和 setNString 都要求用户在 size 参数里声明表定义中对应列的列宽 @@ -635,7 +640,7 @@ TDengine 支持无模式写入功能。无模式写入兼容 InfluxDB 的 行协 **注意**： - JDBC REST 连接目前不支持无模式写入 -- 以下示例代码基于 taos-jdbcdriver-3.0.0 +- 以下示例代码基于 taos-jdbcdriver-3.1.0 ```java public class SchemalessInsertTest { @@ -702,7 +707,7 @@ TaosConsumer consumer = new TaosConsumer<>(config); - td.connect.type: 连接方式。jni：表示使用动态库连接的方式，ws/WebSocket：表示使用 WebSocket 进行数据通信。默认为 jni 方式。 - httpConnectTimeout：创建连接超时参数，单位 ms，默认为 5000 ms。仅在 WebSocket 连接下有效。 - messageWaitTimeout：数据传输超时参数，单位 ms，默认为 10000 ms。仅在 WebSocket 连接下有效。 -其他参数请参考：[Consumer 参数列表](../../../develop/tmq#创建-consumer-以及consumer-group) + 其他参数请参考：[Consumer 参数列表](../../../develop/tmq#创建-consumer-以及consumer-group) #### 订阅消费数据 @@ -968,17 +973,17 @@ public static void main(String[] args) throws Exception { ## 最近更新记录 -| taos-jdbcdriver 版本 | 主要变化 | -| :------------------: | :----------------------------: | -| 3.1.0 | WebSocket 连接支持订阅功能 | -| 3.0.1 - 3.0.4 | 修复一些情况下结果集数据解析错误的问题。3.0.1 在 JDK 11 环境编译，JDK 8 环境下建议使用其他版本 | -| 3.0.0 | 支持 TDengine 3.0 | -| 2.0.42 | 修在 WebSocket 连接中 wasNull 接口返回值 | -| 2.0.41 | 修正 REST 连接中用户名和密码转码方式 | -| 2.0.39 - 2.0.40 | 增加 REST 连接/请求 超时设置 | -| 2.0.38 | JDBC REST 连接增加批量拉取功能 | -| 2.0.37 | 增加对 json tag 支持 | -| 2.0.36 | 增加对 schemaless 写入支持 | +| taos-jdbcdriver 版本 | 主要变化 | +| :------------------: | :--------------------------------------------------------------------------------------------: | +| 3.1.0 | WebSocket 连接支持订阅功能 | +| 3.0.1 - 3.0.4 | 修复一些情况下结果集数据解析错误的问题。3.0.1 在 JDK 11 环境编译，JDK 8 环境下建议使用其他版本 | +| 3.0.0 | 支持 TDengine 3.0 | +| 2.0.42 | 修在 WebSocket 连接中 wasNull 接口返回值 | +| 2.0.41 | 修正 REST 连接中用户名和密码转码方式 | +| 2.0.39 - 2.0.40 | 增加 REST 连接/请求 超时设置 | +| 2.0.38 | JDBC REST 连接增加批量拉取功能 | +| 2.0.37 | 增加对 json tag 支持 | +| 2.0.36 | 增加对 schemaless 写入支持 | ## 常见问题 @@ -1002,9 +1007,9 @@ public static void main(String[] args) throws Exception { 4. java.lang.NoSuchMethodError: setByteArray -**原因**：taos-jdbcdriver 3.* 版本仅支持 TDengine 3.0 及以上版本。 +**原因**：taos-jdbcdriver 3.\* 版本仅支持 TDengine 3.0 及以上版本。 -**解决方法**： 使用 taos-jdbcdriver 2.* 版本连接 TDengine 2.* 版本。 +**解决方法**： 使用 taos-jdbcdriver 2.\* 版本连接 TDengine 2.\* 版本。 5. java.lang.NoSuchMethodError: java.nio.ByteBuffer.position(I)Ljava/nio/ByteBuffer; ... taos-jdbcdriver-3.0.1.jar diff --git a/docs/zh/08-connector/26-rust.mdx b/docs/zh/08-connector/26-rust.mdx index eeb93a6558..d4ca25be81 100644 --- a/docs/zh/08-connector/26-rust.mdx +++ b/docs/zh/08-connector/26-rust.mdx @@ -10,6 +10,7 @@ import TabItem from '@theme/TabItem'; import Preparation from "./_preparation.mdx" import RustInsert from "../07-develop/03-insert-data/_rust_sql.mdx" import RustBind from "../07-develop/03-insert-data/_rust_stmt.mdx" +import RustSml from "../07-develop/03-insert-data/_rust_schemaless.mdx" import RustQuery from "../07-develop/04-query-data/_rust.mdx" [![Crates.io](https://img.shields.io/crates/v/taos)](https://crates.io/crates/taos) ![Crates.io](https://img.shields.io/crates/d/taos) [![docs.rs](https://img.shields.io/docsrs/taos)](https://docs.rs/taos) @@ -230,6 +231,10 @@ async fn demo(taos: &Taos, db: &str) -> Result<(), Error> { +#### Schemaless 写入 + + + ### 查询数据 diff --git a/docs/zh/08-connector/index.md b/docs/zh/08-connector/index.md index eb1f3a9a9a..bb8c95a15a 100644 --- a/docs/zh/08-connector/index.md +++ b/docs/zh/08-connector/index.md @@ -61,7 +61,7 @@ TDengine 版本更新往往会增加新的功能特性，列表中的连接器 | **普通查询** | 支持 | 支持 | 支持 | 支持 | 支持 | 支持 | | **参数绑定** | 暂不支持 | 暂不支持 | 支持 | 支持 | 暂不支持 | 支持 | | **数据订阅（TMQ）** | 支持 | 支持 | 支持 | 暂不支持 | 暂不支持 | 支持 | -| **Schemaless** | 暂不支持 | 暂不支持 | 暂不支持 | 暂不支持 | 暂不支持 | 暂不支持 | +| **Schemaless** | 支持 | 暂不支持 | 暂不支持 | 暂不支持 | 暂不支持 | 暂不支持 | | **批量拉取（基于 WebSocket）** | 支持 | 支持 | 支持 | 支持 | 支持 | 支持 | | **DataFrame** | 不支持 | 支持 | 不支持 | 不支持 | 不支持 | 不支持 | diff --git a/docs/zh/12-taos-sql/02-database.md b/docs/zh/12-taos-sql/02-database.md index dd30635660..a2a0914120 100644 --- a/docs/zh/12-taos-sql/02-database.md +++ b/docs/zh/12-taos-sql/02-database.md @@ -71,8 +71,8 @@ database_option: { - 0：表示可以创建多张超级表。 - 1：表示只可以创建一张超级表。 - STT_TRIGGER：表示落盘文件触发文件合并的个数。默认为 1，范围 1 到 16。对于少表高频场景，此参数建议使用默认配置，或较小的值；而对于多表低频场景，此参数建议配置较大的值。 -- TABLE_PREFIX：内部存储引擎根据表名分配存储该表数据的 VNODE 时要忽略的前缀的长度。 -- TABLE_SUFFIX：内部存储引擎根据表名分配存储该表数据的 VNODE 时要忽略的后缀的长度。 +- TABLE_PREFIX：当其为正值时，在决定把一个表分配到哪个 vgroup 时要忽略表名中指定长度的前缀；当其为负值时，在决定把一个表分配到哪个 vgroup 时只使用表名中指定长度的前缀；例如，假定表名为 "v30001"，当 TSDB_PREFIX = 2 时 使用 "0001" 来决定分配到哪个 vgroup ，当 TSDB_PREFIX = -2 时使用 "v3" 来决定分配到哪个 vgroup +- TABLE_SUFFIX：当其为正值时，在决定把一个表分配到哪个 vgroup 时要忽略表名中指定长度的后缀；当其为负值时，在决定把一个表分配到哪个 vgroup 时只使用表名中指定长度的后缀；例如，假定表名为 "v30001"，当 TSDB_SUFFIX = 2 时 使用 "v300" 来决定分配到哪个 vgroup ，当 TSDB_SUFFIX = -2 时使用 "01" 来决定分配到哪个 vgroup。 - TSDB_PAGESIZE：一个 VNODE 中时序数据存储引擎的页大小，单位为 KB，默认为 4 KB。范围为 1 到 16384，即 1 KB到 16 MB。 - WAL_RETENTION_PERIOD: 为了数据订阅消费，需要WAL日志文件额外保留的最大时长策略。WAL日志清理，不受订阅客户端消费状态影响。单位为 s。默认为 0，表示无需为订阅保留。新建订阅，应先设置恰当的时长策略。 - WAL_RETENTION_SIZE：为了数据订阅消费，需要WAL日志文件额外保留的最大累计大小策略。单位为 KB。默认为 0，表示累计大小无上限。 diff --git a/docs/zh/12-taos-sql/14-stream.md b/docs/zh/12-taos-sql/14-stream.md index 12e466f349..634f50356d 100644 --- a/docs/zh/12-taos-sql/14-stream.md +++ b/docs/zh/12-taos-sql/14-stream.md @@ -15,6 +15,7 @@ stream_options: { IGNORE EXPIRED [0|1] DELETE_MARK time FILL_HISTORY [0|1] + IGNORE UPDATE [0|1] } ``` @@ -169,7 +170,7 @@ T3 时刻，最新事件到达，T 向后推移超过了第二个窗口关闭的 在 window_close 或 max_delay 模式下，窗口关闭直接影响推送结果。在 at_once 模式下，窗口关闭只与内存占用有关。 -## 流式计算的过期数据处理策略 +## 流式计算对于过期数据的处理策略 对于已关闭的窗口，再次落入该窗口中的数据被标记为过期数据. @@ -177,11 +178,20 @@ TDengine 对于过期数据提供两种处理方式，由 IGNORE EXPIRED 选项 1. 重新计算，即 IGNORE EXPIRED 0：从 TSDB 中重新查找对应窗口的所有数据并重新计算得到最新结果 -2. 直接丢弃, 即 IGNORE EXPIRED 1：默认配置，忽略过期数据 +2. 直接丢弃，即 IGNORE EXPIRED 1：默认配置，忽略过期数据 无论在哪种模式下，watermark 都应该被妥善设置，来得到正确结果（直接丢弃模式）或避免频繁触发重算带来的性能开销（重新计算模式）。 +## 流式计算对于修改数据的处理策略 + +TDengine 对于修改数据提供两种处理方式，由 IGNORE UPDATE 选项指定： + +1. 检查数据是否被修改，即 IGNORE UPDATE 0：默认配置，如果被修改，则重新计算对应窗口。 + +2. 不检查数据是否被修改，全部按增量数据计算，即 IGNORE UPDATE 1。 + + ## 写入已存在的超级表 ```sql [field1_name,...] @@ -213,3 +223,29 @@ DELETE_MARK time ``` DELETE_MARK用于删除缓存的窗口状态，也就是删除流计算的中间结果。如果不设置，默认值是10年 T = 最新事件时间 - DELETE_MARK + +## 流式计算支持的函数 + +1. 所有的 [单行函数](../function/#单行函数) 均可用于流计算。 +2. 以下 19 个聚合/选择函数 不能 应用在创建流计算的 SQL 语句。此外的其他类型的函数均可用于流计算。 + +- [leastsquares](../function/#leastsquares) +- [percentile](../function/#percentile) +- [top](../function/#top) +- [bottom](../function/#bottom) +- [elapsed](../function/#elapsed) +- [interp](../function/#interp) +- [derivative](../function/#derivative) +- [irate](../function/#irate) +- [twa](../function/#twa) +- [histogram](../function/#histogram) +- [diff](../function/#diff) +- [statecount](../function/#statecount) +- [stateduration](../function/#stateduration) +- [csum](../function/#csum) +- [mavg](../function/#mavg) +- [sample](../function/#sample) +- [tail](../function/#tail) +- [unique](../function/#unique) +- [mode](../function/#mode) + diff --git a/docs/zh/12-taos-sql/22-meta.md b/docs/zh/12-taos-sql/22-meta.md index 1f2e3fb7d5..7fb60b85a7 100644 --- a/docs/zh/12-taos-sql/22-meta.md +++ b/docs/zh/12-taos-sql/22-meta.md @@ -120,6 +120,10 @@ TDengine 内置了一个名为 `INFORMATION_SCHEMA` 的数据库，提供对数 | 5 | create_time | TIMESTAMP | 创建时间 | | 6 | code_len | INT | 代码长度 | | 7 | bufsize | INT | buffer 大小 | +| 8 | func_language | BINARY(31) | 自定义函数编程语言 | +| 9 | func_body | BINARY(16384) | 函数体定义 | +| 10 | func_version | INT | 函数版本号。初始版本为0，每次替换更新，版本号加1。| + ## INS_INDEXES diff --git a/docs/zh/12-taos-sql/26-udf.md b/docs/zh/12-taos-sql/26-udf.md index 7697944f9a..c1d2761d7d 100644 --- a/docs/zh/12-taos-sql/26-udf.md +++ b/docs/zh/12-taos-sql/26-udf.md @@ -13,27 +13,34 @@ description: 使用 UDF 的详细指南 - 创建标量函数 ```sql -CREATE FUNCTION function_name AS library_path OUTPUTTYPE output_type; +CREATE [OR REPLACE] FUNCTION function_name AS library_path OUTPUTTYPE output_type [LANGUAGE 'C|Python']; ``` - - - function_name：标量函数未来在 SQL 中被调用时的函数名，必须与函数实现中 udf 的实际名称一致； - - library_path：包含 UDF 函数实现的动态链接库的库文件绝对路径（指的是库文件在当前客户端所在主机上的保存路径，通常是指向一个 .so 文件），这个路径需要用英文单引号或英文双引号括起来； + - OR REPLACE: 如果函数已经存在，会修改已有的函数属性。 + - function_name：标量函数未来在 SQL 中被调用时的函数名； + - LANGUAGE 'C|Python'：函数编程语言，目前支持C语言和Python语言。 如果这个从句忽略，编程语言是C语言 + - library_path：如果编程语言是C，路径是包含 UDF 函数实现的动态链接库的库文件绝对路径（指的是库文件在当前客户端所在主机上的保存路径，通常是指向一个 .so 文件）。如果编程语言是Python，路径是包含 UDF 函数实现的Python文件路径。这个路径需要用英文单引号或英文双引号括起来； - output_type：此函数计算结果的数据类型名称； - 例如，如下语句可以把 libbitand.so 创建为系统中可用的 UDF： +例如，如下语句可以把 libbitand.so 创建为系统中可用的 UDF： ```sql CREATE FUNCTION bit_and AS "/home/taos/udf_example/libbitand.so" OUTPUTTYPE INT; ``` +例如，使用以下语句可以修改已经定义的 bit_and 函数，输出类型是 BIGINT，使用Python语言实现。 + + ```sql + CREATE OR REPLACE FUNCTION bit_and AS "/home/taos/udf_example/bit_and.py" OUTPUTTYPE BIGINT LANGUAGE 'Python'; + ``` - 创建聚合函数： ```sql -CREATE AGGREGATE FUNCTION function_name AS library_path OUTPUTTYPE output_type [ BUFSIZE buffer_size ]; +CREATE [OR REPLACE] AGGREGATE FUNCTION function_name AS library_path OUTPUTTYPE output_type [ BUFSIZE buffer_size ] [LANGUAGE 'C|Python']; ``` - + - OR REPLACE: 如果函数已经存在，会修改已有的函数属性。 - function_name：聚合函数未来在 SQL 中被调用时的函数名，必须与函数实现中 udfNormalFunc 的实际名称一致； - - library_path：包含 UDF 函数实现的动态链接库的库文件绝对路径（指的是库文件在当前客户端所在主机上的保存路径，通常是指向一个 .so 文件），这个路径需要用英文单引号或英文双引号括起来； - - output_type：此函数计算结果的数据类型，与上文中 udfNormalFunc 的 itype 参数不同，这里不是使用数字表示法，而是直接写类型名称即可； + - LANGUAGE 'C|Python'：函数编程语言，目前支持C语言和Python语言。 + - library_path：如果编程语言是C，路径是包含 UDF 函数实现的动态链接库的库文件绝对路径（指的是库文件在当前客户端所在主机上的保存路径，通常是指向一个 .so 文件）。如果编程语言是Python，路径是包含 UDF 函数实现的Python文件路径。这个路径需要用英文单引号或英文双引号括起来；； + - output_type：此函数计算结果的数据类型名称； - buffer_size：中间计算结果的缓冲区大小，单位是字节。如果不使用可以不设置。 例如，如下语句可以把 libl2norm.so 创建为系统中可用的 UDF： @@ -41,6 +48,11 @@ CREATE AGGREGATE FUNCTION function_name AS library_path OUTPUTTYPE output_type [ ```sql CREATE AGGREGATE FUNCTION l2norm AS "/home/taos/udf_example/libl2norm.so" OUTPUTTYPE DOUBLE bufsize 8; ``` + 例如，使用以下语句可以修改已经定义的 l2norm 函数的缓冲区大小为64。 + ```sql + CREATE AGGREGATE FUNCTION l2norm AS "/home/taos/udf_example/libl2norm.so" OUTPUTTYPE DOUBLE bufsize 64; + ``` + 关于如何开发自定义函数，请参考 [UDF使用说明](/develop/udf)。 ## 管理 UDF diff --git a/docs/zh/12-taos-sql/29-changes.md b/docs/zh/12-taos-sql/29-changes.md index a797966f57..27dd3294b7 100644 --- a/docs/zh/12-taos-sql/29-changes.md +++ b/docs/zh/12-taos-sql/29-changes.md @@ -27,13 +27,13 @@ description: "TDengine 3.0 版本的语法变更说明" | - | :------- | :-------- | :------- | | 1 | ALTER ACCOUNT | 废除 | 2.x中为企业版功能，3.0不再支持。语法暂时保留了，执行报“This statement is no longer supported”错误。 | 2 | ALTER ALL DNODES | 新增 | 修改所有DNODE的参数。 -| 3 | ALTER DATABASE | 调整 | 废除
  • QUORUM：写入需要的副本确认数。3.0 版本默认行为是强一致性，且不支持修改为弱一致性。
  • BLOCKS：VNODE使用的内存块数。3.0版本使用BUFFER来表示VNODE写入内存池的大小。
  • UPDATE：更新操作的支持模式。3.0版本所有数据库都支持部分列更新。
  • CACHELAST：缓存最新一行数据的模式。3.0版本用CACHEMODEL代替。
  • COMP：3.0版本暂不支持修改。

    • 新增
  • CACHEMODEL：表示是否在内存中缓存子表的最近数据。
  • CACHESIZE：表示缓存子表最近数据的内存大小。
  • WAL_FSYNC_PERIOD：代替原FSYNC参数。
  • WAL_LEVEL：代替原WAL参数。
  • WAL_RETENTION_PERIOD：3.0.4.0版本新增，wal文件的额外保留策略，用于数据订阅。
  • WAL_RETENTION_SIZE：3.0.4.0版本新增，wal文件的额外保留策略，用于数据订阅。
    调整
  • REPLICA：3.0.0版本暂不支持修改。
  • KEEP：3.0版本新增支持带单位的设置方式。
+| 3 | ALTER DATABASE | 调整 |

废除

  • QUORUM：写入需要的副本确认数。3.0 版本默认行为是强一致性，且不支持修改为弱一致性。
  • BLOCKS：VNODE使用的内存块数。3.0版本使用BUFFER来表示VNODE写入内存池的大小。
  • UPDATE：更新操作的支持模式。3.0版本所有数据库都支持部分列更新。
  • CACHELAST：缓存最新一行数据的模式。3.0版本用CACHEMODEL代替。
  • COMP：3.0版本暂不支持修改。

新增

  • CACHEMODEL：表示是否在内存中缓存子表的最近数据。
  • CACHESIZE：表示缓存子表最近数据的内存大小。
  • WAL_FSYNC_PERIOD：代替原FSYNC参数。
  • WAL_LEVEL：代替原WAL参数。
  • WAL_RETENTION_PERIOD：3.0.4.0版本新增，wal文件的额外保留策略，用于数据订阅。
  • WAL_RETENTION_SIZE：3.0.4.0版本新增，wal文件的额外保留策略，用于数据订阅。

调整

  • KEEP：3.0版本新增支持带单位的设置方式。
| 4 | ALTER STABLE | 调整 | 废除
  • CHANGE TAG：修改标签列的名称。3.0版本使用RENAME TAG代替。
    新增
  • RENAME TAG：代替原CHANGE TAG子句。
  • COMMENT：修改超级表的注释。
| 5 | ALTER TABLE | 调整 | 废除
  • CHANGE TAG：修改标签列的名称。3.0版本使用RENAME TAG代替。
    新增
  • RENAME TAG：代替原CHANGE TAG子句。
  • COMMENT：修改表的注释。
  • TTL：修改表的生命周期。
| 6 | ALTER USER | 调整 | 废除
  • PRIVILEGE：修改用户权限。3.0版本使用GRANT和REVOKE来授予和回收权限。
    新增
  • ENABLE：启用或停用此用户。
  • SYSINFO：修改用户是否可查看系统信息。
| 7 | COMPACT VNODES | 暂不支持 | 整理指定VNODE的数据。3.0.0版本暂不支持。 | 8 | CREATE ACCOUNT | 废除 | 2.x中为企业版功能，3.0不再支持。语法暂时保留了，执行报“This statement is no longer supported”错误。 -| 9 | CREATE DATABASE | 调整 | 废除
  • BLOCKS：VNODE使用的内存块数。3.0版本使用BUFFER来表示VNODE写入内存池的大小。
  • CACHE：VNODE使用的内存块的大小。3.0版本使用BUFFER来表示VNODE写入内存池的大小。
  • CACHELAST：缓存最新一行数据的模式。3.0版本用CACHEMODEL代替。
  • DAYS：数据文件存储数据的时间跨度。3.0版本使用DURATION代替。
  • FSYNC：当 WAL 设置为 2 时，执行 fsync 的周期。3.0版本使用WAL_FSYNC_PERIOD代替。
  • QUORUM：写入需要的副本确认数。3.0版本使用STRICT来指定强一致还是弱一致。
  • UPDATE：更新操作的支持模式。3.0版本所有数据库都支持部分列更新。
  • WAL：WAL 级别。3.0版本使用WAL_LEVEL代替。
    新增
  • BUFFER：一个 VNODE 写入内存池大小。
  • CACHEMODEL：表示是否在内存中缓存子表的最近数据。
  • CACHESIZE：表示缓存子表最近数据的内存大小。
  • DURATION：代替原DAYS参数。新增支持带单位的设置方式。
  • PAGES：一个 VNODE 中元数据存储引擎的缓存页个数。
  • PAGESIZE：一个 VNODE 中元数据存储引擎的页大小。
  • RETENTIONS：表示数据的聚合周期和保存时长。
  • STRICT：表示数据同步的一致性要求。
  • SINGLE_STABLE：表示此数据库中是否只可以创建一个超级表。
  • VGROUPS：数据库中初始VGROUP的数目。
  • WAL_FSYNC_PERIOD：代替原FSYNC参数。
  • WAL_LEVEL：代替原WAL参数。
  • WAL_RETENTION_PERIOD：wal文件的额外保留策略，用于数据订阅。
  • WAL_RETENTION_SIZE：wal文件的额外保留策略，用于数据订阅。
  • WAL_ROLL_PERIOD：wal文件切换时长。
  • WAL_SEGMENT_SIZE：wal单个文件大小。
    调整
  • KEEP：3.0版本新增支持带单位的设置方式。
+| 9 | CREATE DATABASE | 调整 |

废除

  • BLOCKS：VNODE使用的内存块数。3.0版本使用BUFFER来表示VNODE写入内存池的大小。
  • CACHE：VNODE使用的内存块的大小。3.0版本使用BUFFER来表示VNODE写入内存池的大小。
  • CACHELAST：缓存最新一行数据的模式。3.0版本用CACHEMODEL代替。
  • DAYS：数据文件存储数据的时间跨度。3.0版本使用DURATION代替。
  • FSYNC：当 WAL 设置为 2 时，执行 fsync 的周期。3.0版本使用WAL_FSYNC_PERIOD代替。
  • QUORUM：写入需要的副本确认数。3.0版本使用STRICT来指定强一致还是弱一致。
  • UPDATE：更新操作的支持模式。3.0版本所有数据库都支持部分列更新。
  • WAL：WAL 级别。3.0版本使用WAL_LEVEL代替。

新增

  • BUFFER：一个 VNODE 写入内存池大小。
  • CACHEMODEL：表示是否在内存中缓存子表的最近数据。
  • CACHESIZE：表示缓存子表最近数据的内存大小。
  • DURATION：代替原DAYS参数。新增支持带单位的设置方式。
  • PAGES：一个 VNODE 中元数据存储引擎的缓存页个数。
  • PAGESIZE：一个 VNODE 中元数据存储引擎的页大小。
  • RETENTIONS：表示数据的聚合周期和保存时长。
  • STRICT：表示数据同步的一致性要求。
  • SINGLE_STABLE：表示此数据库中是否只可以创建一个超级表。
  • VGROUPS：数据库中初始VGROUP的数目。
  • WAL_FSYNC_PERIOD：代替原FSYNC参数。
  • WAL_LEVEL：代替原WAL参数。
  • WAL_RETENTION_PERIOD：wal文件的额外保留策略，用于数据订阅。
  • WAL_RETENTION_SIZE：wal文件的额外保留策略，用于数据订阅。
  • WAL_ROLL_PERIOD：wal文件切换时长。
  • WAL_SEGMENT_SIZE：wal单个文件大小。

调整

  • KEEP：3.0版本新增支持带单位的设置方式。
| 10 | CREATE DNODE | 调整 | 新增主机名和端口号分开指定语法
  • CREATE DNODE dnode_host_name PORT port_val
| 11 | CREATE INDEX | 新增 | 创建SMA索引。 | 12 | CREATE MNODE | 新增 | 创建管理节点。 diff --git a/docs/zh/14-reference/06-taosdump.md b/docs/zh/14-reference/06-taosdump.md index 8a031d1473..8ff1287c3e 100644 --- a/docs/zh/14-reference/06-taosdump.md +++ b/docs/zh/14-reference/06-taosdump.md @@ -79,6 +79,7 @@ Usage: taosdump [OPTION...] dbname [tbname ...] -A, --all-databases Dump all databases. -D, --databases=DATABASES Dump inputted databases. Use comma to separate databases' name. + -e, --escape-character Use escaped character for database name -N, --without-property Dump database without its properties. -s, --schemaonly Only dump tables' schema. -y, --answer-yes Input yes for prompt. It will skip data file diff --git a/docs/zh/14-reference/12-config/index.md b/docs/zh/14-reference/12-config/index.md index fe23684fde..c1d23df409 100644 --- a/docs/zh/14-reference/12-config/index.md +++ b/docs/zh/14-reference/12-config/index.md @@ -356,7 +356,7 @@ charset 的有效值是 UTF-8。 | 适用范围 | 仅服务端适用 | | 含义 | dnode 支持的最大 vnode 数目 | | 取值范围 | 0-4096 | -| 缺省值 | CPU 核数的 2 倍 | +| 缺省值 | CPU 核数的 2 倍 | ## 性能调优 @@ -366,6 +366,7 @@ charset 的有效值是 UTF-8。 | -------- | ---------------------- | | 适用范围 | 仅服务端适用 | | 含义 | 设置写入线程的最大数量 | +| 取值范围 | 0-1024 | | 缺省值 | | ## 日志相关 diff --git a/docs/zh/14-reference/14-taosKeeper.md b/docs/zh/14-reference/14-taosKeeper.md index 12b609584a..b4d35fb240 100644 --- a/docs/zh/14-reference/14-taosKeeper.md +++ b/docs/zh/14-reference/14-taosKeeper.md @@ -111,7 +111,7 @@ Active: inactive (dead) #### 配置文件启动 -执行以下命令即可快速体验 taosKeeper。当不指定 taosKeeper 配置文件时，优先使用 `/etc/taos/keeper.toml` 配置，否则将使用默认配置。 +执行以下命令即可快速体验 taosKeeper。当不指定 taosKeeper 配置文件时，优先使用 `/etc/taos/taoskeeper.toml` 配置，否则将使用默认配置。 ```shell $ taoskeeper -c @@ -156,6 +156,10 @@ database = "log" # 指定需要监控的普通表 tables = [] + +# database options for db storing metrics data +[metrics.databaseoptions] +cachemodel = "none" ``` ### 获取监控指标 @@ -206,7 +210,7 @@ taos_cluster_info_dnodes_total{cluster_id="5981392874047724755"} 1 taos_cluster_info_first_ep{cluster_id="5981392874047724755",value="hlb:6030"} 1 ``` -### check_health +### check\_health ``` $ curl -i http://127.0.0.1:6043/check_health @@ -222,3 +226,29 @@ Content-Length: 19 {"version":"1.0.0"} ``` + +### 集成 Prometheus + +taoskeeper 提供了 `/metrics` 接口，返回了 Prometheus 格式的监控数据，Prometheus 可以从 taoskeeper 抽取监控数据，实现通过 Prometheus 监控 TDengine 的目的。 + +#### 抽取配置 + +Prometheus 提供了 `scrape_configs` 配置如何从 endpoint 抽取监控数据，通常只需要修改 `static_configs` 中的 targets 配置为 taoskeeper 的 endpoint 地址，更多配置信息请参考 [Prometheus 配置文档](https://prometheus.io/docs/prometheus/latest/configuration/configuration/#scrape_config)。 + +``` +# A scrape configuration containing exactly one endpoint to scrape: +# Here it's Prometheus itself. +scrape_configs: + - job_name: "taoskeeper" + # metrics_path defaults to '/metrics' + # scheme defaults to 'http'. + static_configs: + - targets: ["localhost:6043"] +``` + +#### Dashboard + +我们提供了 `TaosKeeper Prometheus Dashboard for 3.x` dashboard，提供了和 TDinsight 类似的监控 dashboard。 + +在 Grafana Dashboard 菜单点击 `import`，dashboard ID 填写 `18587`，点击 `Load` 按钮即可导入 `TaosKeeper Prometheus Dashboard for 3.x` dashboard。 + diff --git a/docs/zh/17-operation/10-monitor.md b/docs/zh/17-operation/10-monitor.md index 01a2257286..7def90916c 100644 --- a/docs/zh/17-operation/10-monitor.md +++ b/docs/zh/17-operation/10-monitor.md @@ -54,7 +54,7 @@ TDinsight dashboard 数据来源于 log 库（存放监控数据的默认db， |first\_ep\_dnode\_id|INT||集群 first ep 的 dnode id| |version|VARCHAR||tdengine version。例如：3.0.4.0| |master\_uptime|FLOAT||当前 master 节点的uptime。单位：天| -|monitor_interval|INT||monitor interval。单位：秒| +|monitor\_interval|INT||monitor interval。单位：秒| |dbs\_total|INT||database 总数| |tbs\_total|BIGINT||当前集群 table 总数| |stbs\_total|INT||当前集群 stable 总数| @@ -107,17 +107,17 @@ TDinsight dashboard 数据来源于 log 库（存放监控数据的默认db， |cpu\_system|FLOAT||服务器 cpu 使用率，从 `/proc/stat` 读取| |cpu\_cores|FLOAT||服务器 cpu 核数| |mem\_engine|INT||taosd 内存使用率，从 `/proc//status` 读取| -|mem\_system|INT||服务器内存使用率| +|mem\_system|INT||服务器可用内存| |mem\_total|INT||服务器内存总量，单位 KB| |disk\_engine|INT||| |disk\_used|BIGINT||data dir 挂载的磁盘使用量，单位 bytes| |disk\_total|BIGINT||data dir 挂载的磁盘总容量，单位 bytes| -|net\_in|FLOAT||网络吞吐率，从 `/proc/net/dev` 中读取的 received bytes。单位 bytes per second| -|net\_out|FLOAT||网络吞吐率，从 `/proc/net/dev` 中读取的 transmit bytes。单位 bytes per second| -|io\_read|FLOAT||io 吞吐率，从 `/proc//io` 中读取的 rchar 与上次数值计算之后，计算得到速度。单位 bytes per second| -|io\_write|FLOAT||io 吞吐率，从 `/proc//io` 中读取的 wchar 与上次数值计算之后，计算得到速度。单位 bytes per second| -|io\_read\_disk|FLOAT||磁盘 io 吞吐率，从 `/proc//io` 中读取的 read_bytes。单位 bytes per second| -|io\_write\_disk|FLOAT||磁盘 io 吞吐率，从 `/proc//io` 中读取的 write_bytes。单位 bytes per second| +|net\_in|FLOAT||网络吞吐率，从 `/proc/net/dev` 中读取的 received bytes。单位 kb/s| +|net\_out|FLOAT||网络吞吐率，从 `/proc/net/dev` 中读取的 transmit bytes。单位 kb/s| +|io\_read|FLOAT||io 吞吐率，从 `/proc//io` 中读取的 rchar 与上次数值计算之后，计算得到速度。单位 kb/s| +|io\_write|FLOAT||io 吞吐率，从 `/proc//io` 中读取的 wchar 与上次数值计算之后，计算得到速度。单位 kb/s| +|io\_read\_disk|FLOAT||磁盘 io 吞吐率，从 `/proc//io` 中读取的 read_bytes。单位 kb/s| +|io\_write\_disk|FLOAT||磁盘 io 吞吐率，从 `/proc//io` 中读取的 write_bytes。单位 kb/s| |req\_select|INT||两个间隔内发生的查询请求数目| |req\_select\_rate|FLOAT||两个间隔内的查询请求速度 = `req_select / monitorInterval`| |req\_insert|INT||两个间隔内发生的写入请求，包含的单条数据数目| diff --git a/docs/zh/28-releases/01-tdengine.md b/docs/zh/28-releases/01-tdengine.md index 0974289c1f..bea0adfa82 100644 --- a/docs/zh/28-releases/01-tdengine.md +++ b/docs/zh/28-releases/01-tdengine.md @@ -10,6 +10,10 @@ TDengine 2.x 各版本安装包请访问[这里](https://www.taosdata.com/all-do import Release from "/components/ReleaseV3"; +## 3.0.4.1 + + + ## 3.0.4.0 diff --git a/docs/zh/28-releases/02-tools.md b/docs/zh/28-releases/02-tools.md index 78926555f1..e13ec68c2e 100644 --- a/docs/zh/28-releases/02-tools.md +++ b/docs/zh/28-releases/02-tools.md @@ -10,6 +10,14 @@ taosTools 各版本安装包下载链接如下： import Release from "/components/ReleaseV3"; +## 2.5.0 + + + +## 2.5.0 + + + ## 2.4.12 diff --git a/examples/c/tmq.c b/examples/c/tmq.c index 83b40c4f20..94545dfaad 100644 --- a/examples/c/tmq.c +++ b/examples/c/tmq.c @@ -20,7 +20,8 @@ #include #include "taos.h" -static int running = 1; +static int running = 1; +const char* topic_name = "topicname"; static int32_t msg_process(TAOS_RES* msg) { char buf[1024]; @@ -243,7 +244,7 @@ _end: tmq_list_t* build_topic_list() { tmq_list_t* topicList = tmq_list_new(); - int32_t code = tmq_list_append(topicList, "topicname"); + int32_t code = tmq_list_append(topicList, topic_name); if (code) { tmq_list_destroy(topicList); return NULL; @@ -269,6 +270,31 @@ void basic_consume_loop(tmq_t* tmq) { fprintf(stderr, "%d msg consumed, include %d rows\n", msgCnt, totalRows); } +void consume_repeatly(tmq_t* tmq) { + int32_t numOfAssignment = 0; + tmq_topic_assignment* pAssign = NULL; + + int32_t code = tmq_get_topic_assignment(tmq, topic_name, &pAssign, &numOfAssignment); + if (code != 0) { + fprintf(stderr, "failed to get assignment, reason:%s", tmq_err2str(code)); + } + + // seek to the earliest offset + for(int32_t i = 0; i < numOfAssignment; ++i) { + tmq_topic_assignment* p = &pAssign[i]; + + code = tmq_offset_seek(tmq, topic_name, p->vgId, p->begin); + if (code != 0) { + fprintf(stderr, "failed to seek to %ld, reason:%s", p->begin, tmq_err2str(code)); + } + } + + free(pAssign); + + // let's do it again + basic_consume_loop(tmq); +} + int main(int argc, char* argv[]) { int32_t code; @@ -294,10 +320,13 @@ int main(int argc, char* argv[]) { if ((code = tmq_subscribe(tmq, topic_list))) { fprintf(stderr, "Failed to tmq_subscribe(): %s\n", tmq_err2str(code)); } + tmq_list_destroy(topic_list); basic_consume_loop(tmq); + consume_repeatly(tmq); + code = tmq_consumer_close(tmq); if (code) { fprintf(stderr, "Failed to close consumer: %s\n", tmq_err2str(code)); diff --git a/include/client/taos.h b/include/client/taos.h index a59e203644..79aab247ab 100644 --- a/include/client/taos.h +++ b/include/client/taos.h @@ -101,6 +101,7 @@ typedef struct TAOS_FIELD_E { #endif typedef void (*__taos_async_fn_t)(void *param, TAOS_RES *res, int code); +typedef void (*__taos_notify_fn_t)(void *param, void *ext, int type); typedef struct TAOS_MULTI_BIND { int buffer_type; @@ -121,6 +122,10 @@ typedef enum { SET_CONF_RET_ERR_TOO_LONG = -6 } SET_CONF_RET_CODE; +typedef enum { + TAOS_NOTIFY_PASSVER = 0, +} TAOS_NOTIFY_TYPE; + #define RET_MSG_LENGTH 1024 typedef struct setConfRet { SET_CONF_RET_CODE retCode; @@ -162,7 +167,7 @@ DLL_EXPORT int taos_stmt_set_sub_tbname(TAOS_STMT *stmt, const char *name DLL_EXPORT int taos_stmt_get_tag_fields(TAOS_STMT *stmt, int *fieldNum, TAOS_FIELD_E **fields); DLL_EXPORT int taos_stmt_get_col_fields(TAOS_STMT *stmt, int *fieldNum, TAOS_FIELD_E **fields); // let stmt to reclaim TAOS_FIELD_E that was allocated by `taos_stmt_get_tag_fields`/`taos_stmt_get_col_fields` -DLL_EXPORT void taos_stmt_reclaim_fields(TAOS_STMT *stmt, TAOS_FIELD_E *fields); +DLL_EXPORT void taos_stmt_reclaim_fields(TAOS_STMT *stmt, TAOS_FIELD_E *fields); DLL_EXPORT int taos_stmt_is_insert(TAOS_STMT *stmt, int *insert); DLL_EXPORT int taos_stmt_num_params(TAOS_STMT *stmt, int *nums); @@ -227,6 +232,7 @@ DLL_EXPORT int taos_load_table_info(TAOS *taos, const char *tableNameList); // set heart beat thread quit mode , if quicByKill 1 then kill thread else quit from inner DLL_EXPORT void taos_set_hb_quit(int8_t quitByKill); +DLL_EXPORT int taos_set_notify_cb(TAOS *taos, __taos_notify_fn_t fp, void *param, int type); /* --------------------------schemaless INTERFACE------------------------------- */ @@ -265,6 +271,12 @@ DLL_EXPORT tmq_t *tmq_consumer_new(tmq_conf_t *conf, char *errstr, int32_t errst DLL_EXPORT const char *tmq_err2str(int32_t code); /* ------------------------TMQ CONSUMER INTERFACE------------------------ */ +typedef struct tmq_topic_assignment { + int32_t vgId; + int64_t currentOffset; + int64_t begin; + int64_t end; +} tmq_topic_assignment; DLL_EXPORT int32_t tmq_subscribe(tmq_t *tmq, const tmq_list_t *topic_list); DLL_EXPORT int32_t tmq_unsubscribe(tmq_t *tmq); @@ -273,6 +285,9 @@ DLL_EXPORT TAOS_RES *tmq_consumer_poll(tmq_t *tmq, int64_t timeout); DLL_EXPORT int32_t tmq_consumer_close(tmq_t *tmq); DLL_EXPORT int32_t tmq_commit_sync(tmq_t *tmq, const TAOS_RES *msg); DLL_EXPORT void tmq_commit_async(tmq_t *tmq, const TAOS_RES *msg, tmq_commit_cb *cb, void *param); +DLL_EXPORT int32_t tmq_get_topic_assignment(tmq_t *tmq, const char *pTopicName, tmq_topic_assignment **assignment, + int32_t *numOfAssignment); +DLL_EXPORT int32_t tmq_offset_seek(tmq_t *tmq, const char *pTopicName, int32_t vgId, int64_t offset); /* ----------------------TMQ CONFIGURATION INTERFACE---------------------- */ diff --git a/include/common/tcommon.h b/include/common/tcommon.h index 2bc67e439f..0544247da2 100644 --- a/include/common/tcommon.h +++ b/include/common/tcommon.h @@ -128,6 +128,7 @@ enum { TMQ_MSG_TYPE__POLL_META_RSP, TMQ_MSG_TYPE__EP_RSP, TMQ_MSG_TYPE__TAOSX_RSP, + TMQ_MSG_TYPE__WALINFO_RSP, TMQ_MSG_TYPE__END_RSP, }; diff --git a/include/common/tglobal.h b/include/common/tglobal.h index 2368614e88..8509d39793 100644 --- a/include/common/tglobal.h +++ b/include/common/tglobal.h @@ -24,6 +24,12 @@ extern "C" { #endif +#define SLOW_LOG_TYPE_QUERY 0x1 +#define SLOW_LOG_TYPE_INSERT 0x2 +#define SLOW_LOG_TYPE_OTHERS 0x4 +#define SLOW_LOG_TYPE_ALL 0xFFFFFFFF + + // cluster extern char tsFirst[]; extern char tsSecond[]; @@ -118,6 +124,8 @@ extern int32_t tsRedirectFactor; extern int32_t tsRedirectMaxPeriod; extern int32_t tsMaxRetryWaitTime; extern bool tsUseAdapter; +extern int32_t tsSlowLogThreshold; +extern int32_t tsSlowLogScope; // client extern int32_t tsMinSlidingTime; diff --git a/include/common/tmsg.h b/include/common/tmsg.h index 4cca2d0311..cbbdd169e6 100644 --- a/include/common/tmsg.h +++ b/include/common/tmsg.h @@ -106,6 +106,7 @@ enum { HEARTBEAT_KEY_DBINFO, HEARTBEAT_KEY_STBINFO, HEARTBEAT_KEY_TMQ, + HEARTBEAT_KEY_USER_PASSINFO, }; typedef enum _mgmt_table { @@ -634,6 +635,7 @@ typedef struct { int8_t connType; SEpSet epSet; int32_t svrTimestamp; + int32_t passVer; char sVer[TSDB_VERSION_LEN]; char sDetailVer[128]; } SConnectRsp; @@ -717,6 +719,14 @@ int32_t tSerializeSGetUserAuthRsp(void* buf, int32_t bufLen, SGetUserAuthRsp* pR int32_t tDeserializeSGetUserAuthRsp(void* buf, int32_t bufLen, SGetUserAuthRsp* pRsp); void tFreeSGetUserAuthRsp(SGetUserAuthRsp* pRsp); +typedef struct SUserPassVersion { + char user[TSDB_USER_LEN]; + int32_t version; +} SUserPassVersion; + +typedef SGetUserAuthReq SGetUserPassReq; +typedef SUserPassVersion SGetUserPassRsp; + /* * for client side struct, only column id, type, bytes are necessary * But for data in vnode side, we need all the following information. @@ -1047,6 +1057,14 @@ int32_t tSerializeSUserAuthBatchRsp(void* buf, int32_t bufLen, SUserAuthBatchRsp int32_t tDeserializeSUserAuthBatchRsp(void* buf, int32_t bufLen, SUserAuthBatchRsp* pRsp); void tFreeSUserAuthBatchRsp(SUserAuthBatchRsp* pRsp); +typedef struct { + SArray* pArray; // Array of SGetUserPassRsp +} SUserPassBatchRsp; + +int32_t tSerializeSUserPassBatchRsp(void* buf, int32_t bufLen, SUserPassBatchRsp* pRsp); +int32_t tDeserializeSUserPassBatchRsp(void* buf, int32_t bufLen, SUserPassBatchRsp* pRsp); +void tFreeSUserPassBatchRsp(SUserPassBatchRsp* pRsp); + typedef struct { char db[TSDB_DB_FNAME_LEN]; STimeWindow timeRange; @@ -1287,6 +1305,9 @@ typedef struct { int16_t hashSuffix; int32_t tsdbPageSize; int64_t reserved[8]; + int8_t learnerReplica; + int8_t learnerSelfIndex; + SReplica learnerReplicas[TSDB_MAX_LEARNER_REPLICA]; } SCreateVnodeReq; int32_t tSerializeSCreateVnodeReq(void* buf, int32_t bufLen, SCreateVnodeReq* pReq); @@ -1358,7 +1379,10 @@ typedef struct { int8_t replica; SReplica replicas[TSDB_MAX_REPLICA]; int64_t reserved[8]; -} SAlterVnodeReplicaReq; + int8_t learnerSelfIndex; + int8_t learnerReplica; + SReplica learnerReplicas[TSDB_MAX_LEARNER_REPLICA]; +} SAlterVnodeReplicaReq, SAlterVnodeTypeReq; int32_t tSerializeSAlterVnodeReplicaReq(void* buf, int32_t bufLen, SAlterVnodeReplicaReq* pReq); int32_t tDeserializeSAlterVnodeReplicaReq(void* buf, int32_t bufLen, SAlterVnodeReplicaReq* pReq); @@ -1630,7 +1654,10 @@ int32_t tDeserializeSCreateDropMQSNodeReq(void* buf, int32_t bufLen, SMCreateQno typedef struct { int8_t replica; SReplica replicas[TSDB_MAX_REPLICA]; -} SDCreateMnodeReq, SDAlterMnodeReq; + int8_t learnerReplica; + SReplica learnerReplicas[TSDB_MAX_LEARNER_REPLICA]; + int64_t lastIndex; +} SDCreateMnodeReq, SDAlterMnodeReq, SDAlterMnodeTypeReq; int32_t tSerializeSDCreateMnodeReq(void* buf, int32_t bufLen, SDCreateMnodeReq* pReq); int32_t tDeserializeSDCreateMnodeReq(void* buf, int32_t bufLen, SDCreateMnodeReq* pReq); @@ -2800,6 +2827,7 @@ typedef struct { } SMqOffset; typedef struct { + int64_t consumerId; int32_t num; SMqOffset* offsets; } SMqCMCommitOffsetReq; @@ -2867,6 +2895,14 @@ typedef struct { int32_t tEncodeSTqOffset(SEncoder* pEncoder, const STqOffset* pOffset); int32_t tDecodeSTqOffset(SDecoder* pDecoder, STqOffset* pOffset); +typedef struct SMqVgOffset { + int64_t consumerId; + STqOffset offset; +} SMqVgOffset; + +int32_t tEncodeMqVgOffset(SEncoder* pEncoder, const SMqVgOffset* pOffset); +int32_t tDecodeMqVgOffset(SDecoder* pDecoder, SMqVgOffset* pOffset); + typedef struct { SMsgHead head; int32_t taskId; @@ -3132,18 +3168,19 @@ typedef struct { int32_t code; int32_t epoch; int64_t consumerId; + int64_t walsver; + int64_t walever; } SMqRspHead; typedef struct { - SMsgHead head; - char subKey[TSDB_SUBSCRIBE_KEY_LEN]; - int8_t withTbName; - int8_t useSnapshot; - int32_t epoch; - uint64_t reqId; - int64_t consumerId; - int64_t timeout; - // int64_t currentOffset; + SMsgHead head; + char subKey[TSDB_SUBSCRIBE_KEY_LEN]; + int8_t withTbName; + int8_t useSnapshot; + int32_t epoch; + uint64_t reqId; + int64_t consumerId; + int64_t timeout; STqOffsetVal reqOffset; } SMqPollReq; @@ -3178,43 +3215,9 @@ typedef struct { SSchemaWrapper schema; } SMqSubTopicEp; -static FORCE_INLINE int32_t tEncodeSMqSubTopicEp(void** buf, const SMqSubTopicEp* pTopicEp) { - int32_t tlen = 0; - tlen += taosEncodeString(buf, pTopicEp->topic); - tlen += taosEncodeString(buf, pTopicEp->db); - int32_t sz = taosArrayGetSize(pTopicEp->vgs); - tlen += taosEncodeFixedI32(buf, sz); - for (int32_t i = 0; i < sz; i++) { - SMqSubVgEp* pVgEp = (SMqSubVgEp*)taosArrayGet(pTopicEp->vgs, i); - tlen += tEncodeSMqSubVgEp(buf, pVgEp); - } - tlen += taosEncodeSSchemaWrapper(buf, &pTopicEp->schema); - return tlen; -} - -static FORCE_INLINE void* tDecodeSMqSubTopicEp(void* buf, SMqSubTopicEp* pTopicEp) { - buf = taosDecodeStringTo(buf, pTopicEp->topic); - buf = taosDecodeStringTo(buf, pTopicEp->db); - int32_t sz; - buf = taosDecodeFixedI32(buf, &sz); - pTopicEp->vgs = taosArrayInit(sz, sizeof(SMqSubVgEp)); - if (pTopicEp->vgs == NULL) { - return NULL; - } - for (int32_t i = 0; i < sz; i++) { - SMqSubVgEp vgEp; - buf = tDecodeSMqSubVgEp(buf, &vgEp); - taosArrayPush(pTopicEp->vgs, &vgEp); - } - buf = taosDecodeSSchemaWrapper(buf, &pTopicEp->schema); - return buf; -} - -static FORCE_INLINE void tDeleteSMqSubTopicEp(SMqSubTopicEp* pSubTopicEp) { - taosMemoryFreeClear(pSubTopicEp->schema.pSchema); - pSubTopicEp->schema.nCols = 0; - taosArrayDestroy(pSubTopicEp->vgs); -} +int32_t tEncodeMqSubTopicEp(void** buf, const SMqSubTopicEp* pTopicEp); +void* tDecodeMqSubTopicEp(void* buf, SMqSubTopicEp* pTopicEp); +void tDeleteMqSubTopicEp(SMqSubTopicEp* pSubTopicEp); typedef struct { SMqRspHead head; @@ -3224,8 +3227,8 @@ typedef struct { void* metaRsp; } SMqMetaRsp; -int32_t tEncodeSMqMetaRsp(SEncoder* pEncoder, const SMqMetaRsp* pRsp); -int32_t tDecodeSMqMetaRsp(SDecoder* pDecoder, SMqMetaRsp* pRsp); +int32_t tEncodeMqMetaRsp(SEncoder* pEncoder, const SMqMetaRsp* pRsp); +int32_t tDecodeMqMetaRsp(SDecoder* pDecoder, SMqMetaRsp* pRsp); typedef struct { SMqRspHead head; @@ -3240,9 +3243,9 @@ typedef struct { SArray* blockSchema; } SMqDataRsp; -int32_t tEncodeSMqDataRsp(SEncoder* pEncoder, const SMqDataRsp* pRsp); -int32_t tDecodeSMqDataRsp(SDecoder* pDecoder, SMqDataRsp* pRsp); -void tDeleteSMqDataRsp(SMqDataRsp* pRsp); +int32_t tEncodeMqDataRsp(SEncoder* pEncoder, const SMqDataRsp* pRsp); +int32_t tDecodeMqDataRsp(SDecoder* pDecoder, SMqDataRsp* pRsp); +void tDeleteMqDataRsp(SMqDataRsp* pRsp); typedef struct { SMqRspHead head; @@ -3278,7 +3281,7 @@ static FORCE_INLINE int32_t tEncodeSMqAskEpRsp(void** buf, const SMqAskEpRsp* pR tlen += taosEncodeFixedI32(buf, sz); for (int32_t i = 0; i < sz; i++) { SMqSubTopicEp* pVgEp = (SMqSubTopicEp*)taosArrayGet(pRsp->topics, i); - tlen += tEncodeSMqSubTopicEp(buf, pVgEp); + tlen += tEncodeMqSubTopicEp(buf, pVgEp); } return tlen; } @@ -3293,14 +3296,14 @@ static FORCE_INLINE void* tDecodeSMqAskEpRsp(void* buf, SMqAskEpRsp* pRsp) { } for (int32_t i = 0; i < sz; i++) { SMqSubTopicEp topicEp; - buf = tDecodeSMqSubTopicEp(buf, &topicEp); + buf = tDecodeMqSubTopicEp(buf, &topicEp); taosArrayPush(pRsp->topics, &topicEp); } return buf; } static FORCE_INLINE void tDeleteSMqAskEpRsp(SMqAskEpRsp* pRsp) { - taosArrayDestroyEx(pRsp->topics, (FDelete)tDeleteSMqSubTopicEp); + taosArrayDestroyEx(pRsp->topics, (FDelete)tDeleteMqSubTopicEp); } #define TD_AUTO_CREATE_TABLE 0x1 diff --git a/include/common/tmsgdef.h b/include/common/tmsgdef.h index 7e42f3ec59..7d877bf146 100644 --- a/include/common/tmsgdef.h +++ b/include/common/tmsgdef.h @@ -83,6 +83,8 @@ enum { TD_DEF_MSG_TYPE(TDMT_DND_CONFIG_DNODE, "config-dnode", NULL, NULL) TD_DEF_MSG_TYPE(TDMT_DND_SYSTABLE_RETRIEVE, "dnode-retrieve", NULL, NULL) TD_DEF_MSG_TYPE(TDMT_DND_MAX_MSG, "dnd-max", NULL, NULL) + TD_DEF_MSG_TYPE(TDMT_DND_ALTER_MNODE_TYPE, "dnode-alter-mnode-type", NULL, NULL) + TD_DEF_MSG_TYPE(TDMT_DND_ALTER_VNODE_TYPE, "dnode-alter-vnode-type", NULL, NULL) TD_NEW_MSG_SEG(TDMT_MND_MSG) TD_DEF_MSG_TYPE(TDMT_MND_CONNECT, "connect", NULL, NULL) @@ -225,7 +227,6 @@ enum { TD_DEF_MSG_TYPE(TDMT_VND_COMMIT, "vnode-commit", NULL, NULL) TD_DEF_MSG_TYPE(TDMT_VND_CREATE_INDEX, "vnode-create-index", NULL, NULL) TD_DEF_MSG_TYPE(TDMT_VND_DROP_INDEX, "vnode-drop-index", NULL, NULL) - TD_DEF_MSG_TYPE(TDMT_VND_DISABLE_WRITE, "vnode-disable-write", NULL, NULL) TD_DEF_MSG_TYPE(TDMT_VND_MAX_MSG, "vnd-max", NULL, NULL) @@ -304,9 +305,11 @@ enum { TD_DEF_MSG_TYPE(TDMT_VND_TMQ_SUBSCRIBE, "vnode-tmq-subscribe", SMqRebVgReq, SMqRebVgRsp) TD_DEF_MSG_TYPE(TDMT_VND_TMQ_DELETE_SUB, "vnode-tmq-delete-sub", SMqVDeleteReq, SMqVDeleteRsp) TD_DEF_MSG_TYPE(TDMT_VND_TMQ_COMMIT_OFFSET, "vnode-tmq-commit-offset", STqOffset, STqOffset) + TD_DEF_MSG_TYPE(TDMT_VND_TMQ_SEEK_TO_OFFSET, "vnode-tmq-seekto-offset", STqOffset, STqOffset) TD_DEF_MSG_TYPE(TDMT_VND_TMQ_ADD_CHECKINFO, "vnode-tmq-add-checkinfo", NULL, NULL) TD_DEF_MSG_TYPE(TDMT_VND_TMQ_DEL_CHECKINFO, "vnode-del-checkinfo", NULL, NULL) TD_DEF_MSG_TYPE(TDMT_VND_TMQ_CONSUME, "vnode-tmq-consume", SMqPollReq, SMqDataBlkRsp) + TD_DEF_MSG_TYPE(TDMT_VND_TMQ_VG_WALINFO, "vnode-tmq-vg-walinfo", SMqPollReq, SMqDataBlkRsp) TD_DEF_MSG_TYPE(TDMT_VND_TMQ_MAX_MSG, "vnd-tmq-max", NULL, NULL) diff --git a/include/dnode/mnode/mnode.h b/include/dnode/mnode/mnode.h index cdb1642a5c..6c3c7497b1 100644 --- a/include/dnode/mnode/mnode.h +++ b/include/dnode/mnode/mnode.h @@ -20,6 +20,7 @@ #include "tmsg.h" #include "tmsgcb.h" #include "trpc.h" +#include "sync.h" #ifdef __cplusplus extern "C" { @@ -33,8 +34,11 @@ typedef struct { bool deploy; int8_t selfIndex; int8_t numOfReplicas; - SReplica replicas[TSDB_MAX_REPLICA]; + int8_t numOfTotalReplicas; + SReplica replicas[TSDB_MAX_REPLICA + TSDB_MAX_LEARNER_REPLICA]; + int32_t nodeRoles[TSDB_MAX_REPLICA + TSDB_MAX_LEARNER_REPLICA]; SMsgCb msgCb; + int64_t lastIndex; } SMnodeOpt; /* ------------------------ SMnode ------------------------ */ @@ -69,6 +73,9 @@ int32_t mndStart(SMnode *pMnode); */ void mndStop(SMnode *pMnode); +int32_t mndIsCatchUp(SMnode *pMnode); +ESyncRole mndGetRole(SMnode *pMnode); + /** * @brief Get mnode monitor info. * diff --git a/include/libs/function/function.h b/include/libs/function/function.h index aa5c78195a..85f7cf7e2c 100644 --- a/include/libs/function/function.h +++ b/include/libs/function/function.h @@ -99,11 +99,11 @@ typedef struct SSubsidiaryResInfo { } SSubsidiaryResInfo; typedef struct SResultDataInfo { - int16_t precision; - int16_t scale; - int16_t type; - int16_t bytes; - int32_t interBufSize; + int16_t precision; + int16_t scale; + int16_t type; + uint16_t bytes; + int32_t interBufSize; } SResultDataInfo; #define GET_RES_INFO(ctx) ((ctx)->resultInfo) diff --git a/include/libs/nodes/plannodes.h b/include/libs/nodes/plannodes.h index ad4b59714c..197a5ecaf9 100644 --- a/include/libs/nodes/plannodes.h +++ b/include/libs/nodes/plannodes.h @@ -185,6 +185,7 @@ typedef struct SMergeLogicNode { int32_t numOfChannels; int32_t srcGroupId; bool groupSort; + bool ignoreGroupId; } SMergeLogicNode; typedef enum EWindowType { @@ -444,6 +445,7 @@ typedef struct SMergePhysiNode { int32_t numOfChannels; int32_t srcGroupId; bool groupSort; + bool ignoreGroupId; } SMergePhysiNode; typedef struct SWinodwPhysiNode { diff --git a/include/libs/parser/parser.h b/include/libs/parser/parser.h index 558203052f..94fb6824d2 100644 --- a/include/libs/parser/parser.h +++ b/include/libs/parser/parser.h @@ -114,7 +114,7 @@ STableDataCxt* smlInitTableDataCtx(SQuery* query, STableMeta* pTableMeta); int32_t smlBindData(SQuery* handle, bool dataFormat, SArray* tags, SArray* colsSchema, SArray* cols, STableMeta* pTableMeta, char* tableName, const char* sTableName, int32_t sTableNameLen, int32_t ttl, - char* msgBuf, int16_t msgBufLen); + char* msgBuf, int32_t msgBufLen); int32_t smlBuildOutput(SQuery* handle, SHashObj* pVgHash); int rawBlockBindData(SQuery *query, STableMeta* pTableMeta, void* data, SVCreateTbReq* pCreateTb, TAOS_FIELD *fields, int numFields, bool needChangeLength); diff --git a/include/libs/sync/sync.h b/include/libs/sync/sync.h index fdb4506cf3..e86a4f9690 100644 --- a/include/libs/sync/sync.h +++ b/include/libs/sync/sync.h @@ -55,6 +55,8 @@ extern "C" { #define SYNC_INDEX_INVALID -1 #define SYNC_TERM_INVALID -1 +#define SYNC_LEARNER_CATCHUP 10 + typedef enum { SYNC_STRATEGY_NO_SNAPSHOT = 0, SYNC_STRATEGY_STANDARD_SNAPSHOT = 1, @@ -76,19 +78,29 @@ typedef enum { TAOS_SYNC_STATE_CANDIDATE = 101, TAOS_SYNC_STATE_LEADER = 102, TAOS_SYNC_STATE_ERROR = 103, + TAOS_SYNC_STATE_LEARNER = 104, } ESyncState; +typedef enum { + TAOS_SYNC_ROLE_VOTER = 0, + TAOS_SYNC_ROLE_LEARNER = 1, + TAOS_SYNC_ROLE_ERROR = 2, +} ESyncRole; + typedef struct SNodeInfo { - int64_t clusterId; - int32_t nodeId; - uint16_t nodePort; - char nodeFqdn[TSDB_FQDN_LEN]; + int64_t clusterId; + int32_t nodeId; + uint16_t nodePort; + char nodeFqdn[TSDB_FQDN_LEN]; + ESyncRole nodeRole; } SNodeInfo; typedef struct SSyncCfg { + int32_t totalReplicaNum; int32_t replicaNum; int32_t myIndex; - SNodeInfo nodeInfo[TSDB_MAX_REPLICA]; + SNodeInfo nodeInfo[TSDB_MAX_REPLICA + TSDB_MAX_LEARNER_REPLICA]; + SyncIndex lastIndex; } SSyncCfg; typedef struct SFsmCbMeta { @@ -155,6 +167,7 @@ typedef struct SSyncFSM { void (*FpBecomeLeaderCb)(const struct SSyncFSM* pFsm); void (*FpBecomeFollowerCb)(const struct SSyncFSM* pFsm); + void (*FpBecomeLearnerCb)(const struct SSyncFSM* pFsm); int32_t (*FpGetSnapshot)(const struct SSyncFSM* pFsm, SSnapshot* pSnapshot, void* pReaderParam, void** ppReader); void (*FpGetSnapshotInfo)(const struct SSyncFSM* pFsm, SSnapshot* pSnapshot); @@ -236,6 +249,8 @@ void syncStop(int64_t rid); void syncPreStop(int64_t rid); void syncPostStop(int64_t rid); int32_t syncPropose(int64_t rid, SRpcMsg* pMsg, bool isWeak, int64_t* seq); +int32_t syncIsCatchUp(int64_t rid); +ESyncRole syncGetRole(int64_t rid); int32_t syncProcessMsg(int64_t rid, SRpcMsg* pMsg); int32_t syncReconfig(int64_t rid, SSyncCfg* pCfg); int32_t syncBeginSnapshot(int64_t rid, int64_t lastApplyIndex); diff --git a/include/libs/wal/wal.h b/include/libs/wal/wal.h index 45050e5d64..f203d5ef77 100644 --- a/include/libs/wal/wal.h +++ b/include/libs/wal/wal.h @@ -127,12 +127,12 @@ typedef struct SWal { typedef struct { int64_t refId; int64_t refVer; -// int64_t refFile; - SWal *pWal; + // int64_t refFile; + SWal *pWal; } SWalRef; typedef struct { -// int8_t scanUncommited; + // int8_t scanUncommited; int8_t scanNotApplied; int8_t scanMeta; int8_t enableRef; @@ -190,15 +190,16 @@ int32_t walApplyVer(SWal *, int64_t ver); // int32_t walDataCorrupted(SWal*); -// read +// wal reader SWalReader *walOpenReader(SWal *, SWalFilterCond *pCond); void walCloseReader(SWalReader *pRead); void walReadReset(SWalReader *pReader); int32_t walReadVer(SWalReader *pRead, int64_t ver); int32_t walReaderSeekVer(SWalReader *pRead, int64_t ver); int32_t walNextValidMsg(SWalReader *pRead); -int64_t walReaderGetCurrentVer(const SWalReader* pReader); -int64_t walReaderGetValidFirstVer(const SWalReader* pReader); +int64_t walReaderGetCurrentVer(const SWalReader *pReader); +int64_t walReaderGetValidFirstVer(const SWalReader *pReader); +void walReaderValidVersionRange(SWalReader *pReader, int64_t *sver, int64_t *ever); // only for tq usage void walSetReaderCapacity(SWalReader *pRead, int32_t capacity); diff --git a/include/util/taoserror.h b/include/util/taoserror.h index 4688c8bb87..acc6977e2e 100644 --- a/include/util/taoserror.h +++ b/include/util/taoserror.h @@ -103,7 +103,7 @@ int32_t* taosGetErrno(); #define TSDB_CODE_CHECKSUM_ERROR TAOS_DEF_ERROR_CODE(0, 0x011F) // internal #define TSDB_CODE_COMPRESS_ERROR TAOS_DEF_ERROR_CODE(0, 0x0120) -#define TSDB_CODE_MSG_NOT_PROCESSED TAOS_DEF_ERROR_CODE(0, 0x0121) // +#define TSDB_CODE_MSG_NOT_PROCESSED TAOS_DEF_ERROR_CODE(0, 0x0121) #define TSDB_CODE_CFG_NOT_FOUND TAOS_DEF_ERROR_CODE(0, 0x0122) #define TSDB_CODE_REPEAT_INIT TAOS_DEF_ERROR_CODE(0, 0x0123) #define TSDB_CODE_DUP_KEY TAOS_DEF_ERROR_CODE(0, 0x0124) @@ -118,9 +118,10 @@ int32_t* taosGetErrno(); #define TSDB_CODE_MSG_ENCODE_ERROR TAOS_DEF_ERROR_CODE(0, 0x012D) #define TSDB_CODE_NO_ENOUGH_DISKSPACE TAOS_DEF_ERROR_CODE(0, 0x012E) -#define TSDB_CODE_APP_IS_STARTING TAOS_DEF_ERROR_CODE(0, 0x0130) // -#define TSDB_CODE_APP_IS_STOPPING TAOS_DEF_ERROR_CODE(0, 0x0131) // -#define TSDB_CODE_IVLD_DATA_FMT TAOS_DEF_ERROR_CODE(0, 0x0132) // +#define TSDB_CODE_APP_IS_STARTING TAOS_DEF_ERROR_CODE(0, 0x0130) +#define TSDB_CODE_APP_IS_STOPPING TAOS_DEF_ERROR_CODE(0, 0x0131) +#define TSDB_CODE_INVALID_DATA_FMT TAOS_DEF_ERROR_CODE(0, 0x0132) +#define TSDB_CODE_INVALID_CFG_VALUE TAOS_DEF_ERROR_CODE(0, 0x0133) //client #define TSDB_CODE_TSC_INVALID_OPERATION TAOS_DEF_ERROR_CODE(0, 0x0200) @@ -404,6 +405,8 @@ int32_t* taosGetErrno(); #define TSDB_CODE_SNODE_ALREADY_DEPLOYED TAOS_DEF_ERROR_CODE(0, 0x040F) #define TSDB_CODE_SNODE_NOT_FOUND TAOS_DEF_ERROR_CODE(0, 0x0410) #define TSDB_CODE_SNODE_NOT_DEPLOYED TAOS_DEF_ERROR_CODE(0, 0x0411) +#define TSDB_CODE_MNODE_NOT_CATCH_UP TAOS_DEF_ERROR_CODE(0, 0x0412) // internal +#define TSDB_CODE_MNODE_ALREADY_IS_VOTER TAOS_DEF_ERROR_CODE(0, 0x0413) // internal // vnode // #define TSDB_CODE_VND_ACTION_IN_PROGRESS TAOS_DEF_ERROR_CODE(0, 0x0500) // 2.x @@ -438,6 +441,8 @@ int32_t* taosGetErrno(); #define TSDB_CODE_VND_STOPPED TAOS_DEF_ERROR_CODE(0, 0x0529) #define TSDB_CODE_VND_DUP_REQUEST TAOS_DEF_ERROR_CODE(0, 0x0530) #define TSDB_CODE_VND_QUERY_BUSY TAOS_DEF_ERROR_CODE(0, 0x0531) +#define TSDB_CODE_VND_NOT_CATCH_UP TAOS_DEF_ERROR_CODE(0, 0x0532) // internal +#define TSDB_CODE_VND_ALREADY_IS_VOTER TAOS_DEF_ERROR_CODE(0, 0x0533) // internal // tsdb #define TSDB_CODE_TDB_INVALID_TABLE_ID TAOS_DEF_ERROR_CODE(0, 0x0600) @@ -535,20 +540,20 @@ int32_t* taosGetErrno(); // #define TSDB_CODE_SYN_INVALID_CHECKSUM TAOS_DEF_ERROR_CODE(0, 0x0908) // 2.x // #define TSDB_CODE_SYN_INVALID_MSGLEN TAOS_DEF_ERROR_CODE(0, 0x0909) // 2.x // #define TSDB_CODE_SYN_INVALID_MSGTYPE TAOS_DEF_ERROR_CODE(0, 0x090A) // 2.x -#define TSDB_CODE_SYN_IS_LEADER TAOS_DEF_ERROR_CODE(0, 0x090B) +// #define TSDB_CODE_SYN_IS_LEADER TAOS_DEF_ERROR_CODE(0, 0x090B) // unused #define TSDB_CODE_SYN_NOT_LEADER TAOS_DEF_ERROR_CODE(0, 0x090C) -#define TSDB_CODE_SYN_ONE_REPLICA TAOS_DEF_ERROR_CODE(0, 0x090D) -#define TSDB_CODE_SYN_NOT_IN_NEW_CONFIG TAOS_DEF_ERROR_CODE(0, 0x090E) -#define TSDB_CODE_SYN_NEW_CONFIG_ERROR TAOS_DEF_ERROR_CODE(0, 0x090F) // internal -#define TSDB_CODE_SYN_RECONFIG_NOT_READY TAOS_DEF_ERROR_CODE(0, 0x0910) +// #define TSDB_CODE_SYN_ONE_REPLICA TAOS_DEF_ERROR_CODE(0, 0x090D) // unused +// #define TSDB_CODE_SYN_NOT_IN_NEW_CONFIG TAOS_DEF_ERROR_CODE(0, 0x090E) // unused +#define TSDB_CODE_SYN_NEW_CONFIG_ERROR TAOS_DEF_ERROR_CODE(0, 0x090F) // internal +// #define TSDB_CODE_SYN_RECONFIG_NOT_READY TAOS_DEF_ERROR_CODE(0, 0x0910) // unused #define TSDB_CODE_SYN_PROPOSE_NOT_READY TAOS_DEF_ERROR_CODE(0, 0x0911) -#define TSDB_CODE_SYN_STANDBY_NOT_READY TAOS_DEF_ERROR_CODE(0, 0x0912) -#define TSDB_CODE_SYN_BATCH_ERROR TAOS_DEF_ERROR_CODE(0, 0x0913) +// #define TSDB_CODE_SYN_STANDBY_NOT_READY TAOS_DEF_ERROR_CODE(0, 0x0912) // unused +// #define TSDB_CODE_SYN_BATCH_ERROR TAOS_DEF_ERROR_CODE(0, 0x0913) // unused #define TSDB_CODE_SYN_RESTORING TAOS_DEF_ERROR_CODE(0, 0x0914) -#define TSDB_CODE_SYN_INVALID_SNAPSHOT_MSG TAOS_DEF_ERROR_CODE(0, 0x0915) // internal +#define TSDB_CODE_SYN_INVALID_SNAPSHOT_MSG TAOS_DEF_ERROR_CODE(0, 0x0915) // internal #define TSDB_CODE_SYN_BUFFER_FULL TAOS_DEF_ERROR_CODE(0, 0x0916) #define TSDB_CODE_SYN_WRITE_STALL TAOS_DEF_ERROR_CODE(0, 0x0917) -#define TSDB_CODE_SYN_NEGO_WIN_EXCEEDED TAOS_DEF_ERROR_CODE(0, 0X0918) +#define TSDB_CODE_SYN_NEGOTIATION_WIN_FULL TAOS_DEF_ERROR_CODE(0, 0x0918) #define TSDB_CODE_SYN_INTERNAL_ERROR TAOS_DEF_ERROR_CODE(0, 0x09FF) // tq @@ -569,7 +574,7 @@ int32_t* taosGetErrno(); // wal // #define TSDB_CODE_WAL_APP_ERROR TAOS_DEF_ERROR_CODE(0, 0x1000) // 2.x #define TSDB_CODE_WAL_FILE_CORRUPTED TAOS_DEF_ERROR_CODE(0, 0x1001) -#define TSDB_CODE_WAL_SIZE_LIMIT TAOS_DEF_ERROR_CODE(0, 0x1002) +// #define TSDB_CODE_WAL_SIZE_LIMIT TAOS_DEF_ERROR_CODE(0, 0x1002) // unused #define TSDB_CODE_WAL_INVALID_VER TAOS_DEF_ERROR_CODE(0, 0x1003) // #define TSDB_CODE_WAL_OUT_OF_MEMORY TAOS_DEF_ERROR_CODE(0, 0x1004) // 2.x #define TSDB_CODE_WAL_LOG_NOT_EXIST TAOS_DEF_ERROR_CODE(0, 0x1005) @@ -713,15 +718,13 @@ int32_t* taosGetErrno(); #define TSDB_CODE_UDF_STOPPING TAOS_DEF_ERROR_CODE(0, 0x2901) #define TSDB_CODE_UDF_PIPE_READ_ERR TAOS_DEF_ERROR_CODE(0, 0x2902) #define TSDB_CODE_UDF_PIPE_CONNECT_ERR TAOS_DEF_ERROR_CODE(0, 0x2903) -#define TSDB_CODE_UDF_PIPE_NO_PIPE TAOS_DEF_ERROR_CODE(0, 0x2904) +#define TSDB_CODE_UDF_PIPE_NOT_EXIST TAOS_DEF_ERROR_CODE(0, 0x2904) #define TSDB_CODE_UDF_LOAD_UDF_FAILURE TAOS_DEF_ERROR_CODE(0, 0x2905) -#define TSDB_CODE_UDF_INVALID_STATE TAOS_DEF_ERROR_CODE(0, 0x2906) -#define TSDB_CODE_UDF_INVALID_INPUT TAOS_DEF_ERROR_CODE(0, 0x2907) -#define TSDB_CODE_UDF_NO_FUNC_HANDLE TAOS_DEF_ERROR_CODE(0, 0x2908) -#define TSDB_CODE_UDF_INVALID_BUFSIZE TAOS_DEF_ERROR_CODE(0, 0x2909) -#define TSDB_CODE_UDF_INVALID_OUTPUT_TYPE TAOS_DEF_ERROR_CODE(0, 0x290A) -#define TSDB_CODE_UDF_SCRIPT_NOT_SUPPORTED TAOS_DEF_ERROR_CODE(0, 0x290B) -#define TSDB_CODE_UDF_FUNC_EXEC_FAILURE TAOS_DEF_ERROR_CODE(0, 0x290C) +#define TSDB_CODE_UDF_INVALID_INPUT TAOS_DEF_ERROR_CODE(0, 0x2906) +#define TSDB_CODE_UDF_INVALID_BUFSIZE TAOS_DEF_ERROR_CODE(0, 0x2907) +#define TSDB_CODE_UDF_INVALID_OUTPUT_TYPE TAOS_DEF_ERROR_CODE(0, 0x2908) +#define TSDB_CODE_UDF_SCRIPT_NOT_SUPPORTED TAOS_DEF_ERROR_CODE(0, 0x2909) +#define TSDB_CODE_UDF_FUNC_EXEC_FAILURE TAOS_DEF_ERROR_CODE(0, 0x290A) // sml #define TSDB_CODE_SML_INVALID_PROTOCOL_TYPE TAOS_DEF_ERROR_CODE(0, 0x3000) diff --git a/include/util/tdef.h b/include/util/tdef.h index a77663fad6..2d1290713a 100644 --- a/include/util/tdef.h +++ b/include/util/tdef.h @@ -232,13 +232,7 @@ typedef enum ELogicConditionType { #define TSDB_QUERY_ID_LEN 26 #define TSDB_TRANS_OPER_LEN 16 -/** - * In some scenarios uint16_t (0~65535) is used to store the row len. - * - Firstly, we use 65531(65535 - 4), as the SDataRow/SKVRow contains 4 bits header. - * - Secondly, if all cols are VarDataT type except primary key, we need 4 bits to store the offset, thus - * the final value is 65531-(4096-1)*4 = 49151. - */ -#define TSDB_MAX_BYTES_PER_ROW 49151 +#define TSDB_MAX_BYTES_PER_ROW 65531 // 49151:65531 #define TSDB_MAX_TAGS_LEN 16384 #define TSDB_MAX_TAGS 128 @@ -285,7 +279,7 @@ typedef enum ELogicConditionType { #define TSDB_DNODE_ROLE_VNODE 2 #define TSDB_MAX_REPLICA 5 - +#define TSDB_MAX_LEARNER_REPLICA 10 #define TSDB_SYNC_LOG_BUFFER_SIZE 4096 #define TSDB_SYNC_LOG_BUFFER_RETENTION 256 #define TSDB_SYNC_APPLYQ_SIZE_LIMIT 512 @@ -410,9 +404,9 @@ typedef enum ELogicConditionType { #define TSDB_EXPLAIN_RESULT_ROW_SIZE (16 * 1024) #define TSDB_EXPLAIN_RESULT_COLUMN_NAME "QUERY_PLAN" -#define TSDB_MAX_FIELD_LEN 16384 -#define TSDB_MAX_BINARY_LEN (TSDB_MAX_FIELD_LEN - TSDB_KEYSIZE) // keep 16384 -#define TSDB_MAX_NCHAR_LEN (TSDB_MAX_FIELD_LEN - TSDB_KEYSIZE) // keep 16384 +#define TSDB_MAX_FIELD_LEN 65519 // 16384:65519 +#define TSDB_MAX_BINARY_LEN TSDB_MAX_FIELD_LEN // 16384-8:65519 +#define TSDB_MAX_NCHAR_LEN TSDB_MAX_FIELD_LEN // 16384-8:65519 #define PRIMARYKEY_TIMESTAMP_COL_ID 1 #define COL_REACH_END(colId, maxColId) ((colId) > (maxColId)) diff --git a/include/util/tlog.h b/include/util/tlog.h index 541b7589b7..5a421033c9 100644 --- a/include/util/tlog.h +++ b/include/util/tlog.h @@ -83,6 +83,12 @@ void taosPrintLongString(const char *flags, ELogLevel level, int32_t dflag, cons #endif ; +void taosPrintSlowLog(const char *format, ...) +#ifdef __GNUC__ + __attribute__((format(printf, 1, 2))) +#endif + ; + bool taosAssertDebug(bool condition, const char *file, int32_t line, const char *format, ...); bool taosAssertRelease(bool condition); diff --git a/source/client/inc/clientInt.h b/source/client/inc/clientInt.h index 46d44d7443..18891bb932 100644 --- a/source/client/inc/clientInt.h +++ b/source/client/inc/clientInt.h @@ -63,6 +63,7 @@ typedef struct { // statistics int32_t reportCnt; int32_t connKeyCnt; + int32_t passKeyCnt; // with passVer call back int64_t reportBytes; // not implemented int64_t startTime; // ctl @@ -126,6 +127,12 @@ typedef struct SAppInfo { TdThreadMutex mutex; } SAppInfo; +typedef struct { + int32_t ver; + void* param; + __taos_notify_fn_t fp; +} SPassInfo; + typedef struct STscObj { char user[TSDB_USER_LEN]; char pass[TSDB_PASSWORD_LEN]; @@ -141,6 +148,7 @@ typedef struct STscObj { int32_t numOfReqs; // number of sqlObj bound to this connection SAppInstInfo* pAppInfo; SHashObj* pRequests; + SPassInfo passInfo; } STscObj; typedef struct STscDbg { @@ -354,7 +362,7 @@ void stopAllRequests(SHashObj* pRequests); // conn level int hbRegisterConn(SAppHbMgr* pAppHbMgr, int64_t tscRefId, int64_t clusterId, int8_t connType); -void hbDeregisterConn(SAppHbMgr* pAppHbMgr, SClientHbKey connKey); +void hbDeregisterConn(STscObj* pTscObj, SClientHbKey connKey); typedef struct SSqlCallbackWrapper { SParseContext* pParseCtx; diff --git a/source/client/src/clientEnv.c b/source/client/src/clientEnv.c index 418103f2a6..c8f3feb2d4 100644 --- a/source/client/src/clientEnv.c +++ b/source/client/src/clientEnv.c @@ -42,7 +42,7 @@ SAppInfo appInfo; int64_t lastClusterId = 0; int32_t clientReqRefPool = -1; int32_t clientConnRefPool = -1; -int32_t clientStop = 0; +int32_t clientStop = -1; int32_t timestampDeltaLimit = 900; // s @@ -69,7 +69,6 @@ static int32_t registerRequest(SRequestObj *pRequest, STscObj *pTscObj) { } static void deregisterRequest(SRequestObj *pRequest) { - const static int64_t SLOW_QUERY_INTERVAL = 3000000L; // todo configurable if (pRequest == NULL) { tscError("pRequest == NULL"); return; @@ -80,6 +79,7 @@ static void deregisterRequest(SRequestObj *pRequest) { int32_t currentInst = atomic_sub_fetch_64((int64_t *)&pActivity->currentRequests, 1); int32_t num = atomic_sub_fetch_32(&pTscObj->numOfReqs, 1); + int32_t reqType = SLOW_LOG_TYPE_OTHERS; int64_t duration = taosGetTimestampUs() - pRequest->metric.start; tscDebug("0x%" PRIx64 " free Request from connObj: 0x%" PRIx64 ", reqId:0x%" PRIx64 @@ -95,6 +95,7 @@ static void deregisterRequest(SRequestObj *pRequest) { duration, pRequest->metric.parseCostUs, pRequest->metric.ctgCostUs, pRequest->metric.analyseCostUs, pRequest->metric.planCostUs, pRequest->metric.execCostUs); atomic_add_fetch_64((int64_t *)&pActivity->insertElapsedTime, duration); + reqType = SLOW_LOG_TYPE_INSERT; } else if (QUERY_NODE_SELECT_STMT == pRequest->stmtType) { tscDebug("query duration %" PRId64 "us: parseCost:%" PRId64 "us, ctgCost:%" PRId64 "us, analyseCost:%" PRId64 "us, planCost:%" PRId64 "us, exec:%" PRId64 "us", @@ -102,12 +103,16 @@ static void deregisterRequest(SRequestObj *pRequest) { pRequest->metric.planCostUs, pRequest->metric.execCostUs); atomic_add_fetch_64((int64_t *)&pActivity->queryElapsedTime, duration); + reqType = SLOW_LOG_TYPE_QUERY; } } - if (duration >= SLOW_QUERY_INTERVAL) { + if (duration >= (tsSlowLogThreshold * 1000000UL)) { atomic_add_fetch_64((int64_t *)&pActivity->numOfSlowQueries, 1); - tscWarnL("slow query: %s, duration:%" PRId64, pRequest->sqlstr, duration); + if (tsSlowLogScope & reqType) { + taosPrintSlowLog("PID:%d, Conn:%u, QID:0x%" PRIx64 ", Start:%" PRId64 ", Duration:%" PRId64 "us, SQL:%s", + taosGetPId(), pTscObj->connId, pRequest->requestId, pRequest->metric.start, duration, pRequest->sqlstr); + } } releaseTscObj(pTscObj->id); @@ -239,7 +244,7 @@ void destroyTscObj(void *pObj) { tscTrace("begin to destroy tscObj %" PRIx64 " p:%p", tscId, pTscObj); SClientHbKey connKey = {.tscRid = pTscObj->id, .connType = pTscObj->connType}; - hbDeregisterConn(pTscObj->pAppInfo->pAppHbMgr, connKey); + hbDeregisterConn(pTscObj, connKey); destroyAllRequests(pTscObj->pRequests); taosHashCleanup(pTscObj->pRequests); @@ -427,8 +432,12 @@ static void *tscCrashReportThreadFp(void *param) { } #endif + if (-1 != atomic_val_compare_exchange_32(&clientStop, -1, 0)) { + return NULL; + } + while (1) { - if (clientStop) break; + if (clientStop > 0) break; if (loopTimes++ < reportPeriodNum) { taosMsleep(sleepTime); continue; @@ -466,7 +475,7 @@ static void *tscCrashReportThreadFp(void *param) { loopTimes = 0; } - clientStop = -1; + clientStop = -2; return NULL; } diff --git a/source/client/src/clientHb.c b/source/client/src/clientHb.c index 8d082ab60b..78092b65a8 100644 --- a/source/client/src/clientHb.c +++ b/source/client/src/clientHb.c @@ -19,6 +19,15 @@ #include "scheduler.h" #include "trpc.h" +typedef struct { + union { + struct { + int64_t clusterId; + int32_t passKeyCnt; + }; + }; +} SHbParam; + static SClientHbMgr clientHbMgr = {0}; static int32_t hbCreateThread(); @@ -49,14 +58,60 @@ static int32_t hbProcessUserAuthInfoRsp(void *value, int32_t valueLen, struct SC return TSDB_CODE_SUCCESS; } +static int32_t hbProcessUserPassInfoRsp(void *value, int32_t valueLen, SClientHbKey *connKey, SAppHbMgr *pAppHbMgr) { + int32_t code = 0; + int32_t numOfBatchs = 0; + SUserPassBatchRsp batchRsp = {0}; + if (tDeserializeSUserPassBatchRsp(value, valueLen, &batchRsp) != 0) { + code = TSDB_CODE_INVALID_MSG; + return code; + } + + numOfBatchs = taosArrayGetSize(batchRsp.pArray); + + SClientHbReq *pReq = NULL; + while ((pReq = taosHashIterate(pAppHbMgr->activeInfo, pReq))) { + STscObj *pTscObj = (STscObj *)acquireTscObj(pReq->connKey.tscRid); + if (!pTscObj) { + continue; + } + SPassInfo *passInfo = &pTscObj->passInfo; + if (!passInfo->fp) { + releaseTscObj(pReq->connKey.tscRid); + continue; + } + + for (int32_t i = 0; i < numOfBatchs; ++i) { + SGetUserPassRsp *rsp = taosArrayGet(batchRsp.pArray, i); + if (0 == strncmp(rsp->user, pTscObj->user, TSDB_USER_LEN)) { + int32_t oldVer = atomic_load_32(&passInfo->ver); + if (oldVer < rsp->version) { + atomic_store_32(&passInfo->ver, rsp->version); + if (passInfo->fp) { + (*passInfo->fp)(passInfo->param, &passInfo->ver, TAOS_NOTIFY_PASSVER); + } + tscDebug("update passVer of user %s from %d to %d, tscRid:%" PRIi64, rsp->user, oldVer, + atomic_load_32(&passInfo->ver), pTscObj->id); + } + break; + } + } + releaseTscObj(pReq->connKey.tscRid); + } + + taosArrayDestroy(batchRsp.pArray); + + return code; +} + static int32_t hbGenerateVgInfoFromRsp(SDBVgInfo **pInfo, SUseDbRsp *rsp) { - int32_t code = 0; + int32_t code = 0; SDBVgInfo *vgInfo = taosMemoryCalloc(1, sizeof(SDBVgInfo)); if (NULL == vgInfo) { code = TSDB_CODE_OUT_OF_MEMORY; return code; } - + vgInfo->vgVersion = rsp->vgVersion; vgInfo->stateTs = rsp->stateTs; vgInfo->hashMethod = rsp->hashMethod; @@ -69,7 +124,7 @@ static int32_t hbGenerateVgInfoFromRsp(SDBVgInfo **pInfo, SUseDbRsp *rsp) { code = TSDB_CODE_OUT_OF_MEMORY; goto _return; } - + for (int32_t j = 0; j < rsp->vgNum; ++j) { SVgroupInfo *pInfo = taosArrayGet(rsp->pVgroupInfos, j); if (taosHashPut(vgInfo->vgHash, &pInfo->vgId, sizeof(int32_t), pInfo, sizeof(SVgroupInfo)) != 0) { @@ -123,7 +178,8 @@ static int32_t hbProcessDBInfoRsp(void *value, int32_t valueLen, struct SCatalog goto _return; } - catalogUpdateDBVgInfo(pCatalog, (rsp->db[0] == 'i') ? TSDB_PERFORMANCE_SCHEMA_DB : TSDB_INFORMATION_SCHEMA_DB, rsp->uid, vgInfo); + catalogUpdateDBVgInfo(pCatalog, (rsp->db[0] == 'i') ? TSDB_PERFORMANCE_SCHEMA_DB : TSDB_INFORMATION_SCHEMA_DB, + rsp->uid, vgInfo); } } @@ -291,6 +347,15 @@ static int32_t hbQueryHbRspHandle(SAppHbMgr *pAppHbMgr, SClientHbRsp *pRsp) { hbProcessStbInfoRsp(kv->value, kv->valueLen, pCatalog); break; } + case HEARTBEAT_KEY_USER_PASSINFO: { + if (kv->valueLen <= 0 || NULL == kv->value) { + tscError("invalid hb user pass info, len:%d, value:%p", kv->valueLen, kv->value); + break; + } + + hbProcessUserPassInfoRsp(kv->value, kv->valueLen, &pRsp->connKey, pAppHbMgr); + break; + } default: tscError("invalid hb key type:%d", kv->key); break; @@ -308,7 +373,7 @@ static int32_t hbAsyncCallBack(void *param, SDataBuf *pMsg, int32_t code) { } static int32_t emptyRspNum = 0; - int32_t idx = *(int32_t *)param; + int32_t idx = *(int32_t *)param; SClientHbBatchRsp pRsp = {0}; if (TSDB_CODE_SUCCESS == code) { tDeserializeSClientHbBatchRsp(pMsg->pData, pMsg->len, &pRsp); @@ -339,8 +404,7 @@ static int32_t hbAsyncCallBack(void *param, SDataBuf *pMsg, int32_t code) { if (code != 0) { pInst->onlineDnodes = pInst->totalDnodes ? 0 : -1; - tscDebug("hb rsp error %s, update server status %d/%d", tstrerror(code), pInst->onlineDnodes, - pInst->totalDnodes); + tscDebug("hb rsp error %s, update server status %d/%d", tstrerror(code), pInst->onlineDnodes, pInst->totalDnodes); } if (rspNum) { @@ -472,6 +536,49 @@ int32_t hbGetQueryBasicInfo(SClientHbKey *connKey, SClientHbReq *req) { return TSDB_CODE_SUCCESS; } +static int32_t hbGetUserBasicInfo(SClientHbKey *connKey, SClientHbReq *req) { + STscObj *pTscObj = (STscObj *)acquireTscObj(connKey->tscRid); + if (!pTscObj) { + tscWarn("tscObj rid %" PRIx64 " not exist", connKey->tscRid); + return TSDB_CODE_APP_ERROR; + } + + int32_t code = 0; + SUserPassVersion *user = taosMemoryMalloc(sizeof(SUserPassVersion)); + if (!user) { + code = TSDB_CODE_OUT_OF_MEMORY; + goto _return; + } + strncpy(user->user, pTscObj->user, TSDB_USER_LEN); + user->version = htonl(pTscObj->passInfo.ver); + + SKv kv = { + .key = HEARTBEAT_KEY_USER_PASSINFO, + .valueLen = sizeof(SUserPassVersion), + .value = user, + }; + + tscDebug("hb got user basic info, valueLen:%d, user:%s, passVer:%d, tscRid:%" PRIi64, kv.valueLen, user->user, + pTscObj->passInfo.ver, connKey->tscRid); + + if (!req->info) { + req->info = taosHashInit(64, hbKeyHashFunc, 1, HASH_ENTRY_LOCK); + } + + if (taosHashPut(req->info, &kv.key, sizeof(kv.key), &kv, sizeof(kv)) < 0) { + code = terrno ? terrno : TSDB_CODE_APP_ERROR; + goto _return; + } + +_return: + releaseTscObj(connKey->tscRid); + if (code) { + tscError("hb got user basic info failed since %s", terrstr(code)); + } + + return code; +} + int32_t hbGetExpiredUserInfo(SClientHbKey *connKey, struct SCatalog *pCatalog, SClientHbReq *req) { SUserAuthVersion *users = NULL; uint32_t userNum = 0; @@ -526,8 +633,8 @@ int32_t hbGetExpiredDBInfo(SClientHbKey *connKey, struct SCatalog *pCatalog, SCl for (int32_t i = 0; i < dbNum; ++i) { SDbVgVersion *db = &dbs[i]; - tscDebug("the %dth expired dbFName:%s, dbId:%" PRId64 ", vgVersion:%d, numOfTable:%d, startTs:%" PRId64, - i, db->dbFName, db->dbId, db->vgVersion, db->numOfTable, db->stateTs); + tscDebug("the %dth expired dbFName:%s, dbId:%" PRId64 ", vgVersion:%d, numOfTable:%d, startTs:%" PRId64, i, + db->dbFName, db->dbId, db->vgVersion, db->numOfTable, db->stateTs); db->dbId = htobe64(db->dbId); db->vgVersion = htonl(db->vgVersion); @@ -607,19 +714,23 @@ int32_t hbGetAppInfo(int64_t clusterId, SClientHbReq *req) { } int32_t hbQueryHbReqHandle(SClientHbKey *connKey, void *param, SClientHbReq *req) { - int64_t *clusterId = (int64_t *)param; + SHbParam *hbParam = (SHbParam *)param; struct SCatalog *pCatalog = NULL; - int32_t code = catalogGetHandle(*clusterId, &pCatalog); + int32_t code = catalogGetHandle(hbParam->clusterId, &pCatalog); if (code != TSDB_CODE_SUCCESS) { - tscWarn("catalogGetHandle failed, clusterId:%" PRIx64 ", error:%s", *clusterId, tstrerror(code)); + tscWarn("catalogGetHandle failed, clusterId:%" PRIx64 ", error:%s", hbParam->clusterId, tstrerror(code)); return code; } - hbGetAppInfo(*clusterId, req); + hbGetAppInfo(hbParam->clusterId, req); hbGetQueryBasicInfo(connKey, req); + if (hbParam->passKeyCnt > 0) { + hbGetUserBasicInfo(connKey, req); + } + code = hbGetExpiredUserInfo(connKey, pCatalog, req); if (TSDB_CODE_SUCCESS != code) { return code; @@ -673,7 +784,26 @@ SClientHbBatchReq *hbGatherAllInfo(SAppHbMgr *pAppHbMgr) { while (pIter != NULL) { pOneReq = taosArrayPush(pBatchReq->reqs, pOneReq); - code = (*clientHbMgr.reqHandle[pOneReq->connKey.connType])(&pOneReq->connKey, &pOneReq->clusterId, pOneReq); + SHbParam param; + switch (pOneReq->connKey.connType) { + case CONN_TYPE__QUERY: { + param.clusterId = pOneReq->clusterId; + param.passKeyCnt = atomic_load_32(&pAppHbMgr->passKeyCnt); + break; + } + default: + break; + } + if (clientHbMgr.reqHandle[pOneReq->connKey.connType]) { + code = (*clientHbMgr.reqHandle[pOneReq->connKey.connType])(&pOneReq->connKey, ¶m, pOneReq); + if (code) { + tscWarn("hbGatherAllInfo failed since %s, tscRid:%" PRIi64 ", connType:%" PRIi8, tstrerror(code), + pOneReq->connKey.tscRid, pOneReq->connKey.connType); + } + } + break; + +#if 0 if (code) { pIter = taosHashIterate(pAppHbMgr->activeInfo, pIter); pOneReq = pIter; @@ -682,6 +812,7 @@ SClientHbBatchReq *hbGatherAllInfo(SAppHbMgr *pAppHbMgr) { pIter = taosHashIterate(pAppHbMgr->activeInfo, pIter); pOneReq = pIter; +#endif } releaseTscObj(rid); @@ -856,7 +987,7 @@ static void hbStopThread() { } SAppHbMgr *appHbMgrInit(SAppInstInfo *pAppInstInfo, char *key) { - if(hbMgrInit() != 0){ + if (hbMgrInit() != 0) { terrno = TSDB_CODE_TSC_INTERNAL_ERROR; return NULL; } @@ -868,6 +999,7 @@ SAppHbMgr *appHbMgrInit(SAppInstInfo *pAppInstInfo, char *key) { // init stat pAppHbMgr->startTime = taosGetTimestampMs(); pAppHbMgr->connKeyCnt = 0; + pAppHbMgr->passKeyCnt = 0; pAppHbMgr->reportCnt = 0; pAppHbMgr->reportBytes = 0; pAppHbMgr->key = taosStrdup(key); @@ -946,27 +1078,23 @@ int hbMgrInit() { TdThreadMutexAttr attr = {0}; int ret = taosThreadMutexAttrInit(&attr); - if(ret != 0){ - uError("hbMgrInit:taosThreadMutexAttrInit error") - return ret; + if (ret != 0) { + uError("hbMgrInit:taosThreadMutexAttrInit error") return ret; } ret = taosThreadMutexAttrSetType(&attr, PTHREAD_MUTEX_RECURSIVE); - if(ret != 0){ - uError("hbMgrInit:taosThreadMutexAttrSetType error") - return ret; + if (ret != 0) { + uError("hbMgrInit:taosThreadMutexAttrSetType error") return ret; } ret = taosThreadMutexInit(&clientHbMgr.lock, &attr); - if(ret != 0){ - uError("hbMgrInit:taosThreadMutexInit error") - return ret; + if (ret != 0) { + uError("hbMgrInit:taosThreadMutexInit error") return ret; } ret = taosThreadMutexAttrDestroy(&attr); - if(ret != 0){ - uError("hbMgrInit:taosThreadMutexAttrDestroy error") - return ret; + if (ret != 0) { + uError("hbMgrInit:taosThreadMutexAttrDestroy error") return ret; } // init handle funcs @@ -1028,7 +1156,8 @@ int hbRegisterConn(SAppHbMgr *pAppHbMgr, int64_t tscRefId, int64_t clusterId, in } } -void hbDeregisterConn(SAppHbMgr *pAppHbMgr, SClientHbKey connKey) { +void hbDeregisterConn(STscObj *pTscObj, SClientHbKey connKey) { + SAppHbMgr *pAppHbMgr = pTscObj->pAppInfo->pAppHbMgr; SClientHbReq *pReq = taosHashAcquire(pAppHbMgr->activeInfo, &connKey, sizeof(SClientHbKey)); if (pReq) { tFreeClientHbReq(pReq); @@ -1041,9 +1170,12 @@ void hbDeregisterConn(SAppHbMgr *pAppHbMgr, SClientHbKey connKey) { } atomic_sub_fetch_32(&pAppHbMgr->connKeyCnt, 1); + taosThreadMutexLock(&pTscObj->mutex); + if (pTscObj->passInfo.fp) { + atomic_sub_fetch_32(&pAppHbMgr->passKeyCnt, 1); + } + taosThreadMutexUnlock(&pTscObj->mutex); } // set heart beat thread quit mode , if quicByKill 1 then kill thread else quit from inner -void taos_set_hb_quit(int8_t quitByKill) { - clientHbMgr.quitByKill = quitByKill; -} +void taos_set_hb_quit(int8_t quitByKill) { clientHbMgr.quitByKill = quitByKill; } diff --git a/source/client/src/clientImpl.c b/source/client/src/clientImpl.c index ce174744ef..f8eade1d7c 100644 --- a/source/client/src/clientImpl.c +++ b/source/client/src/clientImpl.c @@ -1248,6 +1248,11 @@ STscObj* taosConnectImpl(const char* user, const char* auth, const char* db, __t return NULL; } + pRequest->sqlstr = taosStrdup("taos_connect"); + if (pRequest->sqlstr) { + pRequest->sqlLen = strlen(pRequest->sqlstr); + } + SMsgSendInfo* body = buildConnectMsg(pRequest); int64_t transporterId = 0; @@ -1257,7 +1262,7 @@ STscObj* taosConnectImpl(const char* user, const char* auth, const char* db, __t if (pRequest->code != TSDB_CODE_SUCCESS) { const char* errorMsg = (pRequest->code == TSDB_CODE_RPC_FQDN_ERROR) ? taos_errstr(pRequest) : tstrerror(pRequest->code); - fprintf(stderr, "failed to connect to server, reason: %s\n\n", errorMsg); + tscError("failed to connect to server, reason: %s", errorMsg); terrno = pRequest->code; destroyRequest(pRequest); diff --git a/source/client/src/clientMain.c b/source/client/src/clientMain.c index fd70598efb..55337d4a0b 100644 --- a/source/client/src/clientMain.c +++ b/source/client/src/clientMain.c @@ -119,6 +119,43 @@ TAOS *taos_connect(const char *ip, const char *user, const char *pass, const cha return NULL; } +int taos_set_notify_cb(TAOS *taos, __taos_notify_fn_t fp, void *param, int type) { + if (taos == NULL) { + terrno = TSDB_CODE_INVALID_PARA; + return terrno; + } + + STscObj *pObj = acquireTscObj(*(int64_t *)taos); + if (NULL == pObj) { + terrno = TSDB_CODE_TSC_DISCONNECTED; + tscError("invalid parameter for %s", __func__); + return terrno; + } + + switch (type) { + case TAOS_NOTIFY_PASSVER: { + taosThreadMutexLock(&pObj->mutex); + if (fp && !pObj->passInfo.fp) { + atomic_add_fetch_32(&pObj->pAppInfo->pAppHbMgr->passKeyCnt, 1); + } else if (!fp && pObj->passInfo.fp) { + atomic_sub_fetch_32(&pObj->pAppInfo->pAppHbMgr->passKeyCnt, 1); + } + pObj->passInfo.fp = fp; + pObj->passInfo.param = param; + taosThreadMutexUnlock(&pObj->mutex); + break; + } + default: { + terrno = TSDB_CODE_INVALID_PARA; + releaseTscObj(*(int64_t *)taos); + return terrno; + } + } + + releaseTscObj(*(int64_t *)taos); + return 0; +} + void taos_close_internal(void *taos) { if (taos == NULL) { return; @@ -1307,6 +1344,8 @@ int taos_load_table_info(TAOS *taos, const char *tableNameList) { goto _return; } + pRequest->syncQuery = true; + STscObj *pTscObj = pRequest->pTscObj; code = transferTableNameList(tableNameList, pTscObj->acctId, pTscObj->db, &catalogReq.pTableMeta); if (code) { @@ -1333,7 +1372,7 @@ int taos_load_table_info(TAOS *taos, const char *tableNameList) { tsem_wait(&pParam->sem); _return: - taosArrayDestroy(catalogReq.pTableMeta); + taosArrayDestroyEx(catalogReq.pTableMeta, destoryTablesReq); destroyRequest(pRequest); return code; } diff --git a/source/client/src/clientMsgHandler.c b/source/client/src/clientMsgHandler.c index 032517cafc..6d53f2b4c5 100644 --- a/source/client/src/clientMsgHandler.c +++ b/source/client/src/clientMsgHandler.c @@ -129,6 +129,7 @@ int32_t processConnectRsp(void* param, SDataBuf* pMsg, int32_t code) { lastClusterId = connectRsp.clusterId; pTscObj->connType = connectRsp.connType; + pTscObj->passInfo.ver = connectRsp.passVer; hbRegisterConn(pTscObj->pAppInfo->pAppHbMgr, pTscObj->id, connectRsp.clusterId, connectRsp.connType); diff --git a/source/client/src/clientRawBlockWrite.c b/source/client/src/clientRawBlockWrite.c index a09780dc15..32f28e4563 100644 --- a/source/client/src/clientRawBlockWrite.c +++ b/source/client/src/clientRawBlockWrite.c @@ -1511,7 +1511,7 @@ static int32_t tmqWriteRawDataImpl(TAOS* taos, void* data, int32_t dataLen) { rspObj.resType = RES_TYPE__TMQ; tDecoderInit(&decoder, data, dataLen); - code = tDecodeSMqDataRsp(&decoder, &rspObj.rsp); + code = tDecodeMqDataRsp(&decoder, &rspObj.rsp); if (code != 0) { uError("WriteRaw:decode smqDataRsp error"); code = TSDB_CODE_INVALID_MSG; @@ -1615,7 +1615,7 @@ static int32_t tmqWriteRawDataImpl(TAOS* taos, void* data, int32_t dataLen) { code = pRequest->code; end: - tDeleteSMqDataRsp(&rspObj.rsp); + tDeleteMqDataRsp(&rspObj.rsp); tDecoderClear(&decoder); qDestroyQuery(pQuery); destroyRequest(pRequest); @@ -1858,7 +1858,7 @@ int32_t tmq_get_raw(TAOS_RES* res, tmq_raw_data* raw) { int32_t len = 0; int32_t code = 0; - tEncodeSize(tEncodeSMqDataRsp, &rspObj->rsp, len, code); + tEncodeSize(tEncodeMqDataRsp, &rspObj->rsp, len, code); if (code < 0) { return -1; } @@ -1866,7 +1866,7 @@ int32_t tmq_get_raw(TAOS_RES* res, tmq_raw_data* raw) { void* buf = taosMemoryCalloc(1, len); SEncoder encoder = {0}; tEncoderInit(&encoder, buf, len); - tEncodeSMqDataRsp(&encoder, &rspObj->rsp); + tEncodeMqDataRsp(&encoder, &rspObj->rsp); tEncoderClear(&encoder); raw->raw = buf; diff --git a/source/client/src/clientSml.c b/source/client/src/clientSml.c index c8fb5f7bd8..33fabbed50 100644 --- a/source/client/src/clientSml.c +++ b/source/client/src/clientSml.c @@ -195,17 +195,19 @@ int32_t smlSetCTableName(SSmlTableInfo *oneTable) { return TSDB_CODE_SUCCESS; } -void getTableUid(SSmlHandle *info, SSmlLineInfo *currElement, SSmlTableInfo *tinfo){ - char key[TSDB_TABLE_NAME_LEN * 2 + 1] = {0}; +void getTableUid(SSmlHandle *info, SSmlLineInfo *currElement, SSmlTableInfo *tinfo) { + char key[TSDB_TABLE_NAME_LEN * 2 + 1] = {0}; size_t nLen = strlen(tinfo->childTableName); memcpy(key, currElement->measure, currElement->measureLen); memcpy(key + currElement->measureLen + 1, tinfo->childTableName, nLen); - void *uid = taosHashGet(info->tableUids, key, currElement->measureLen + 1 + nLen); // use \0 as separator for stable name and child table name + void *uid = + taosHashGet(info->tableUids, key, + currElement->measureLen + 1 + nLen); // use \0 as separator for stable name and child table name if (uid == NULL) { tinfo->uid = info->uid++; taosHashPut(info->tableUids, key, currElement->measureLen + 1 + nLen, &tinfo->uid, sizeof(uint64_t)); - }else{ - tinfo->uid = *(uint64_t*)uid; + } else { + tinfo->uid = *(uint64_t *)uid; } } @@ -548,7 +550,8 @@ static int32_t smlGenerateSchemaAction(SSchema *colField, SHashObj *colHash, SSm uint16_t *index = colHash ? (uint16_t *)taosHashGet(colHash, kv->key, kv->keyLen) : NULL; if (index) { if (colField[*index].type != kv->type) { - uError("SML:0x%" PRIx64 " point type and db type mismatch. db type: %d, point type: %d, key: %s", info->id, colField[*index].type, kv->type, kv->key); + uError("SML:0x%" PRIx64 " point type and db type mismatch. db type: %d, point type: %d, key: %s", info->id, + colField[*index].type, kv->type, kv->key); return TSDB_CODE_SML_INVALID_DATA; } @@ -575,17 +578,18 @@ static int32_t smlGenerateSchemaAction(SSchema *colField, SHashObj *colHash, SSm #define BOUNDARY 1024 static int32_t smlFindNearestPowerOf2(int32_t length, uint8_t type) { int32_t result = 1; - if (length >= BOUNDARY){ + if (length >= BOUNDARY) { result = length; - }else{ + } else { while (result <= length) { - result *= 2; + result <<= 1; } } + if (type == TSDB_DATA_TYPE_BINARY && result > TSDB_MAX_BINARY_LEN - VARSTR_HEADER_SIZE) { result = TSDB_MAX_BINARY_LEN - VARSTR_HEADER_SIZE; - } else if (type == TSDB_DATA_TYPE_NCHAR && result > (TSDB_MAX_BINARY_LEN - VARSTR_HEADER_SIZE) / TSDB_NCHAR_SIZE) { - result = (TSDB_MAX_BINARY_LEN - VARSTR_HEADER_SIZE) / TSDB_NCHAR_SIZE; + } else if (type == TSDB_DATA_TYPE_NCHAR && result > (TSDB_MAX_NCHAR_LEN - VARSTR_HEADER_SIZE) / TSDB_NCHAR_SIZE) { + result = (TSDB_MAX_NCHAR_LEN - VARSTR_HEADER_SIZE) / TSDB_NCHAR_SIZE; } if (type == TSDB_DATA_TYPE_NCHAR) { @@ -675,7 +679,7 @@ static int32_t smlBuildFieldsList(SSmlHandle *info, SSchema *schemaField, SHashO SField *field = taosArrayGet(results, j); len += field->bytes; } - if(len > maxLen){ + if (len > maxLen) { return isTag ? TSDB_CODE_PAR_INVALID_TAGS_LENGTH : TSDB_CODE_PAR_INVALID_ROW_LENGTH; } @@ -690,6 +694,7 @@ static int32_t smlSendMetaMsg(SSmlHandle *info, SName *pName, SArray *pColumns, SMCreateStbReq pReq = {0}; int32_t code = TSDB_CODE_SUCCESS; SCmdMsgInfo pCmdMsg = {0}; + char *pSql = NULL; // put front for free pReq.numOfColumns = taosArrayGetSize(pColumns); @@ -697,7 +702,27 @@ static int32_t smlSendMetaMsg(SSmlHandle *info, SName *pName, SArray *pColumns, pReq.numOfTags = taosArrayGetSize(pTags); pReq.pTags = pTags; - code = buildRequest(info->taos->id, "", 0, NULL, false, &pRequest, 0); + if (action == SCHEMA_ACTION_CREATE_STABLE) { + pReq.colVer = 1; + pReq.tagVer = 1; + pReq.suid = 0; + pReq.source = TD_REQ_FROM_APP; + pSql = "sml_create_stable"; + } else if (action == SCHEMA_ACTION_ADD_TAG || action == SCHEMA_ACTION_CHANGE_TAG_SIZE) { + pReq.colVer = pTableMeta->sversion; + pReq.tagVer = pTableMeta->tversion + 1; + pReq.suid = pTableMeta->uid; + pReq.source = TD_REQ_FROM_TAOX; + pSql = (action == SCHEMA_ACTION_ADD_TAG) ? "sml_add_tag" : "sml_modify_tag_size"; + } else if (action == SCHEMA_ACTION_ADD_COLUMN || action == SCHEMA_ACTION_CHANGE_COLUMN_SIZE) { + pReq.colVer = pTableMeta->sversion + 1; + pReq.tagVer = pTableMeta->tversion; + pReq.suid = pTableMeta->uid; + pReq.source = TD_REQ_FROM_TAOX; + pSql = (action == SCHEMA_ACTION_ADD_COLUMN) ? "sml_add_column" : "sml_modify_column_size"; + } + + code = buildRequest(info->taos->id, pSql, strlen(pSql), NULL, false, &pRequest, 0); if (code != TSDB_CODE_SUCCESS) { goto end; } @@ -708,23 +733,6 @@ static int32_t smlSendMetaMsg(SSmlHandle *info, SName *pName, SArray *pColumns, goto end; } - if (action == SCHEMA_ACTION_CREATE_STABLE) { - pReq.colVer = 1; - pReq.tagVer = 1; - pReq.suid = 0; - pReq.source = TD_REQ_FROM_APP; - } else if (action == SCHEMA_ACTION_ADD_TAG || action == SCHEMA_ACTION_CHANGE_TAG_SIZE) { - pReq.colVer = pTableMeta->sversion; - pReq.tagVer = pTableMeta->tversion + 1; - pReq.suid = pTableMeta->uid; - pReq.source = TD_REQ_FROM_TAOX; - } else if (action == SCHEMA_ACTION_ADD_COLUMN || action == SCHEMA_ACTION_CHANGE_COLUMN_SIZE) { - pReq.colVer = pTableMeta->sversion + 1; - pReq.tagVer = pTableMeta->tversion; - pReq.suid = pTableMeta->uid; - pReq.source = TD_REQ_FROM_TAOX; - } - if (pReq.numOfTags == 0) { pReq.numOfTags = 1; SField field = {0}; @@ -1336,7 +1344,7 @@ static int32_t smlInsertData(SSmlHandle *info) { if (info->pRequest->dbList == NULL) { info->pRequest->dbList = taosArrayInit(1, TSDB_DB_FNAME_LEN); } - char *data = (char*)taosArrayReserve(info->pRequest->dbList, 1); + char *data = (char *)taosArrayReserve(info->pRequest->dbList, 1); SName pName = {TSDB_TABLE_NAME_T, info->taos->acctId, {0}, {0}}; tstrncpy(pName.dbname, info->pRequest->pDb, sizeof(pName.dbname)); tNameGetFullDbName(&pName, data); @@ -1578,9 +1586,9 @@ static int smlProcess(SSmlHandle *info, char *lines[], char *rawLine, char *rawL do { code = smlModifyDBSchemas(info); - if (code == 0 || code == TSDB_CODE_SML_INVALID_DATA || code == TSDB_CODE_PAR_TOO_MANY_COLUMNS - || code == TSDB_CODE_PAR_INVALID_TAGS_NUM || code == TSDB_CODE_PAR_INVALID_TAGS_LENGTH - || code == TSDB_CODE_PAR_INVALID_ROW_LENGTH || code == TSDB_CODE_MND_FIELD_VALUE_OVERFLOW) { + if (code == 0 || code == TSDB_CODE_SML_INVALID_DATA || code == TSDB_CODE_PAR_TOO_MANY_COLUMNS || + code == TSDB_CODE_PAR_INVALID_TAGS_NUM || code == TSDB_CODE_PAR_INVALID_TAGS_LENGTH || + code == TSDB_CODE_PAR_INVALID_ROW_LENGTH || code == TSDB_CODE_MND_FIELD_VALUE_OVERFLOW) { break; } taosMsleep(100); @@ -1602,6 +1610,44 @@ static int smlProcess(SSmlHandle *info, char *lines[], char *rawLine, char *rawL return code; } +void smlSetReqSQL(SRequestObj *request, char *lines[], char *rawLine, char *rawLineEnd) { + if (tsSlowLogScope & SLOW_LOG_TYPE_INSERT) { + int32_t len = 0; + int32_t rlen = 0; + char *p = NULL; + + if (lines && lines[0]) { + len = strlen(lines[0]); + p = lines[0]; + } else if (rawLine) { + if (rawLineEnd) { + len = rawLineEnd - rawLine; + } else { + len = strlen(rawLine); + } + p = rawLine; + } + + if (NULL == p) { + return; + } + + rlen = TMIN(len, TSDB_MAX_ALLOWED_SQL_LEN); + rlen = TMAX(rlen, 0); + + char *sql = taosMemoryMalloc(rlen + 1); + if (NULL == sql) { + uError("malloc %d for sml sql failed", rlen + 1); + return; + } + memcpy(sql, p, rlen); + sql[rlen] = 0; + + request->sqlstr = sql; + request->sqlLen = rlen; + } +} + TAOS_RES *taos_schemaless_insert_inner(TAOS *taos, char *lines[], char *rawLine, char *rawLineEnd, int numLines, int protocol, int precision, int32_t ttl, int64_t reqid) { int32_t code = TSDB_CODE_SUCCESS; @@ -1634,6 +1680,8 @@ TAOS_RES *taos_schemaless_insert_inner(TAOS *taos, char *lines[], char *rawLine, info->msgBuf.len = ERROR_MSG_BUF_DEFAULT_SIZE; info->lineNum = numLines; + smlSetReqSQL(request, lines, rawLine, rawLineEnd); + SSmlMsgBuf msg = {ERROR_MSG_BUF_DEFAULT_SIZE, request->msgBuf}; if (request->pDb == NULL) { request->code = TSDB_CODE_PAR_DB_NOT_SPECIFIED; diff --git a/source/client/src/clientSmlLine.c b/source/client/src/clientSmlLine.c index 2f7e8a0f97..e79093398e 100644 --- a/source/client/src/clientSmlLine.c +++ b/source/client/src/clientSmlLine.c @@ -27,29 +27,28 @@ // equal = #define IS_EQUAL(sql) (*(sql) == EQUAL && *((sql)-1) != SLASH) // quote " -//#define IS_QUOTE(sql) (*(sql) == QUOTE && *((sql)-1) != SLASH) +// #define IS_QUOTE(sql) (*(sql) == QUOTE && *((sql)-1) != SLASH) // SLASH -#define IS_SLASH_LETTER_IN_FIELD_VALUE(sql) \ - (*((sql)-1) == SLASH && (*(sql) == QUOTE || *(sql) == SLASH)) +#define IS_SLASH_LETTER_IN_FIELD_VALUE(sql) (*((sql)-1) == SLASH && (*(sql) == QUOTE || *(sql) == SLASH)) -#define IS_SLASH_LETTER_IN_TAG_FIELD_KEY(sql) \ +#define IS_SLASH_LETTER_IN_TAG_FIELD_KEY(sql) \ (*((sql)-1) == SLASH && (*(sql) == COMMA || *(sql) == SPACE || *(sql) == EQUAL)) -#define PROCESS_SLASH_IN_FIELD_VALUE(key, keyLen) \ - for (int i = 1; i < keyLen; ++i) { \ - if (IS_SLASH_LETTER_IN_FIELD_VALUE(key + i)) { \ - MOVE_FORWARD_ONE(key + i, keyLen - i); \ - keyLen--; \ - } \ +#define PROCESS_SLASH_IN_FIELD_VALUE(key, keyLen) \ + for (int i = 1; i < keyLen; ++i) { \ + if (IS_SLASH_LETTER_IN_FIELD_VALUE(key + i)) { \ + MOVE_FORWARD_ONE(key + i, keyLen - i); \ + keyLen--; \ + } \ } -#define PROCESS_SLASH_IN_TAG_FIELD_KEY(key, keyLen) \ - for (int i = 1; i < keyLen; ++i) { \ - if (IS_SLASH_LETTER_IN_TAG_FIELD_KEY(key + i)) { \ - MOVE_FORWARD_ONE(key + i, keyLen - i); \ - keyLen--; \ - } \ +#define PROCESS_SLASH_IN_TAG_FIELD_KEY(key, keyLen) \ + for (int i = 1; i < keyLen; ++i) { \ + if (IS_SLASH_LETTER_IN_TAG_FIELD_KEY(key + i)) { \ + MOVE_FORWARD_ONE(key + i, keyLen - i); \ + keyLen--; \ + } \ } #define BINARY_ADD_LEN 2 // "binary" 2 means " " @@ -152,15 +151,15 @@ static int32_t smlParseTagKv(SSmlHandle *info, char **sql, char *sqlEnd, SSmlLin SSmlSTableMeta *sMeta = NULL; if (unlikely(tmp == NULL)) { - char* measure = currElement->measure; + char *measure = currElement->measure; int measureLen = currElement->measureLen; - if(currElement->measureEscaped){ - measure = (char*)taosMemoryMalloc(currElement->measureLen); + if (currElement->measureEscaped) { + measure = (char *)taosMemoryMalloc(currElement->measureLen); memcpy(measure, currElement->measure, currElement->measureLen); PROCESS_SLASH_IN_MEASUREMENT(measure, measureLen); } STableMeta *pTableMeta = smlGetMeta(info, measure, measureLen); - if(currElement->measureEscaped){ + if (currElement->measureEscaped) { taosMemoryFree(measure); } if (pTableMeta == NULL) { @@ -171,9 +170,13 @@ static int32_t smlParseTagKv(SSmlHandle *info, char **sql, char *sqlEnd, SSmlLin sMeta = smlBuildSTableMeta(info->dataFormat); sMeta->tableMeta = pTableMeta; taosHashPut(info->superTables, currElement->measure, currElement->measureLen, &sMeta, POINTER_BYTES); - for(int i = pTableMeta->tableInfo.numOfColumns; i < pTableMeta->tableInfo.numOfTags + pTableMeta->tableInfo.numOfColumns; i++){ + for (int i = pTableMeta->tableInfo.numOfColumns; + i < pTableMeta->tableInfo.numOfTags + pTableMeta->tableInfo.numOfColumns; i++) { SSchema *tag = pTableMeta->schema + i; - SSmlKv kv = {.key = tag->name, .keyLen = strlen(tag->name), .type = tag->type, .length = (tag->bytes - VARSTR_HEADER_SIZE) / TSDB_NCHAR_SIZE }; + SSmlKv kv = {.key = tag->name, + .keyLen = strlen(tag->name), + .type = tag->type, + .length = (tag->bytes - VARSTR_HEADER_SIZE) / TSDB_NCHAR_SIZE}; taosArrayPush(sMeta->tags, &kv); } tmp = &sMeta; @@ -248,19 +251,25 @@ static int32_t smlParseTagKv(SSmlHandle *info, char **sql, char *sqlEnd, SSmlLin return TSDB_CODE_PAR_INVALID_VAR_COLUMN_LEN; } - if (keyEscaped){ - char *tmp = (char*)taosMemoryMalloc(keyLen); + if (keyEscaped) { + char *tmp = (char *)taosMemoryMalloc(keyLen); memcpy(tmp, key, keyLen); PROCESS_SLASH_IN_TAG_FIELD_KEY(tmp, keyLen); key = tmp; } - if (valueEscaped){ - char *tmp = (char*)taosMemoryMalloc(valueLen); + if (valueEscaped) { + char *tmp = (char *)taosMemoryMalloc(valueLen); memcpy(tmp, value, valueLen); PROCESS_SLASH_IN_TAG_FIELD_KEY(tmp, valueLen); value = tmp; } - SSmlKv kv = {.key = key, .keyLen = keyLen, .type = TSDB_DATA_TYPE_NCHAR, .value = value, .length = valueLen, .keyEscaped = keyEscaped, .valueEscaped = valueEscaped}; + SSmlKv kv = {.key = key, + .keyLen = keyLen, + .type = TSDB_DATA_TYPE_NCHAR, + .value = value, + .length = valueLen, + .keyEscaped = keyEscaped, + .valueEscaped = valueEscaped}; taosArrayPush(preLineKV, &kv); if (info->dataFormat) { if (unlikely(cnt + 1 > info->currSTableMeta->tableInfo.numOfTags)) { @@ -305,10 +314,10 @@ static int32_t smlParseTagKv(SSmlHandle *info, char **sql, char *sqlEnd, SSmlLin return TSDB_CODE_OUT_OF_MEMORY; } tinfo->tags = taosArrayDup(preLineKV, NULL); - for(size_t i = 0; i < taosArrayGetSize(preLineKV); i++){ + for (size_t i = 0; i < taosArrayGetSize(preLineKV); i++) { SSmlKv *kv = (SSmlKv *)taosArrayGet(preLineKV, i); - if(kv->keyEscaped)kv->key = NULL; - if(kv->valueEscaped)kv->value = NULL; + if (kv->keyEscaped) kv->key = NULL; + if (kv->valueEscaped) kv->value = NULL; } smlSetCTableName(tinfo); @@ -330,7 +339,7 @@ static int32_t smlParseTagKv(SSmlHandle *info, char **sql, char *sqlEnd, SSmlLin static int32_t smlParseColKv(SSmlHandle *info, char **sql, char *sqlEnd, SSmlLineInfo *currElement, bool isSameMeasure, bool isSameCTable) { - int cnt = 0; + int cnt = 0; if (info->dataFormat) { if (unlikely(!isSameCTable)) { SSmlTableInfo **oneTable = @@ -346,15 +355,15 @@ static int32_t smlParseColKv(SSmlHandle *info, char **sql, char *sqlEnd, SSmlLin SSmlSTableMeta **tmp = (SSmlSTableMeta **)taosHashGet(info->superTables, currElement->measure, currElement->measureLen); if (unlikely(tmp == NULL)) { - char* measure = currElement->measure; + char *measure = currElement->measure; int measureLen = currElement->measureLen; - if(currElement->measureEscaped){ - measure = (char*)taosMemoryMalloc(currElement->measureLen); + if (currElement->measureEscaped) { + measure = (char *)taosMemoryMalloc(currElement->measureLen); memcpy(measure, currElement->measure, currElement->measureLen); PROCESS_SLASH_IN_MEASUREMENT(measure, measureLen); } STableMeta *pTableMeta = smlGetMeta(info, measure, measureLen); - if(currElement->measureEscaped){ + if (currElement->measureEscaped) { taosMemoryFree(measure); } if (pTableMeta == NULL) { @@ -366,12 +375,12 @@ static int32_t smlParseColKv(SSmlHandle *info, char **sql, char *sqlEnd, SSmlLin (*tmp)->tableMeta = pTableMeta; taosHashPut(info->superTables, currElement->measure, currElement->measureLen, tmp, POINTER_BYTES); - for(int i = 0; i < pTableMeta->tableInfo.numOfColumns; i++){ + for (int i = 0; i < pTableMeta->tableInfo.numOfColumns; i++) { SSchema *tag = pTableMeta->schema + i; - SSmlKv kv = {.key = tag->name, .keyLen = strlen(tag->name), .type = tag->type }; - if(tag->type == TSDB_DATA_TYPE_NCHAR){ + SSmlKv kv = {.key = tag->name, .keyLen = strlen(tag->name), .type = tag->type}; + if (tag->type == TSDB_DATA_TYPE_NCHAR) { kv.length = (tag->bytes - VARSTR_HEADER_SIZE) / TSDB_NCHAR_SIZE; - }else if(tag->type == TSDB_DATA_TYPE_BINARY){ + } else if (tag->type == TSDB_DATA_TYPE_BINARY) { kv.length = tag->bytes - VARSTR_HEADER_SIZE; } taosArrayPush((*tmp)->cols, &kv); @@ -459,16 +468,16 @@ static int32_t smlParseColKv(SSmlHandle *info, char **sql, char *sqlEnd, SSmlLin return ret; } - if (keyEscaped){ - char *tmp = (char*)taosMemoryMalloc(kv.keyLen); + if (keyEscaped) { + char *tmp = (char *)taosMemoryMalloc(kv.keyLen); memcpy(tmp, key, kv.keyLen); PROCESS_SLASH_IN_TAG_FIELD_KEY(tmp, kv.keyLen); kv.key = tmp; kv.keyEscaped = keyEscaped; } - if (valueEscaped){ - char *tmp = (char*)taosMemoryMalloc(kv.length); + if (valueEscaped) { + char *tmp = (char *)taosMemoryMalloc(kv.length); memcpy(tmp, kv.value, kv.length); PROCESS_SLASH_IN_FIELD_VALUE(tmp, kv.length); kv.value = tmp; @@ -643,9 +652,13 @@ int32_t smlParseInfluxString(SSmlHandle *info, char *sql, char *sqlEnd, SSmlLine if (info->dataFormat) { uDebug("SML:0x%" PRIx64 " smlParseInfluxString format true, ts:%" PRId64, info->id, ts); ret = smlBuildCol(info->currTableDataCtx, info->currSTableMeta->schema, &kv, 0); - if(ret != TSDB_CODE_SUCCESS){return ret;} + if (ret != TSDB_CODE_SUCCESS) { + return ret; + } ret = smlBuildRow(info->currTableDataCtx); - if(ret != TSDB_CODE_SUCCESS){return ret;} + if (ret != TSDB_CODE_SUCCESS) { + return ret; + } clearColValArray(info->currTableDataCtx->pValues); } else { uDebug("SML:0x%" PRIx64 " smlParseInfluxString format false, ts:%" PRId64, info->id, ts); diff --git a/source/client/src/clientSmlTelnet.c b/source/client/src/clientSmlTelnet.c index c5dd20ba7b..42b8001e59 100644 --- a/source/client/src/clientSmlTelnet.c +++ b/source/client/src/clientSmlTelnet.c @@ -158,7 +158,7 @@ static int32_t smlParseTelnetTags(SSmlHandle *info, char *data, char *sqlEnd, SS return TSDB_CODE_TSC_INVALID_VALUE; } - if (unlikely(valueLen > (TSDB_MAX_NCHAR_LEN - VARSTR_HEADER_SIZE) / TSDB_NCHAR_SIZE)) { + if (unlikely(valueLen > (TSDB_MAX_TAGS_LEN - VARSTR_HEADER_SIZE) / TSDB_NCHAR_SIZE)) { return TSDB_CODE_PAR_INVALID_VAR_COLUMN_LEN; } diff --git a/source/client/src/clientTmq.c b/source/client/src/clientTmq.c index b488af9ba1..9e0065a5f4 100644 --- a/source/client/src/clientTmq.c +++ b/source/client/src/clientTmq.c @@ -133,16 +133,23 @@ enum { TMQ_DELAYED_TASK__COMMIT, }; -typedef struct { - int64_t pollCnt; - int64_t numOfRows; +typedef struct SVgOffsetInfo { STqOffsetVal committedOffset; STqOffsetVal currentOffset; - int32_t vgId; - int32_t vgStatus; - int32_t vgSkipCnt; - int64_t emptyBlockReceiveTs; // once empty block is received, idle for ignoreCnt then start to poll data - SEpSet epSet; + int64_t walVerBegin; + int64_t walVerEnd; +} SVgOffsetInfo; + +typedef struct { + int64_t pollCnt; + int64_t numOfRows; + SVgOffsetInfo offsetInfo; + int32_t vgId; + int32_t vgStatus; + int32_t vgSkipCnt; // here used to mark the slow vgroups + bool receiveInfo; + int64_t emptyBlockReceiveTs; // once empty block is received, idle for ignoreCnt then start to poll data + SEpSet epSet; } SMqClientVg; typedef struct { @@ -190,6 +197,23 @@ typedef struct { uint64_t requestId; // request id for debug purpose } SMqPollCbParam; +typedef struct SMqVgCommon { + tsem_t rsp; + int32_t numOfRsp; + SArray* pList; + TdThreadMutex mutex; + int64_t consumerId; + char* pTopicName; + int32_t code; +} SMqVgCommon; + +typedef struct SMqVgWalInfoParam { + int32_t vgId; + int32_t epoch; + int32_t totalReq; + SMqVgCommon* pCommon; +} SMqVgWalInfoParam; + typedef struct { int64_t refId; int32_t epoch; @@ -204,7 +228,7 @@ typedef struct { typedef struct { SMqCommitCbParamSet* params; - STqOffset* pOffset; + SMqVgOffset* pOffset; char topicName[TSDB_TOPIC_FNAME_LEN]; int32_t vgId; tmq_t* pTmq; @@ -219,7 +243,7 @@ static int32_t doAskEp(tmq_t* tmq); static int32_t makeTopicVgroupKey(char* dst, const char* topicName, int32_t vg); static int32_t tmqCommitDone(SMqCommitCbParamSet* pParamSet); static int32_t doSendCommitMsg(tmq_t* tmq, SMqClientVg* pVg, const char* pTopicName, SMqCommitCbParamSet* pParamSet, - int32_t index, int32_t totalVgroups); + int32_t index, int32_t totalVgroups, int32_t type); static void commitRspCountDown(SMqCommitCbParamSet* pParamSet, int64_t consumerId, const char* pTopic, int32_t vgId); static void asyncAskEp(tmq_t* pTmq, __tmq_askep_fn_t askEpFn, void* param); static void addToQueueCallbackFn(tmq_t* pTmq, int32_t code, SDataBuf* pDataBuf, void* param); @@ -438,7 +462,7 @@ static int32_t tmqCommitCb(void* param, SDataBuf* pBuf, int32_t code) { // if (code1 != TSDB_CODE_SUCCESS) { // retry failed. // tscError("consumer:0x%" PRIx64 " topic:%s vgId:%d offset:%" PRId64 // " retry failed, ignore this commit. code:%s ordinal:%d/%d", - // pParam->pTmq->consumerId, pParam->topicName, pVg->vgId, pVg->committedOffset.version, + // pParam->pTmq->consumerId, pParam->topicName, pVg->vgId, pVg->offsetInfo.committedOffset.version, // tstrerror(terrno), index + 1, numOfVgroups); // } // } @@ -464,22 +488,23 @@ static int32_t tmqCommitCb(void* param, SDataBuf* pBuf, int32_t code) { } static int32_t doSendCommitMsg(tmq_t* tmq, SMqClientVg* pVg, const char* pTopicName, SMqCommitCbParamSet* pParamSet, - int32_t index, int32_t totalVgroups) { - STqOffset* pOffset = taosMemoryCalloc(1, sizeof(STqOffset)); + int32_t index, int32_t totalVgroups, int32_t type) { + SMqVgOffset* pOffset = taosMemoryCalloc(1, sizeof(SMqVgOffset)); if (pOffset == NULL) { return TSDB_CODE_OUT_OF_MEMORY; } - pOffset->val = pVg->currentOffset; + pOffset->consumerId = tmq->consumerId; + pOffset->offset.val = pVg->offsetInfo.currentOffset; int32_t groupLen = strlen(tmq->groupId); - memcpy(pOffset->subKey, tmq->groupId, groupLen); - pOffset->subKey[groupLen] = TMQ_SEPARATOR; - strcpy(pOffset->subKey + groupLen + 1, pTopicName); + memcpy(pOffset->offset.subKey, tmq->groupId, groupLen); + pOffset->offset.subKey[groupLen] = TMQ_SEPARATOR; + strcpy(pOffset->offset.subKey + groupLen + 1, pTopicName); int32_t len = 0; int32_t code = 0; - tEncodeSize(tEncodeSTqOffset, pOffset, len, code); + tEncodeSize(tEncodeMqVgOffset, pOffset, len, code); if (code < 0) { return TSDB_CODE_INVALID_PARA; } @@ -496,7 +521,7 @@ static int32_t doSendCommitMsg(tmq_t* tmq, SMqClientVg* pVg, const char* pTopicN SEncoder encoder; tEncoderInit(&encoder, abuf, len); - tEncodeSTqOffset(&encoder, pOffset); + tEncodeMqVgOffset(&encoder, pOffset); tEncoderClear(&encoder); // build param @@ -530,19 +555,19 @@ static int32_t doSendCommitMsg(tmq_t* tmq, SMqClientVg* pVg, const char* pTopicN pMsgSendInfo->param = pParam; pMsgSendInfo->paramFreeFp = taosMemoryFree; pMsgSendInfo->fp = tmqCommitCb; - pMsgSendInfo->msgType = TDMT_VND_TMQ_COMMIT_OFFSET; + pMsgSendInfo->msgType = type; atomic_add_fetch_32(&pParamSet->waitingRspNum, 1); atomic_add_fetch_32(&pParamSet->totalRspNum, 1); SEp* pEp = GET_ACTIVE_EP(&pVg->epSet); char offsetBuf[80] = {0}; - tFormatOffset(offsetBuf, tListLen(offsetBuf), &pOffset->val); + tFormatOffset(offsetBuf, tListLen(offsetBuf), &pOffset->offset.val); char commitBuf[80] = {0}; - tFormatOffset(commitBuf, tListLen(commitBuf), &pVg->committedOffset); + tFormatOffset(commitBuf, tListLen(commitBuf), &pVg->offsetInfo.committedOffset); tscDebug("consumer:0x%" PRIx64 " topic:%s on vgId:%d send offset:%s prev:%s, ep:%s:%d, ordinal:%d/%d, req:0x%" PRIx64, - tmq->consumerId, pOffset->subKey, pVg->vgId, offsetBuf, commitBuf, pEp->fqdn, pEp->port, index + 1, + tmq->consumerId, pOffset->offset.subKey, pVg->vgId, offsetBuf, commitBuf, pEp->fqdn, pEp->port, index + 1, totalVgroups, pMsgSendInfo->requestId); int64_t transporterId = 0; @@ -551,7 +576,22 @@ static int32_t doSendCommitMsg(tmq_t* tmq, SMqClientVg* pVg, const char* pTopicN return TSDB_CODE_SUCCESS; } -static void asyncCommitOffset(tmq_t* tmq, const TAOS_RES* pRes, tmq_commit_cb* pCommitFp, void* userParam) { +static SMqClientTopic* getTopicByName(tmq_t* tmq, const char* pTopicName) { + int32_t numOfTopics = taosArrayGetSize(tmq->clientTopics); + for (int32_t i = 0; i < numOfTopics; ++i) { + SMqClientTopic* pTopic = taosArrayGet(tmq->clientTopics, i); + if (strcmp(pTopic->topicName, pTopicName) != 0) { + continue; + } + + return pTopic; + } + + tscError("consumer:0x%" PRIx64 ", total:%d, failed to find topic:%s", tmq->consumerId, numOfTopics, pTopicName); + return NULL; +} + +static void asyncCommitOffset(tmq_t* tmq, const TAOS_RES* pRes, int32_t type, tmq_commit_cb* pCommitFp, void* userParam) { char* pTopicName = NULL; int32_t vgId = 0; int32_t code = 0; @@ -593,15 +633,8 @@ static void asyncCommitOffset(tmq_t* tmq, const TAOS_RES* pRes, tmq_commit_cb* p tscDebug("consumer:0x%" PRIx64 " do manual commit offset for %s, vgId:%d", tmq->consumerId, pTopicName, vgId); - int32_t i = 0; - for (; i < numOfTopics; i++) { - SMqClientTopic* pTopic = taosArrayGet(tmq->clientTopics, i); - if (strcmp(pTopic->topicName, pTopicName) == 0) { - break; - } - } - - if (i == numOfTopics) { + SMqClientTopic* pTopic = getTopicByName(tmq, pTopicName); + if (pTopic == NULL) { tscWarn("consumer:0x%" PRIx64 " failed to find the specified topic:%s, total topics:%d", tmq->consumerId, pTopicName, numOfTopics); taosMemoryFree(pParamSet); @@ -609,8 +642,6 @@ static void asyncCommitOffset(tmq_t* tmq, const TAOS_RES* pRes, tmq_commit_cb* p return; } - SMqClientTopic* pTopic = taosArrayGet(tmq->clientTopics, i); - int32_t j = 0; int32_t numOfVgroups = taosArrayGetSize(pTopic->vgs); for (j = 0; j < numOfVgroups; j++) { @@ -629,8 +660,8 @@ static void asyncCommitOffset(tmq_t* tmq, const TAOS_RES* pRes, tmq_commit_cb* p } SMqClientVg* pVg = taosArrayGet(pTopic->vgs, j); - if (pVg->currentOffset.type > 0 && !tOffsetEqual(&pVg->currentOffset, &pVg->committedOffset)) { - code = doSendCommitMsg(tmq, pVg, pTopic->topicName, pParamSet, j, numOfVgroups); + if (pVg->offsetInfo.currentOffset.type > 0 && !tOffsetEqual(&pVg->offsetInfo.currentOffset, &pVg->offsetInfo.committedOffset)) { + code = doSendCommitMsg(tmq, pVg, pTopic->topicName, pParamSet, j, numOfVgroups, type); // failed to commit, callback user function directly. if (code != TSDB_CODE_SUCCESS) { @@ -670,20 +701,20 @@ static void asyncCommitAllOffsets(tmq_t* tmq, tmq_commit_cb* pCommitFp, void* us for (int32_t j = 0; j < numOfVgroups; j++) { SMqClientVg* pVg = taosArrayGet(pTopic->vgs, j); - if (pVg->currentOffset.type > 0 && !tOffsetEqual(&pVg->currentOffset, &pVg->committedOffset)) { - int32_t code = doSendCommitMsg(tmq, pVg, pTopic->topicName, pParamSet, j, numOfVgroups); + if (pVg->offsetInfo.currentOffset.type > 0 && !tOffsetEqual(&pVg->offsetInfo.currentOffset, &pVg->offsetInfo.committedOffset)) { + int32_t code = doSendCommitMsg(tmq, pVg, pTopic->topicName, pParamSet, j, numOfVgroups, TDMT_VND_TMQ_COMMIT_OFFSET); if (code != TSDB_CODE_SUCCESS) { tscError("consumer:0x%" PRIx64 " topic:%s vgId:%d offset:%" PRId64 " failed, code:%s ordinal:%d/%d", - tmq->consumerId, pTopic->topicName, pVg->vgId, pVg->committedOffset.version, tstrerror(terrno), + tmq->consumerId, pTopic->topicName, pVg->vgId, pVg->offsetInfo.committedOffset.version, tstrerror(terrno), j + 1, numOfVgroups); continue; } // update the offset value. - pVg->committedOffset = pVg->currentOffset; + pVg->offsetInfo.committedOffset = pVg->offsetInfo.currentOffset; } else { tscDebug("consumer:0x%" PRIx64 " topic:%s vgId:%d, no commit, current:%" PRId64 ", ordinal:%d/%d", - tmq->consumerId, pTopic->topicName, pVg->vgId, pVg->currentOffset.version, j + 1, numOfVgroups); + tmq->consumerId, pTopic->topicName, pVg->vgId, pVg->offsetInfo.currentOffset.version, j + 1, numOfVgroups); } } } @@ -1085,7 +1116,7 @@ _failed: } int32_t tmq_subscribe(tmq_t* tmq, const tmq_list_t* topic_list) { - const int32_t MAX_RETRY_COUNT = 120 * 2; // let's wait for 2 mins at most + const int32_t MAX_RETRY_COUNT = 120 * 4; // let's wait for 4 mins at most const SArray* container = &topic_list->container; int32_t sz = taosArrayGetSize(container); void* buf = NULL; @@ -1138,22 +1169,13 @@ int32_t tmq_subscribe(tmq_t* tmq, const tmq_list_t* topic_list) { goto FAIL; } - SMqSubscribeCbParam param = { - .rspErr = 0, - .refId = tmq->refId, - .epoch = tmq->epoch, - }; - + SMqSubscribeCbParam param = { .rspErr = 0, .refId = tmq->refId, .epoch = tmq->epoch }; if (tsem_init(¶m.rspSem, 0, 0) != 0) { code = TSDB_CODE_TSC_INTERNAL_ERROR; goto FAIL; } - sendInfo->msgInfo = (SDataBuf){ - .pData = buf, - .len = tlen, - .handle = NULL, - }; + sendInfo->msgInfo = (SDataBuf){.pData = buf, .len = tlen, .handle = NULL}; sendInfo->requestId = generateRequestId(); sendInfo->requestObjRefId = 0; @@ -1181,7 +1203,7 @@ int32_t tmq_subscribe(tmq_t* tmq, const tmq_list_t* topic_list) { int32_t retryCnt = 0; while (TSDB_CODE_MND_CONSUMER_NOT_READY == doAskEp(tmq)) { if (retryCnt++ > MAX_RETRY_COUNT) { - tscError("consumer:0x%" PRIx64 ", mnd not ready for subscribe, retry:%d in 500ms", tmq->consumerId, retryCnt); + tscError("consumer:0x%" PRIx64 ", mnd not ready for subscribe, max retry reached:%d", tmq->consumerId, retryCnt); code = TSDB_CODE_TSC_INTERNAL_ERROR; goto FAIL; } @@ -1217,7 +1239,7 @@ void tmq_conf_set_auto_commit_cb(tmq_conf_t* conf, tmq_commit_cb* cb, void* para conf->commitCbUserParam = param; } -int32_t tmqPollCb(void* param, SDataBuf* pMsg, int32_t code) { +static int32_t tmqPollCb(void* param, SDataBuf* pMsg, int32_t code) { SMqPollCbParam* pParam = (SMqPollCbParam*)param; int64_t refId = pParam->refId; @@ -1270,12 +1292,12 @@ int32_t tmqPollCb(void* param, SDataBuf* pMsg, int32_t code) { } int32_t msgEpoch = ((SMqRspHead*)pMsg->pData)->epoch; - int32_t tmqEpoch = atomic_load_32(&tmq->epoch); - if (msgEpoch < tmqEpoch) { + int32_t clientEpoch = atomic_load_32(&tmq->epoch); + if (msgEpoch < clientEpoch) { // do not write into queue since updating epoch reset tscWarn("consumer:0x%" PRIx64 " msg discard from vgId:%d since from earlier epoch, rsp epoch %d, current epoch %d, reqId:0x%" PRIx64, - tmq->consumerId, vgId, msgEpoch, tmqEpoch, requestId); + tmq->consumerId, vgId, msgEpoch, clientEpoch, requestId); tsem_post(&tmq->rspSem); taosReleaseRef(tmqMgmt.rsetId, refId); @@ -1285,9 +1307,9 @@ int32_t tmqPollCb(void* param, SDataBuf* pMsg, int32_t code) { return 0; } - if (msgEpoch != tmqEpoch) { + if (msgEpoch != clientEpoch) { tscWarn("consumer:0x%" PRIx64 " mismatch rsp from vgId:%d, epoch %d, current epoch %d, reqId:0x%" PRIx64, - tmq->consumerId, vgId, msgEpoch, tmqEpoch, requestId); + tmq->consumerId, vgId, msgEpoch, clientEpoch, requestId); } // handle meta rsp @@ -1313,7 +1335,7 @@ int32_t tmqPollCb(void* param, SDataBuf* pMsg, int32_t code) { if (rspType == TMQ_MSG_TYPE__POLL_RSP) { SDecoder decoder; tDecoderInit(&decoder, POINTER_SHIFT(pMsg->pData, sizeof(SMqRspHead)), pMsg->len - sizeof(SMqRspHead)); - tDecodeSMqDataRsp(&decoder, &pRspWrapper->dataRsp); + tDecodeMqDataRsp(&decoder, &pRspWrapper->dataRsp); tDecoderClear(&decoder); memcpy(&pRspWrapper->dataRsp, pMsg->pData, sizeof(SMqRspHead)); @@ -1324,7 +1346,7 @@ int32_t tmqPollCb(void* param, SDataBuf* pMsg, int32_t code) { } else if (rspType == TMQ_MSG_TYPE__POLL_META_RSP) { SDecoder decoder; tDecoderInit(&decoder, POINTER_SHIFT(pMsg->pData, sizeof(SMqRspHead)), pMsg->len - sizeof(SMqRspHead)); - tDecodeSMqMetaRsp(&decoder, &pRspWrapper->metaRsp); + tDecodeMqMetaRsp(&decoder, &pRspWrapper->metaRsp); tDecoderClear(&decoder); memcpy(&pRspWrapper->metaRsp, pMsg->pData, sizeof(SMqRspHead)); } else if (rspType == TMQ_MSG_TYPE__TAOSX_RSP) { @@ -1395,7 +1417,6 @@ static void initClientTopicFromRsp(SMqClientTopic* pTopic, SMqSubTopicEp* pTopic SMqClientVg clientVg = { .pollCnt = 0, - .currentOffset = offsetNew, .vgId = pVgEp->vgId, .epSet = pVgEp->epSet, .vgStatus = TMQ_VG_STATUS__IDLE, @@ -1404,6 +1425,10 @@ static void initClientTopicFromRsp(SMqClientTopic* pTopic, SMqSubTopicEp* pTopic .numOfRows = numOfRows, }; + clientVg.offsetInfo.currentOffset = offsetNew; + clientVg.offsetInfo.committedOffset = offsetNew; + clientVg.offsetInfo.walVerBegin = -1; + clientVg.offsetInfo.walVerEnd = -1; taosArrayPush(pTopic->vgs, &clientVg); } } @@ -1453,11 +1478,16 @@ static bool doUpdateLocalEp(tmq_t* tmq, int32_t epoch, const SMqAskEpRsp* pRsp) makeTopicVgroupKey(vgKey, pTopicCur->topicName, pVgCur->vgId); char buf[80]; +<<<<<<< HEAD tFormatOffset(buf, 80, &pVgCur->currentOffset); tscDebug("consumer:0x%" PRIx64 ", doUpdateLocalEp current vg, epoch:%d vgId:%d vgKey:%s, offset:%s", tmq->consumerId, tmq->epoch, pVgCur->vgId, +======= + tFormatOffset(buf, 80, &pVgCur->offsetInfo.currentOffset); + tscDebug("consumer:0x%" PRIx64 ", epoch:%d vgId:%d vgKey:%s, offset:%s", tmq->consumerId, epoch, pVgCur->vgId, +>>>>>>> enh/3.0 vgKey, buf); - SVgroupSaveInfo info = {.offset = pVgCur->currentOffset, .numOfRows = pVgCur->numOfRows}; + SVgroupSaveInfo info = {.offset = pVgCur->offsetInfo.currentOffset, .numOfRows = pVgCur->numOfRows}; taosHashPut(pVgOffsetHashMap, vgKey, strlen(vgKey), &info, sizeof(SVgroupSaveInfo)); } } @@ -1533,8 +1563,8 @@ int32_t askEpCallbackFn(void* param, SDataBuf* pMsg, int32_t code) { tscDebug("consumer:0x%" PRIx64 ", recv ep, msg epoch %d, current epoch %d, update local ep", tmq->consumerId, head->epoch, epoch); } - pParam->pUserFn(tmq, code, pMsg, pParam->pParam); + pParam->pUserFn(tmq, code, pMsg, pParam->pParam); taosReleaseRef(tmqMgmt.rsetId, pParam->refId); taosMemoryFree(pMsg->pEpSet); @@ -1553,8 +1583,7 @@ void tmqBuildConsumeReqImpl(SMqPollReq* pReq, tmq_t* tmq, int64_t timeout, SMqCl pReq->consumerId = tmq->consumerId; pReq->timeout = timeout; pReq->epoch = tmq->epoch; - /*pReq->currentOffset = reqOffset;*/ - pReq->reqOffset = pVg->currentOffset; + pReq->reqOffset = pVg->offsetInfo.currentOffset; pReq->head.vgId = pVg->vgId; pReq->useSnapshot = tmq->useSnapshot; pReq->reqId = generateRequestId(); @@ -1652,7 +1681,7 @@ static int32_t doTmqPollImpl(tmq_t* pTmq, SMqClientTopic* pTopic, SMqClientVg* p pParam->refId = pTmq->refId; pParam->epoch = pTmq->epoch; - pParam->pVg = pVg; // pVg may be released,fix it + pParam->pVg = pVg; pParam->pTopic = pTopic; pParam->vgId = pVg->vgId; pParam->requestId = req.reqId; @@ -1664,8 +1693,12 @@ static int32_t doTmqPollImpl(tmq_t* pTmq, SMqClientTopic* pTopic, SMqClientVg* p return handleErrorBeforePoll(pVg, pTmq); } +<<<<<<< HEAD sendInfo->msgInfo = (SDataBuf){ .pData = msg, .len = msgSize, .handle = NULL }; +======= + sendInfo->msgInfo = (SDataBuf){.pData = msg, .len = msgSize, .handle = NULL}; +>>>>>>> enh/3.0 sendInfo->requestId = req.reqId; sendInfo->requestObjRefId = 0; sendInfo->param = pParam; @@ -1674,7 +1707,7 @@ static int32_t doTmqPollImpl(tmq_t* pTmq, SMqClientTopic* pTopic, SMqClientVg* p int64_t transporterId = 0; char offsetFormatBuf[80]; - tFormatOffset(offsetFormatBuf, tListLen(offsetFormatBuf), &pVg->currentOffset); + tFormatOffset(offsetFormatBuf, tListLen(offsetFormatBuf), &pVg->offsetInfo.currentOffset); tscDebug("consumer:0x%" PRIx64 " send poll to %s vgId:%d, epoch %d, req:%s, reqId:0x%" PRIx64, pTmq->consumerId, pTopic->topicName, pVg->vgId, pTmq->epoch, offsetFormatBuf, req.reqId); @@ -1709,13 +1742,6 @@ static int32_t tmqPollImpl(tmq_t* tmq, int64_t timeout) { tscTrace("consumer:0x%" PRIx64 " epoch %d wait poll-rsp, skip vgId:%d skip cnt %d", tmq->consumerId, tmq->epoch, pVg->vgId, vgSkipCnt); continue; -#if 0 - if (skipCnt < 30000) { - continue; - } else { - tscDebug("consumer:0x%" PRIx64 ",skip vgId:%d skip too much reset", tmq->consumerId, pVg->vgId); - } -#endif } atomic_store_32(&pVg->vgSkipCnt, 0); @@ -1790,11 +1816,23 @@ static void* tmqHandleAllRsp(tmq_t* tmq, int64_t timeout, bool pollIfReset) { pVg->epSet = *pollRspWrapper->pEpset; } +<<<<<<< HEAD if(pDataRsp->rspOffset.type != 0){ // if offset is validate pVg->currentOffset = pDataRsp->rspOffset; // update the local offset value only for the returned values. } +======= + // update the local offset value only for the returned values. + pVg->offsetInfo.currentOffset = pDataRsp->rspOffset; + + // update the status +>>>>>>> enh/3.0 atomic_store_32(&pVg->vgStatus, TMQ_VG_STATUS__IDLE); + // update the valid wal version range + pVg->offsetInfo.walVerBegin = pDataRsp->head.walsver; + pVg->offsetInfo.walVerEnd = pDataRsp->head.walever; + pVg->receiveInfo = true; + char buf[80]; tFormatOffset(buf, 80, &pDataRsp->rspOffset); if (pDataRsp->blockNum == 0) { @@ -1823,6 +1861,7 @@ static void* tmqHandleAllRsp(tmq_t* tmq, int64_t timeout, bool pollIfReset) { taosFreeQitem(pollRspWrapper); } } else if (pRspWrapper->tmqRspType == TMQ_MSG_TYPE__POLL_META_RSP) { + // todo handle the wal range and epset for each vgroup SMqPollRspWrapper* pollRspWrapper = (SMqPollRspWrapper*)pRspWrapper; int32_t consumerEpoch = atomic_load_32(&tmq->epoch); @@ -1830,9 +1869,13 @@ static void* tmqHandleAllRsp(tmq_t* tmq, int64_t timeout, bool pollIfReset) { if (pollRspWrapper->metaRsp.head.epoch == consumerEpoch) { SMqClientVg* pVg = pollRspWrapper->vgHandle; +<<<<<<< HEAD if(pollRspWrapper->metaRsp.rspOffset.type != 0){ // if offset is validate pVg->currentOffset = pollRspWrapper->metaRsp.rspOffset; } +======= + pVg->offsetInfo.currentOffset = pollRspWrapper->metaRsp.rspOffset; +>>>>>>> enh/3.0 atomic_store_32(&pVg->vgStatus, TMQ_VG_STATUS__IDLE); // build rsp SMqMetaRspObj* pRsp = tmqBuildMetaRspFromWrapper(pollRspWrapper); @@ -1850,9 +1893,13 @@ static void* tmqHandleAllRsp(tmq_t* tmq, int64_t timeout, bool pollIfReset) { if (pollRspWrapper->taosxRsp.head.epoch == consumerEpoch) { SMqClientVg* pVg = pollRspWrapper->vgHandle; +<<<<<<< HEAD if(pollRspWrapper->taosxRsp.rspOffset.type != 0){ // if offset is validate pVg->currentOffset = pollRspWrapper->taosxRsp.rspOffset; } +======= + pVg->offsetInfo.currentOffset = pollRspWrapper->taosxRsp.rspOffset; +>>>>>>> enh/3.0 atomic_store_32(&pVg->vgStatus, TMQ_VG_STATUS__IDLE); if (pollRspWrapper->taosxRsp.blockNum == 0) { @@ -1878,7 +1925,7 @@ static void* tmqHandleAllRsp(tmq_t* tmq, int64_t timeout, bool pollIfReset) { tmq->totalRows += numOfRows; char buf[80]; - tFormatOffset(buf, 80, &pVg->currentOffset); + tFormatOffset(buf, 80, &pVg->offsetInfo.currentOffset); tscDebug("consumer:0x%" PRIx64 " process taosx poll rsp, vgId:%d, offset:%s, blocks:%d, rows:%" PRId64 ", vg total:%" PRId64 " total:%" PRId64 " reqId:0x%" PRIx64, tmq->consumerId, pVg->vgId, buf, pollRspWrapper->dataRsp.blockNum, numOfRows, pVg->numOfRows, @@ -1914,15 +1961,6 @@ TAOS_RES* tmq_consumer_poll(tmq_t* tmq, int64_t timeout) { tscDebug("consumer:0x%" PRIx64 " start to poll at %" PRId64 ", timeout:%" PRId64, tmq->consumerId, startTime, timeout); -#if 0 - tmqHandleAllDelayedTask(tmq); - tmqPollImpl(tmq, timeout); - rspObj = tmqHandleAllRsp(tmq, timeout, false); - if (rspObj) { - return (TAOS_RES*)rspObj; - } -#endif - // in no topic status, delayed task also need to be processed if (atomic_load_8(&tmq->status) == TMQ_CONSUMER_STATUS__INIT) { tscDebug("consumer:0x%" PRIx64 " poll return since consumer is init", tmq->consumerId); @@ -2118,11 +2156,11 @@ void tmq_commit_async(tmq_t* tmq, const TAOS_RES* pRes, tmq_commit_cb* cb, void* if (pRes == NULL) { // here needs to commit all offsets. asyncCommitAllOffsets(tmq, cb, param); } else { // only commit one offset - asyncCommitOffset(tmq, pRes, cb, param); + asyncCommitOffset(tmq, pRes, TDMT_VND_TMQ_COMMIT_OFFSET, cb, param); } } -static void commitCallBackFn(tmq_t *pTmq, int32_t code, void* param) { +static void commitCallBackFn(tmq_t *UNUSED_PARAM(tmq), int32_t code, void* param) { SSyncCommitInfo* pInfo = (SSyncCommitInfo*) param; pInfo->code = code; tsem_post(&pInfo->sem); @@ -2138,7 +2176,7 @@ int32_t tmq_commit_sync(tmq_t* tmq, const TAOS_RES* pRes) { if (pRes == NULL) { asyncCommitAllOffsets(tmq, commitCallBackFn, pInfo); } else { - asyncCommitOffset(tmq, pRes, commitCallBackFn, pInfo); + asyncCommitOffset(tmq, pRes, TDMT_VND_TMQ_COMMIT_OFFSET, commitCallBackFn, pInfo); } tsem_wait(&pInfo->sem); @@ -2322,4 +2360,272 @@ SReqResultInfo* tmqGetNextResInfo(TAOS_RES* res, bool convertUcs4) { } return NULL; +} + +static int32_t tmqGetWalInfoCb(void* param, SDataBuf* pMsg, int32_t code) { + SMqVgWalInfoParam* pParam = param; + SMqVgCommon* pCommon = pParam->pCommon; + + int32_t total = atomic_add_fetch_32(&pCommon->numOfRsp, 1); + if (code != TSDB_CODE_SUCCESS) { + tscError("consumer:0x%" PRIx64 " failed to get the wal info from vgId:%d for topic:%s", pCommon->consumerId, + pParam->vgId, pCommon->pTopicName); + pCommon->code = code; + } else { + SMqDataRsp rsp; + SDecoder decoder; + tDecoderInit(&decoder, POINTER_SHIFT(pMsg->pData, sizeof(SMqRspHead)), pMsg->len - sizeof(SMqRspHead)); + tDecodeMqDataRsp(&decoder, &rsp); + tDecoderClear(&decoder); + + SMqRspHead* pHead = pMsg->pData; + + tmq_topic_assignment assignment = {.begin = pHead->walsver, + .end = pHead->walever, + .currentOffset = rsp.rspOffset.version, + .vgId = pParam->vgId}; + + taosThreadMutexLock(&pCommon->mutex); + taosArrayPush(pCommon->pList, &assignment); + taosThreadMutexUnlock(&pCommon->mutex); + } + + if (total == pParam->totalReq) { + tsem_post(&pCommon->rsp); + } + + taosMemoryFree(pParam); + return 0; +} + +static void destroyCommonInfo(SMqVgCommon* pCommon) { + taosArrayDestroy(pCommon->pList); + tsem_destroy(&pCommon->rsp); + taosThreadMutexDestroy(&pCommon->mutex); + taosMemoryFree(pCommon->pTopicName); + taosMemoryFree(pCommon); +} + +int32_t tmq_get_topic_assignment(tmq_t* tmq, const char* pTopicName, tmq_topic_assignment** assignment, + int32_t* numOfAssignment) { + *numOfAssignment = 0; + *assignment = NULL; + + int32_t accId = tmq->pTscObj->acctId; + char tname[128] = {0}; + sprintf(tname, "%d.%s", accId, pTopicName); + + SMqClientTopic* pTopic = getTopicByName(tmq, tname); + if (pTopic == NULL) { + return TSDB_CODE_INVALID_PARA; + } + + // in case of snapshot is opened, no valid offset will return + *numOfAssignment = taosArrayGetSize(pTopic->vgs); + + *assignment = taosMemoryCalloc(*numOfAssignment, sizeof(tmq_topic_assignment)); + if (*assignment == NULL) { + tscError("consumer:0x%" PRIx64 " failed to malloc buffer, size:%" PRIzu, tmq->consumerId, + (*numOfAssignment) * sizeof(tmq_topic_assignment)); + return TSDB_CODE_OUT_OF_MEMORY; + } + + bool needFetch = false; + + for (int32_t j = 0; j < (*numOfAssignment); ++j) { + SMqClientVg* pClientVg = taosArrayGet(pTopic->vgs, j); + if (!pClientVg->receiveInfo) { + needFetch = true; + break; + } + + tmq_topic_assignment* pAssignment = &(*assignment)[j]; + if (pClientVg->offsetInfo.currentOffset.type == TMQ_OFFSET__LOG) { + pAssignment->currentOffset = pClientVg->offsetInfo.currentOffset.version; + } else { + pAssignment->currentOffset = 0; + } + + pAssignment->begin = pClientVg->offsetInfo.walVerBegin; + pAssignment->end = pClientVg->offsetInfo.walVerEnd; + pAssignment->vgId = pClientVg->vgId; + } + + if (needFetch) { + SMqVgCommon* pCommon = taosMemoryCalloc(1, sizeof(SMqVgCommon)); + if (pCommon == NULL) { + terrno = TSDB_CODE_OUT_OF_MEMORY; + return terrno; + } + + pCommon->pList= taosArrayInit(4, sizeof(tmq_topic_assignment)); + tsem_init(&pCommon->rsp, 0, 0); + taosThreadMutexInit(&pCommon->mutex, 0); + pCommon->pTopicName = taosStrdup(pTopic->topicName); + pCommon->consumerId = tmq->consumerId; + + terrno = TSDB_CODE_OUT_OF_MEMORY; + for (int32_t i = 0; i < (*numOfAssignment); ++i) { + SMqClientVg* pClientVg = taosArrayGet(pTopic->vgs, i); + + SMqVgWalInfoParam* pParam = taosMemoryMalloc(sizeof(SMqVgWalInfoParam)); + if (pParam == NULL) { + destroyCommonInfo(pCommon); + return terrno; + } + + pParam->epoch = tmq->epoch; + pParam->vgId = pClientVg->vgId; + pParam->totalReq = *numOfAssignment; + pParam->pCommon = pCommon; + + SMqPollReq req = {0}; + tmqBuildConsumeReqImpl(&req, tmq, 10, pTopic, pClientVg); + + int32_t msgSize = tSerializeSMqPollReq(NULL, 0, &req); + if (msgSize < 0) { + taosMemoryFree(pParam); + destroyCommonInfo(pCommon); + return terrno; + } + + char* msg = taosMemoryCalloc(1, msgSize); + if (NULL == msg) { + taosMemoryFree(pParam); + destroyCommonInfo(pCommon); + return terrno; + } + + if (tSerializeSMqPollReq(msg, msgSize, &req) < 0) { + taosMemoryFree(msg); + taosMemoryFree(pParam); + destroyCommonInfo(pCommon); + return terrno; + } + + SMsgSendInfo* sendInfo = taosMemoryCalloc(1, sizeof(SMsgSendInfo)); + if (sendInfo == NULL) { + taosMemoryFree(pParam); + taosMemoryFree(msg); + destroyCommonInfo(pCommon); + return terrno; + } + + sendInfo->msgInfo = (SDataBuf){.pData = msg, .len = msgSize, .handle = NULL}; + sendInfo->requestId = req.reqId; + sendInfo->requestObjRefId = 0; + sendInfo->param = pParam; + sendInfo->fp = tmqGetWalInfoCb; + sendInfo->msgType = TDMT_VND_TMQ_VG_WALINFO; + + int64_t transporterId = 0; + char offsetFormatBuf[80]; + tFormatOffset(offsetFormatBuf, tListLen(offsetFormatBuf), &pClientVg->offsetInfo.currentOffset); + + tscDebug("consumer:0x%" PRIx64 " %s retrieve wal info vgId:%d, epoch %d, req:%s, reqId:0x%" PRIx64, + tmq->consumerId, pTopic->topicName, pClientVg->vgId, tmq->epoch, offsetFormatBuf, req.reqId); + asyncSendMsgToServer(tmq->pTscObj->pAppInfo->pTransporter, &pClientVg->epSet, &transporterId, sendInfo); + } + + tsem_wait(&pCommon->rsp); + int32_t code = pCommon->code; + + terrno = code; + if (code != TSDB_CODE_SUCCESS) { + taosMemoryFree(*assignment); + *assignment = NULL; + *numOfAssignment = 0; + } else { + int32_t num = taosArrayGetSize(pCommon->pList); + for(int32_t i = 0; i < num; ++i) { + (*assignment)[i] = *(tmq_topic_assignment*)taosArrayGet(pCommon->pList, i); + } + *numOfAssignment = num; + } + + destroyCommonInfo(pCommon); + return code; + } else { + return TSDB_CODE_SUCCESS; + } +} + +int32_t tmq_offset_seek(tmq_t* tmq, const char* pTopicName, int32_t vgId, int64_t offset) { + if (tmq == NULL) { + tscError("invalid tmq handle, null"); + return TSDB_CODE_INVALID_PARA; + } + + int32_t accId = tmq->pTscObj->acctId; + char tname[128] = {0}; + sprintf(tname, "%d.%s", accId, pTopicName); + + SMqClientTopic* pTopic = getTopicByName(tmq, tname); + if (pTopic == NULL) { + tscError("consumer:0x%" PRIx64 " invalid topic name:%s", tmq->consumerId, pTopicName); + return TSDB_CODE_INVALID_PARA; + } + + SMqClientVg* pVg = NULL; + int32_t numOfVgs = taosArrayGetSize(pTopic->vgs); + for (int32_t i = 0; i < numOfVgs; ++i) { + SMqClientVg* pClientVg = taosArrayGet(pTopic->vgs, i); + if (pClientVg->vgId == vgId) { + pVg = pClientVg; + break; + } + } + + if (pVg == NULL) { + tscError("consumer:0x%" PRIx64 " invalid vgroup id:%d", tmq->consumerId, vgId); + return TSDB_CODE_INVALID_PARA; + } + + SVgOffsetInfo* pOffsetInfo = &pVg->offsetInfo; + + int32_t type = pOffsetInfo->currentOffset.type; + if (type != TMQ_OFFSET__LOG) { + tscError("consumer:0x%" PRIx64 " offset type:%d not wal version, seek not allowed", tmq->consumerId, type); + return TSDB_CODE_INVALID_PARA; + } + + if (offset < pOffsetInfo->walVerBegin || offset > pOffsetInfo->walVerEnd) { + tscError("consumer:0x%" PRIx64 " invalid seek params, offset:%" PRId64, tmq->consumerId, offset); + return TSDB_CODE_INVALID_PARA; + } + + // update the offset, and then commit to vnode + if (pOffsetInfo->currentOffset.type == TMQ_OFFSET__LOG) { + pOffsetInfo->currentOffset.version = offset; + pOffsetInfo->committedOffset.version = INT64_MIN; + } + + SMqRspObj rspObj = {.resType = RES_TYPE__TMQ, .vgId = pVg->vgId}; + tstrncpy(rspObj.topic, tname, tListLen(rspObj.topic)); + + tscDebug("consumer:0x%" PRIx64 " seek to %" PRId64 " on vgId:%d", tmq->consumerId, offset, pVg->vgId); + + SSyncCommitInfo* pInfo = taosMemoryMalloc(sizeof(SSyncCommitInfo)); + if (pInfo == NULL) { + tscError("consumer:0x%"PRIx64" failed to prepare seek operation", tmq->consumerId); + return TSDB_CODE_OUT_OF_MEMORY; + } + + tsem_init(&pInfo->sem, 0, 0); + pInfo->code = 0; + + asyncCommitOffset(tmq, &rspObj, TDMT_VND_TMQ_SEEK_TO_OFFSET, commitCallBackFn, pInfo); + + tsem_wait(&pInfo->sem); + int32_t code = pInfo->code; + + tsem_destroy(&pInfo->sem); + taosMemoryFree(pInfo); + + if (code != TSDB_CODE_SUCCESS) { + tscError("consumer:0x%" PRIx64 " failed to send seek to vgId:%d, code:%s", tmq->consumerId, pVg->vgId, + tstrerror(code)); + } + + return code; } \ No newline at end of file diff --git a/source/client/test/clientTests.cpp b/source/client/test/clientTests.cpp index 56f68e5972..105dfde139 100644 --- a/source/client/test/clientTests.cpp +++ b/source/client/test/clientTests.cpp @@ -30,6 +30,27 @@ #include "taos.h" namespace { + +void printSubResults(void* pRes, int32_t* totalRows) { + char buf[1024]; + + while (1) { + TAOS_ROW row = taos_fetch_row(pRes); + if (row == NULL) { + break; + } + + TAOS_FIELD* fields = taos_fetch_fields(pRes); + int32_t numOfFields = taos_field_count(pRes); + int32_t precision = taos_result_precision(pRes); + taos_print_row(buf, row, fields, numOfFields); + *totalRows += 1; + printf("precision: %d, row content: %s\n", precision, buf); + } + +// taos_free_result(pRes); +} + void showDB(TAOS* pConn) { TAOS_RES* pRes = taos_query(pConn, "show databases"); TAOS_ROW pRow = NULL; @@ -112,7 +133,7 @@ void createNewTable(TAOS* pConn, int32_t index) { } taos_free_result(pRes); - for (int32_t i = 0; i < 100; i += 20) { + for (int32_t i = 0; i < 10000; i += 20) { char sql[1024] = {0}; sprintf(sql, "insert into tu%d values(now+%da, %d)(now+%da, %d)(now+%da, %d)(now+%da, %d)" @@ -803,7 +824,7 @@ TEST(clientCase, projection_query_tables) { } taos_free_result(pRes); - for (int32_t i = 0; i < 10000; ++i) { + for (int32_t i = 0; i < 1; ++i) { printf("create table :%d\n", i); createNewTable(pConn, i); } @@ -990,7 +1011,7 @@ TEST(clientCase, sub_db_test) { tmq_conf_set(conf, "td.connect.user", "root"); tmq_conf_set(conf, "td.connect.pass", "taosdata"); tmq_conf_set(conf, "auto.offset.reset", "earliest"); - tmq_conf_set(conf, "experimental.snapshot.enable", "true"); + tmq_conf_set(conf, "experimental.snapshot.enable", "false"); tmq_conf_set(conf, "msg.with.table.name", "true"); tmq_conf_set_auto_commit_cb(conf, tmq_commit_cb_print, NULL); @@ -1000,7 +1021,7 @@ TEST(clientCase, sub_db_test) { // 创建订阅 topics 列表 tmq_list_t* topicList = tmq_list_new(); tmq_list_append(topicList, "topic_t1"); - tmq_list_append(topicList, "topic_s2"); +// tmq_list_append(topicList, "topic_s2"); // 启动订阅 tmq_subscribe(tmq, topicList); @@ -1059,6 +1080,7 @@ TEST(clientCase, sub_tb_test) { ASSERT_NE(pConn, nullptr); tmq_conf_t* conf = tmq_conf_new(); +<<<<<<< HEAD int32_t ts = taosGetTimestampMs()%INT32_MAX; char consumerGroupid[128] = {0}; @@ -1067,6 +1089,11 @@ TEST(clientCase, sub_tb_test) { tmq_conf_set(conf, "enable.auto.commit", "true"); tmq_conf_set(conf, "auto.commit.interval.ms", "2000"); tmq_conf_set(conf, "group.id", consumerGroupid); +======= + tmq_conf_set(conf, "enable.auto.commit", "false"); + tmq_conf_set(conf, "auto.commit.interval.ms", "1000"); + tmq_conf_set(conf, "group.id", "cgrpName1024"); +>>>>>>> enh/3.0 tmq_conf_set(conf, "td.connect.user", "root"); tmq_conf_set(conf, "td.connect.pass", "taosdata"); tmq_conf_set(conf, "auto.offset.reset", "earliest"); @@ -1083,7 +1110,6 @@ TEST(clientCase, sub_tb_test) { // 启动订阅 tmq_subscribe(tmq, topicList); - tmq_list_destroy(topicList); TAOS_FIELD* fields = NULL; @@ -1095,11 +1121,27 @@ TEST(clientCase, sub_tb_test) { int32_t count = 0; + tmq_topic_assignment* pAssign = NULL; + int32_t numOfAssign = 0; + + int32_t code = tmq_get_topic_assignment(tmq, "topic_t1", &pAssign, &numOfAssign); + if (code != 0) { + printf("error occurs:%s\n", tmq_err2str(code)); + tmq_consumer_close(tmq); + taos_close(pConn); + fprintf(stderr, "%d msg consumed, include %d rows\n", msgCnt, totalRows); + return; + } + while (1) { TAOS_RES* pRes = tmq_consumer_poll(tmq, timeout); +<<<<<<< HEAD if (pRes) { char buf[128]; +======= + if (pRes != NULL) { +>>>>>>> enh/3.0 const char* topicName = tmq_get_topic_name(pRes); // const char* dbName = tmq_get_db_name(pRes); // int32_t vgroupId = tmq_get_vgroup_id(pRes); @@ -1108,6 +1150,7 @@ TEST(clientCase, sub_tb_test) { // printf("db: %s\n", dbName); // printf("vgroup id: %d\n", vgroupId); +<<<<<<< HEAD while (1) { TAOS_ROW row = taos_fetch_row(pRes); if (row == NULL) { @@ -1124,9 +1167,20 @@ TEST(clientCase, sub_tb_test) { } taos_free_result(pRes); +======= + printSubResults(pRes, &totalRows); +>>>>>>> enh/3.0 } else { - break; +// tmq_offset_seek(tmq, "topic_t1", pAssign[0].vgroupHandle, pAssign[0].begin); +// break; } + + tmq_commit_sync(tmq, pRes); + if (pRes != NULL) { + taos_free_result(pRes); + } + + tmq_offset_seek(tmq, "topic_t1", pAssign[0].vgId, pAssign[0].begin); } tmq_consumer_close(tmq); diff --git a/source/common/src/tdataformat.c b/source/common/src/tdataformat.c index b18bd882ae..846ca44b83 100644 --- a/source/common/src/tdataformat.c +++ b/source/common/src/tdataformat.c @@ -500,7 +500,7 @@ int32_t tRowGet(SRow *pRow, STSchema *pTSchema, int32_t iCol, SColVal *pColVal) break; default: ASSERTS(0, "invalid row format"); - return TSDB_CODE_IVLD_DATA_FMT; + return TSDB_CODE_INVALID_DATA_FMT; } if (bv == BIT_FLG_NONE) { @@ -755,7 +755,7 @@ SColVal *tRowIterNext(SRowIter *pIter) { } if (pIter->pRow->flag == HAS_NULL) { - pIter->cv = COL_VAL_NULL(pTColumn->type, pTColumn->colId); + pIter->cv = COL_VAL_NULL(pTColumn->colId, pTColumn->type); goto _exit; } @@ -938,7 +938,7 @@ static int32_t tRowTupleUpsertColData(SRow *pRow, STSchema *pTSchema, SColData * break; default: ASSERTS(0, "Invalid row flag"); - return TSDB_CODE_IVLD_DATA_FMT; + return TSDB_CODE_INVALID_DATA_FMT; } while (pColData) { @@ -963,7 +963,7 @@ static int32_t tRowTupleUpsertColData(SRow *pRow, STSchema *pTSchema, SColData * break; default: ASSERTS(0, "Invalid row flag"); - return TSDB_CODE_IVLD_DATA_FMT; + return TSDB_CODE_INVALID_DATA_FMT; } if (bv == BIT_FLG_NONE) { @@ -1054,7 +1054,7 @@ static int32_t tRowKVUpsertColData(SRow *pRow, STSchema *pTSchema, SColData *aCo pData = pv + ((uint32_t *)pKVIdx->idx)[iCol]; } else { ASSERTS(0, "Invalid KV row format"); - return TSDB_CODE_IVLD_DATA_FMT; + return TSDB_CODE_INVALID_DATA_FMT; } int16_t cid; @@ -2441,7 +2441,7 @@ _exit: int32_t tColDataAddValueByDataBlock(SColData *pColData, int8_t type, int32_t bytes, int32_t nRows, char *lengthOrbitmap, char *data) { int32_t code = 0; - if(data == NULL){ + if (data == NULL) { for (int32_t i = 0; i < nRows; ++i) { code = tColDataAppendValueImpl[pColData->flag][CV_FLAG_NONE](pColData, NULL, 0); } @@ -2455,8 +2455,9 @@ int32_t tColDataAddValueByDataBlock(SColData *pColData, int8_t type, int32_t byt code = tColDataAppendValueImpl[pColData->flag][CV_FLAG_NULL](pColData, NULL, 0); if (code) goto _exit; } else { - if(ASSERT(varDataTLen(data + offset) <= bytes)){ - uError("var data length invalid, varDataTLen(data + offset):%d <= bytes:%d", (int)varDataTLen(data + offset), bytes); + if (ASSERT(varDataTLen(data + offset) <= bytes)) { + uError("var data length invalid, varDataTLen(data + offset):%d <= bytes:%d", (int)varDataTLen(data + offset), + bytes); code = TSDB_CODE_INVALID_PARA; goto _exit; } diff --git a/source/common/src/tglobal.c b/source/common/src/tglobal.c index 81b9b47e5c..4748e98913 100644 --- a/source/common/src/tglobal.c +++ b/source/common/src/tglobal.c @@ -117,6 +117,10 @@ int32_t tsRedirectFactor = 2; int32_t tsRedirectMaxPeriod = 1000; int32_t tsMaxRetryWaitTime = 10000; bool tsUseAdapter = false; +int32_t tsSlowLogThreshold = 3; // seconds +int32_t tsSlowLogScope = SLOW_LOG_TYPE_ALL; + + /* @@ -347,6 +351,8 @@ static int32_t taosAddClientCfg(SConfig *pCfg) { if (cfgAddBool(pCfg, "useAdapter", tsUseAdapter, true) != 0) return -1; if (cfgAddBool(pCfg, "crashReporting", tsEnableCrashReport, true) != 0) return -1; if (cfgAddInt64(pCfg, "queryMaxConcurrentTables", tsQueryMaxConcurrentTables, INT64_MIN, INT64_MAX, 1) != 0) return -1; + if (cfgAddInt32(pCfg, "slowLogThreshold", tsSlowLogThreshold, 0, INT32_MAX, true) != 0) return -1; + if (cfgAddString(pCfg, "slowLogScope", "", true) != 0) return -1; tsNumOfRpcThreads = tsNumOfCores / 2; tsNumOfRpcThreads = TRANGE(tsNumOfRpcThreads, 2, TSDB_MAX_RPC_THREADS); @@ -696,6 +702,42 @@ static void taosSetServerLogCfg(SConfig *pCfg) { metaDebugFlag = cfgGetItem(pCfg, "metaDebugFlag")->i32; } +static int32_t taosSetSlowLogScope(char *pScope) { + if (NULL == pScope || 0 == strlen(pScope)) { + tsSlowLogScope = SLOW_LOG_TYPE_ALL; + return 0; + } + + if (0 == strcasecmp(pScope, "all")) { + tsSlowLogScope = SLOW_LOG_TYPE_ALL; + return 0; + } + + if (0 == strcasecmp(pScope, "query")) { + tsSlowLogScope = SLOW_LOG_TYPE_QUERY; + return 0; + } + + if (0 == strcasecmp(pScope, "insert")) { + tsSlowLogScope = SLOW_LOG_TYPE_INSERT; + return 0; + } + + if (0 == strcasecmp(pScope, "others")) { + tsSlowLogScope = SLOW_LOG_TYPE_OTHERS; + return 0; + } + + if (0 == strcasecmp(pScope, "none")) { + tsSlowLogScope = 0; + return 0; + } + + uError("Invalid slowLog scope value:%s", pScope); + terrno = TSDB_CODE_INVALID_CFG_VALUE; + return -1; +} + static int32_t taosSetClientCfg(SConfig *pCfg) { tstrncpy(tsLocalFqdn, cfgGetItem(pCfg, "fqdn")->str, TSDB_FQDN_LEN); tsServerPort = (uint16_t)cfgGetItem(pCfg, "serverPort")->i32; @@ -746,6 +788,10 @@ static int32_t taosSetClientCfg(SConfig *pCfg) { tsUseAdapter = cfgGetItem(pCfg, "useAdapter")->bval; tsEnableCrashReport = cfgGetItem(pCfg, "crashReporting")->bval; tsQueryMaxConcurrentTables = cfgGetItem(pCfg, "queryMaxConcurrentTables")->i64; + tsSlowLogThreshold = cfgGetItem(pCfg, "slowLogThreshold")->i32; + if (taosSetSlowLogScope(cfgGetItem(pCfg, "slowLogScope")->str)) { + return -1; + } tsMaxRetryWaitTime = cfgGetItem(pCfg, "maxRetryWaitTime")->i32; @@ -1162,6 +1208,12 @@ int32_t taosSetCfg(SConfig *pCfg, char *name) { sDebugFlag = cfgGetItem(pCfg, "sDebugFlag")->i32; } else if (strcasecmp("smaDebugFlag", name) == 0) { smaDebugFlag = cfgGetItem(pCfg, "smaDebugFlag")->i32; + } else if (strcasecmp("slowLogThreshold", name) == 0) { + tsSlowLogThreshold = cfgGetItem(pCfg, "slowLogThreshold")->i32; + } else if (strcasecmp("slowLogScope", name) == 0) { + if (taosSetSlowLogScope(cfgGetItem(pCfg, "slowLogScope")->str)) { + return -1; + } } break; } diff --git a/source/common/src/tmsg.c b/source/common/src/tmsg.c index 40fd55f0a6..bb6b11f301 100644 --- a/source/common/src/tmsg.c +++ b/source/common/src/tmsg.c @@ -2937,6 +2937,59 @@ void tFreeSUserAuthBatchRsp(SUserAuthBatchRsp *pRsp) { taosArrayDestroy(pRsp->pArray); } +int32_t tSerializeSUserPassBatchRsp(void *buf, int32_t bufLen, SUserPassBatchRsp *pRsp) { + SEncoder encoder = {0}; + tEncoderInit(&encoder, buf, bufLen); + + if (tStartEncode(&encoder) < 0) return -1; + + int32_t numOfBatch = taosArrayGetSize(pRsp->pArray); + if (tEncodeI32(&encoder, numOfBatch) < 0) return -1; + for (int32_t i = 0; i < numOfBatch; ++i) { + SGetUserPassRsp *pUserPassRsp = taosArrayGet(pRsp->pArray, i); + if (tEncodeCStr(&encoder, pUserPassRsp->user) < 0) return -1; + if (tEncodeI32(&encoder, pUserPassRsp->version) < 0) return -1; + } + tEndEncode(&encoder); + + int32_t tlen = encoder.pos; + tEncoderClear(&encoder); + return tlen; +} + +int32_t tDeserializeSUserPassBatchRsp(void *buf, int32_t bufLen, SUserPassBatchRsp *pRsp) { + SDecoder decoder = {0}; + tDecoderInit(&decoder, buf, bufLen); + + if (tStartDecode(&decoder) < 0) return -1; + + int32_t numOfBatch = taosArrayGetSize(pRsp->pArray); + if (tDecodeI32(&decoder, &numOfBatch) < 0) return -1; + + pRsp->pArray = taosArrayInit(numOfBatch, sizeof(SGetUserPassRsp)); + if (pRsp->pArray == NULL) { + terrno = TSDB_CODE_OUT_OF_MEMORY; + return -1; + } + + for (int32_t i = 0; i < numOfBatch; ++i) { + SGetUserPassRsp rsp = {0}; + if (tDecodeCStrTo(&decoder, rsp.user) < 0) return -1; + if (tDecodeI32(&decoder, &rsp.version) < 0) return -1; + taosArrayPush(pRsp->pArray, &rsp); + } + tEndDecode(&decoder); + + tDecoderClear(&decoder); + return 0; +} + +void tFreeSUserPassBatchRsp(SUserPassBatchRsp *pRsp) { + if(pRsp) { + taosArrayDestroy(pRsp->pArray); + } +} + int32_t tSerializeSDbCfgReq(void *buf, int32_t bufLen, SDbCfgReq *pReq) { SEncoder encoder = {0}; tEncoderInit(&encoder, buf, bufLen); @@ -3977,6 +4030,7 @@ int32_t tSerializeSConnectRsp(void *buf, int32_t bufLen, SConnectRsp *pRsp) { if (tEncodeI32(&encoder, pRsp->svrTimestamp) < 0) return -1; if (tEncodeCStr(&encoder, pRsp->sVer) < 0) return -1; if (tEncodeCStr(&encoder, pRsp->sDetailVer) < 0) return -1; + if (tEncodeI32(&encoder, pRsp->passVer) < 0) return -1; tEndEncode(&encoder); int32_t tlen = encoder.pos; @@ -4000,6 +4054,13 @@ int32_t tDeserializeSConnectRsp(void *buf, int32_t bufLen, SConnectRsp *pRsp) { if (tDecodeI32(&decoder, &pRsp->svrTimestamp) < 0) return -1; if (tDecodeCStrTo(&decoder, pRsp->sVer) < 0) return -1; if (tDecodeCStrTo(&decoder, pRsp->sDetailVer) < 0) return -1; + + if (!tDecodeIsEnd(&decoder)) { + if (tDecodeI32(&decoder, &pRsp->passVer) < 0) return -1; + } else { + pRsp->passVer = 0; + } + tEndDecode(&decoder); tDecoderClear(&decoder); @@ -4130,6 +4191,12 @@ int32_t tSerializeSCreateVnodeReq(void *buf, int32_t bufLen, SCreateVnodeReq *pR for (int32_t i = 0; i < 8; ++i) { if (tEncodeI64(&encoder, pReq->reserved[i]) < 0) return -1; } + if (tEncodeI8(&encoder, pReq->learnerReplica) < 0) return -1; + if (tEncodeI8(&encoder, pReq->learnerSelfIndex) < 0) return -1; + for (int32_t i = 0; i < TSDB_MAX_LEARNER_REPLICA; ++i) { + SReplica *pReplica = &pReq->learnerReplicas[i]; + if (tEncodeSReplica(&encoder, pReplica) < 0) return -1; + } tEndEncode(&encoder); @@ -4208,6 +4275,14 @@ int32_t tDeserializeSCreateVnodeReq(void *buf, int32_t bufLen, SCreateVnodeReq * for (int32_t i = 0; i < 8; ++i) { if (tDecodeI64(&decoder, &pReq->reserved[i]) < 0) return -1; } + if (!tDecodeIsEnd(&decoder)) { + if (tDecodeI8(&decoder, &pReq->learnerReplica) < 0) return -1; + if (tDecodeI8(&decoder, &pReq->learnerSelfIndex) < 0) return -1; + for (int32_t i = 0; i < TSDB_MAX_LEARNER_REPLICA; ++i) { + SReplica *pReplica = &pReq->learnerReplicas[i]; + if (tDecodeSReplica(&decoder, pReplica) < 0) return -1; + } + } tEndDecode(&decoder); tDecoderClear(&decoder); @@ -4434,6 +4509,12 @@ int32_t tSerializeSAlterVnodeReplicaReq(void *buf, int32_t bufLen, SAlterVnodeRe for (int32_t i = 0; i < 8; ++i) { if (tEncodeI64(&encoder, pReq->reserved[i]) < 0) return -1; } + if (tEncodeI8(&encoder, pReq->learnerSelfIndex) < 0) return -1; + if (tEncodeI8(&encoder, pReq->learnerReplica) < 0) return -1; + for (int32_t i = 0; i < TSDB_MAX_LEARNER_REPLICA; ++i) { + SReplica *pReplica = &pReq->learnerReplicas[i]; + if (tEncodeSReplica(&encoder, pReplica) < 0) return -1; + } tEndEncode(&encoder); int32_t tlen = encoder.pos; @@ -4457,7 +4538,15 @@ int32_t tDeserializeSAlterVnodeReplicaReq(void *buf, int32_t bufLen, SAlterVnode for (int32_t i = 0; i < 8; ++i) { if (tDecodeI64(&decoder, &pReq->reserved[i]) < 0) return -1; } - + if (!tDecodeIsEnd(&decoder)) { + if (tDecodeI8(&decoder, &pReq->learnerSelfIndex) < 0) return -1; + if (tDecodeI8(&decoder, &pReq->learnerReplica) < 0) return -1; + for (int32_t i = 0; i < TSDB_MAX_LEARNER_REPLICA; ++i) { + SReplica *pReplica = &pReq->learnerReplicas[i]; + if (tDecodeSReplica(&decoder, pReplica) < 0) return -1; + } + } + tEndDecode(&decoder); tDecoderClear(&decoder); return 0; @@ -4768,6 +4857,12 @@ int32_t tSerializeSDCreateMnodeReq(void *buf, int32_t bufLen, SDCreateMnodeReq * SReplica *pReplica = &pReq->replicas[i]; if (tEncodeSReplica(&encoder, pReplica) < 0) return -1; } + if (tEncodeI8(&encoder, pReq->learnerReplica) < 0) return -1; + for (int32_t i = 0; i < TSDB_MAX_LEARNER_REPLICA; ++i) { + SReplica *pReplica = &pReq->learnerReplicas[i]; + if (tEncodeSReplica(&encoder, pReplica) < 0) return -1; + } + if (tEncodeI64(&encoder, pReq->lastIndex) < 0) return -1; tEndEncode(&encoder); int32_t tlen = encoder.pos; @@ -4785,6 +4880,14 @@ int32_t tDeserializeSDCreateMnodeReq(void *buf, int32_t bufLen, SDCreateMnodeReq SReplica *pReplica = &pReq->replicas[i]; if (tDecodeSReplica(&decoder, pReplica) < 0) return -1; } + if (!tDecodeIsEnd(&decoder)) { + if (tDecodeI8(&decoder, &pReq->learnerReplica) < 0) return -1; + for (int32_t i = 0; i < TSDB_MAX_LEARNER_REPLICA; ++i) { + SReplica *pReplica = &pReq->learnerReplicas[i]; + if (tDecodeSReplica(&decoder, pReplica) < 0) return -1; + } + if (tDecodeI64(&decoder, &pReq->lastIndex) < 0) return -1; + } tEndDecode(&decoder); tDecoderClear(&decoder); @@ -6906,6 +7009,18 @@ int32_t tDecodeSTqOffset(SDecoder *pDecoder, STqOffset *pOffset) { return 0; } +int32_t tEncodeMqVgOffset(SEncoder* pEncoder, const SMqVgOffset* pOffset) { + if (tEncodeSTqOffset(pEncoder, &pOffset->offset) < 0) return -1; + if (tEncodeI64(pEncoder, pOffset->consumerId) < 0) return -1; + return 0; +} + +int32_t tDecodeMqVgOffset(SDecoder* pDecoder, SMqVgOffset* pOffset) { + if (tDecodeSTqOffset(pDecoder, &pOffset->offset) < 0) return -1; + if (tDecodeI64(pDecoder, &pOffset->consumerId) < 0) return -1; + return 0; +} + int32_t tEncodeSTqCheckInfo(SEncoder *pEncoder, const STqCheckInfo *pInfo) { if (tEncodeCStr(pEncoder, pInfo->topic) < 0) return -1; if (tEncodeI64(pEncoder, pInfo->ntbUid) < 0) return -1; @@ -6970,21 +7085,21 @@ int32_t tDecodeDeleteRes(SDecoder *pCoder, SDeleteRes *pRes) { return 0; } -int32_t tEncodeSMqMetaRsp(SEncoder *pEncoder, const SMqMetaRsp *pRsp) { +int32_t tEncodeMqMetaRsp(SEncoder *pEncoder, const SMqMetaRsp *pRsp) { if (tEncodeSTqOffsetVal(pEncoder, &pRsp->rspOffset) < 0) return -1; if (tEncodeI16(pEncoder, pRsp->resMsgType)) return -1; if (tEncodeBinary(pEncoder, pRsp->metaRsp, pRsp->metaRspLen)) return -1; return 0; } -int32_t tDecodeSMqMetaRsp(SDecoder *pDecoder, SMqMetaRsp *pRsp) { +int32_t tDecodeMqMetaRsp(SDecoder *pDecoder, SMqMetaRsp *pRsp) { if (tDecodeSTqOffsetVal(pDecoder, &pRsp->rspOffset) < 0) return -1; if (tDecodeI16(pDecoder, &pRsp->resMsgType) < 0) return -1; if (tDecodeBinaryAlloc(pDecoder, &pRsp->metaRsp, (uint64_t *)&pRsp->metaRspLen) < 0) return -1; return 0; } -int32_t tEncodeSMqDataRsp(SEncoder *pEncoder, const SMqDataRsp *pRsp) { +int32_t tEncodeMqDataRsp(SEncoder *pEncoder, const SMqDataRsp *pRsp) { if (tEncodeSTqOffsetVal(pEncoder, &pRsp->reqOffset) < 0) return -1; if (tEncodeSTqOffsetVal(pEncoder, &pRsp->rspOffset) < 0) return -1; if (tEncodeI32(pEncoder, pRsp->blockNum) < 0) return -1; @@ -7009,7 +7124,7 @@ int32_t tEncodeSMqDataRsp(SEncoder *pEncoder, const SMqDataRsp *pRsp) { return 0; } -int32_t tDecodeSMqDataRsp(SDecoder *pDecoder, SMqDataRsp *pRsp) { +int32_t tDecodeMqDataRsp(SDecoder *pDecoder, SMqDataRsp *pRsp) { if (tDecodeSTqOffsetVal(pDecoder, &pRsp->reqOffset) < 0) return -1; if (tDecodeSTqOffsetVal(pDecoder, &pRsp->rspOffset) < 0) return -1; if (tDecodeI32(pDecoder, &pRsp->blockNum) < 0) return -1; @@ -7054,7 +7169,7 @@ int32_t tDecodeSMqDataRsp(SDecoder *pDecoder, SMqDataRsp *pRsp) { return 0; } -void tDeleteSMqDataRsp(SMqDataRsp *pRsp) { +void tDeleteMqDataRsp(SMqDataRsp *pRsp) { pRsp->blockDataLen = taosArrayDestroy(pRsp->blockDataLen); taosArrayDestroyP(pRsp->blockData, (FDelete)taosMemoryFree); pRsp->blockData = NULL; @@ -7155,8 +7270,7 @@ int32_t tDecodeSTaosxRsp(SDecoder *pDecoder, STaosxRsp *pRsp) { } void tDeleteSTaosxRsp(STaosxRsp *pRsp) { - taosArrayDestroy(pRsp->blockDataLen); - pRsp->blockDataLen = NULL; + pRsp->blockDataLen = taosArrayDestroy(pRsp->blockDataLen); taosArrayDestroyP(pRsp->blockData, (FDelete)taosMemoryFree); pRsp->blockData = NULL; taosArrayDestroyP(pRsp->blockSchema, (FDelete)tDeleteSchemaWrapper); @@ -7164,8 +7278,7 @@ void tDeleteSTaosxRsp(STaosxRsp *pRsp) { taosArrayDestroyP(pRsp->blockTbName, (FDelete)taosMemoryFree); pRsp->blockTbName = NULL; - taosArrayDestroy(pRsp->createTableLen); - pRsp->createTableLen = NULL; + pRsp->createTableLen = taosArrayDestroy(pRsp->createTableLen); taosArrayDestroyP(pRsp->createTableReq, (FDelete)taosMemoryFree); pRsp->createTableReq = NULL; } @@ -7539,6 +7652,7 @@ void tDestroySSubmitRsp2(SSubmitRsp2 *pRsp, int32_t flag) { } } +<<<<<<< HEAD int32_t tSerializeSMPauseStreamReq(void *buf, int32_t bufLen, const SMPauseStreamReq *pReq) { SEncoder encoder = {0}; tEncoderInit(&encoder, buf, bufLen); @@ -7591,3 +7705,42 @@ int32_t tDeserializeSMResumeStreamReq(void *buf, int32_t bufLen, SMResumeStreamR return 0; } +======= +int32_t tEncodeMqSubTopicEp(void **buf, const SMqSubTopicEp *pTopicEp) { + int32_t tlen = 0; + tlen += taosEncodeString(buf, pTopicEp->topic); + tlen += taosEncodeString(buf, pTopicEp->db); + int32_t sz = taosArrayGetSize(pTopicEp->vgs); + tlen += taosEncodeFixedI32(buf, sz); + for (int32_t i = 0; i < sz; i++) { + SMqSubVgEp *pVgEp = (SMqSubVgEp *)taosArrayGet(pTopicEp->vgs, i); + tlen += tEncodeSMqSubVgEp(buf, pVgEp); + } + tlen += taosEncodeSSchemaWrapper(buf, &pTopicEp->schema); + return tlen; +} + +void *tDecodeMqSubTopicEp(void *buf, SMqSubTopicEp *pTopicEp) { + buf = taosDecodeStringTo(buf, pTopicEp->topic); + buf = taosDecodeStringTo(buf, pTopicEp->db); + int32_t sz; + buf = taosDecodeFixedI32(buf, &sz); + pTopicEp->vgs = taosArrayInit(sz, sizeof(SMqSubVgEp)); + if (pTopicEp->vgs == NULL) { + return NULL; + } + for (int32_t i = 0; i < sz; i++) { + SMqSubVgEp vgEp; + buf = tDecodeSMqSubVgEp(buf, &vgEp); + taosArrayPush(pTopicEp->vgs, &vgEp); + } + buf = taosDecodeSSchemaWrapper(buf, &pTopicEp->schema); + return buf; +} + +void tDeleteMqSubTopicEp(SMqSubTopicEp *pSubTopicEp) { + taosMemoryFreeClear(pSubTopicEp->schema.pSchema); + pSubTopicEp->schema.nCols = 0; + taosArrayDestroy(pSubTopicEp->vgs); +} +>>>>>>> enh/3.0 diff --git a/source/dnode/mgmt/mgmt_dnode/inc/dmInt.h b/source/dnode/mgmt/mgmt_dnode/inc/dmInt.h index 600a1f6829..6c40fa808f 100644 --- a/source/dnode/mgmt/mgmt_dnode/inc/dmInt.h +++ b/source/dnode/mgmt/mgmt_dnode/inc/dmInt.h @@ -32,6 +32,7 @@ typedef struct SDnodeMgmt { TdThread crashReportThread; SSingleWorker mgmtWorker; ProcessCreateNodeFp processCreateNodeFp; + ProcessAlterNodeTypeFp processAlterNodeTypeFp; ProcessDropNodeFp processDropNodeFp; SendMonitorReportFp sendMonitorReportFp; GetVnodeLoadsFp getVnodeLoadsFp; diff --git a/source/dnode/mgmt/mgmt_dnode/src/dmHandle.c b/source/dnode/mgmt/mgmt_dnode/src/dmHandle.c index 228f301aec..1998269896 100644 --- a/source/dnode/mgmt/mgmt_dnode/src/dmHandle.c +++ b/source/dnode/mgmt/mgmt_dnode/src/dmHandle.c @@ -352,6 +352,7 @@ SArray *dmGetMsgHandles() { if (dmSetMgmtHandle(pArray, TDMT_DND_CONFIG_DNODE, dmPutNodeMsgToMgmtQueue, 0) == NULL) goto _OVER; if (dmSetMgmtHandle(pArray, TDMT_DND_SERVER_STATUS, dmPutNodeMsgToMgmtQueue, 0) == NULL) goto _OVER; if (dmSetMgmtHandle(pArray, TDMT_DND_SYSTABLE_RETRIEVE, dmPutNodeMsgToMgmtQueue, 0) == NULL) goto _OVER; + if (dmSetMgmtHandle(pArray, TDMT_DND_ALTER_MNODE_TYPE, dmPutNodeMsgToMgmtQueue, 0) == NULL) goto _OVER; // Requests handled by MNODE if (dmSetMgmtHandle(pArray, TDMT_MND_GRANT, dmPutNodeMsgToMgmtQueue, 0) == NULL) goto _OVER; diff --git a/source/dnode/mgmt/mgmt_dnode/src/dmInt.c b/source/dnode/mgmt/mgmt_dnode/src/dmInt.c index 51df293ba7..09783a5ea9 100644 --- a/source/dnode/mgmt/mgmt_dnode/src/dmInt.c +++ b/source/dnode/mgmt/mgmt_dnode/src/dmInt.c @@ -48,6 +48,7 @@ static int32_t dmOpenMgmt(SMgmtInputOpt *pInput, SMgmtOutputOpt *pOutput) { pMgmt->path = pInput->path; pMgmt->name = pInput->name; pMgmt->processCreateNodeFp = pInput->processCreateNodeFp; + pMgmt->processAlterNodeTypeFp = pInput->processAlterNodeTypeFp; pMgmt->processDropNodeFp = pInput->processDropNodeFp; pMgmt->sendMonitorReportFp = pInput->sendMonitorReportFp; pMgmt->getVnodeLoadsFp = pInput->getVnodeLoadsFp; diff --git a/source/dnode/mgmt/mgmt_dnode/src/dmWorker.c b/source/dnode/mgmt/mgmt_dnode/src/dmWorker.c index 0d3b423771..06b6221940 100644 --- a/source/dnode/mgmt/mgmt_dnode/src/dmWorker.c +++ b/source/dnode/mgmt/mgmt_dnode/src/dmWorker.c @@ -227,6 +227,9 @@ static void dmProcessMgmtQueue(SQueueInfo *pInfo, SRpcMsg *pMsg) { case TDMT_DND_DROP_SNODE: code = (*pMgmt->processDropNodeFp)(SNODE, pMsg); break; + case TDMT_DND_ALTER_MNODE_TYPE: + code = (*pMgmt->processAlterNodeTypeFp)(MNODE, pMsg); + break; case TDMT_DND_SERVER_STATUS: code = dmProcessServerRunStatus(pMgmt, pMsg); break; diff --git a/source/dnode/mgmt/mgmt_mnode/src/mmFile.c b/source/dnode/mgmt/mgmt_mnode/src/mmFile.c index f06669a610..cb0849f4b9 100644 --- a/source/dnode/mgmt/mgmt_mnode/src/mmFile.c +++ b/source/dnode/mgmt/mgmt_mnode/src/mmFile.c @@ -24,12 +24,16 @@ static int32_t mmDecodeOption(SJson *pJson, SMnodeOpt *pOption) { if (code < 0) return -1; tjsonGetInt32ValueFromDouble(pJson, "selfIndex", pOption->selfIndex, code); if (code < 0) return 0; + tjsonGetInt32ValueFromDouble(pJson, "lastIndex", pOption->lastIndex, code); + if (code < 0) return 0; SJson *replicas = tjsonGetObjectItem(pJson, "replicas"); if (replicas == NULL) return 0; - pOption->numOfReplicas = tjsonGetArraySize(replicas); + pOption->numOfTotalReplicas = tjsonGetArraySize(replicas); - for (int32_t i = 0; i < pOption->numOfReplicas; ++i) { + pOption->numOfReplicas = 0; + + for (int32_t i = 0; i < pOption->numOfTotalReplicas; ++i) { SJson *replica = tjsonGetArrayItem(replicas, i); if (replica == NULL) return -1; @@ -40,6 +44,14 @@ static int32_t mmDecodeOption(SJson *pJson, SMnodeOpt *pOption) { if (code < 0) return -1; tjsonGetUInt16ValueFromDouble(replica, "port", pReplica->port, code); if (code < 0) return -1; + tjsonGetInt32ValueFromDouble(replica, "role", pOption->nodeRoles[i], code); + if (code < 0) return -1; + if(pOption->nodeRoles[i] == TAOS_SYNC_ROLE_VOTER){ + pOption->numOfReplicas++; + } + } + + for (int32_t i = 0; i < pOption->numOfTotalReplicas; ++i) { } return 0; @@ -112,14 +124,14 @@ _OVER: } static int32_t mmEncodeOption(SJson *pJson, const SMnodeOpt *pOption) { - if (pOption->deploy && pOption->numOfReplicas > 0) { + if (pOption->deploy && pOption->numOfTotalReplicas > 0) { if (tjsonAddDoubleToObject(pJson, "selfIndex", pOption->selfIndex) < 0) return -1; SJson *replicas = tjsonCreateArray(); if (replicas == NULL) return -1; if (tjsonAddItemToObject(pJson, "replicas", replicas) < 0) return -1; - for (int32_t i = 0; i < pOption->numOfReplicas; ++i) { + for (int32_t i = 0; i < pOption->numOfTotalReplicas; ++i) { SJson *replica = tjsonCreateObject(); if (replica == NULL) return -1; @@ -127,10 +139,13 @@ static int32_t mmEncodeOption(SJson *pJson, const SMnodeOpt *pOption) { if (tjsonAddDoubleToObject(replica, "id", pReplica->id) < 0) return -1; if (tjsonAddStringToObject(replica, "fqdn", pReplica->fqdn) < 0) return -1; if (tjsonAddDoubleToObject(replica, "port", pReplica->port) < 0) return -1; + if (tjsonAddDoubleToObject(replica, "role", pOption->nodeRoles[i]) < 0) return -1; if (tjsonAddItemToArray(replicas, replica) < 0) return -1; } } + if (tjsonAddDoubleToObject(pJson, "lastIndex", pOption->lastIndex) < 0) return -1; + if (tjsonAddDoubleToObject(pJson, "deployed", pOption->deploy) < 0) return -1; return 0; diff --git a/source/dnode/mgmt/mgmt_mnode/src/mmHandle.c b/source/dnode/mgmt/mgmt_mnode/src/mmHandle.c index 3bf748ea29..d9678c713d 100644 --- a/source/dnode/mgmt/mgmt_mnode/src/mmHandle.c +++ b/source/dnode/mgmt/mgmt_mnode/src/mmHandle.c @@ -33,12 +33,24 @@ int32_t mmProcessCreateReq(const SMgmtInputOpt *pInput, SRpcMsg *pMsg) { return -1; } - SMnodeOpt option = {.deploy = true, .numOfReplicas = createReq.replica, .selfIndex = -1}; + SMnodeOpt option = {.deploy = true, .numOfReplicas = createReq.replica, + .numOfTotalReplicas = createReq.replica + createReq.learnerReplica, + .selfIndex = -1, .lastIndex = createReq.lastIndex}; + memcpy(option.replicas, createReq.replicas, sizeof(createReq.replicas)); - for (int32_t i = 0; i < option.numOfReplicas; ++i) { + for (int32_t i = 0; i < createReq.replica; ++i) { if (createReq.replicas[i].id == pInput->pData->dnodeId) { option.selfIndex = i; } + option.nodeRoles[i] = TAOS_SYNC_ROLE_VOTER; + } + + memcpy(&(option.replicas[createReq.replica]), createReq.learnerReplicas, sizeof(createReq.learnerReplicas)); + for (int32_t i = 0; i < createReq.learnerReplica; ++i) { + if (createReq.learnerReplicas[i].id == pInput->pData->dnodeId) { + option.selfIndex = createReq.replica + i; + } + option.nodeRoles[createReq.replica + i] = TAOS_SYNC_ROLE_LEARNER; } if (option.selfIndex == -1) { @@ -92,6 +104,8 @@ SArray *mmGetMsgHandles() { if (dmSetMgmtHandle(pArray, TDMT_DND_CREATE_VNODE_RSP, mmPutMsgToWriteQueue, 0) == NULL) goto _OVER; if (dmSetMgmtHandle(pArray, TDMT_DND_DROP_VNODE_RSP, mmPutMsgToWriteQueue, 0) == NULL) goto _OVER; if (dmSetMgmtHandle(pArray, TDMT_DND_CONFIG_DNODE_RSP, mmPutMsgToReadQueue, 0) == NULL) goto _OVER; + if (dmSetMgmtHandle(pArray, TDMT_DND_ALTER_MNODE_TYPE_RSP, mmPutMsgToReadQueue, 0) == NULL) goto _OVER; + if (dmSetMgmtHandle(pArray, TDMT_DND_ALTER_VNODE_TYPE_RSP, mmPutMsgToReadQueue, 0) == NULL) goto _OVER; if (dmSetMgmtHandle(pArray, TDMT_MND_CONNECT, mmPutMsgToReadQueue, 0) == NULL) goto _OVER; if (dmSetMgmtHandle(pArray, TDMT_MND_CREATE_ACCT, mmPutMsgToWriteQueue, 0) == NULL) goto _OVER; diff --git a/source/dnode/mgmt/mgmt_mnode/src/mmInt.c b/source/dnode/mgmt/mgmt_mnode/src/mmInt.c index e7f71ad420..7840528db9 100644 --- a/source/dnode/mgmt/mgmt_mnode/src/mmInt.c +++ b/source/dnode/mgmt/mgmt_mnode/src/mmInt.c @@ -45,9 +45,11 @@ static void mmBuildOptionForDeploy(SMnodeMgmt *pMgmt, const SMgmtInputOpt *pInpu pOption->dnodeId = pMgmt->pData->dnodeId; pOption->selfIndex = 0; pOption->numOfReplicas = 1; + pOption->numOfTotalReplicas = 1; pOption->replicas[0].id = 1; pOption->replicas[0].port = tsServerPort; tstrncpy(pOption->replicas[0].fqdn, tsLocalFqdn, TSDB_FQDN_LEN); + pOption->lastIndex = SYNC_INDEX_INVALID; } static void mmBuildOptionForOpen(SMnodeMgmt *pMgmt, SMnodeOpt *pOption) { @@ -123,9 +125,10 @@ static int32_t mmOpen(SMgmtInputOpt *pInput, SMgmtOutputOpt *pOutput) { } tmsgReportStartup("mnode-worker", "initialized"); - if (option.numOfReplicas > 0) { + if (option.numOfTotalReplicas > 0) { option.deploy = true; option.numOfReplicas = 0; + option.numOfTotalReplicas = 0; if (mmWriteFile(pMgmt->path, &option) != 0) { dError("failed to write mnode file since %s", terrstr()); return -1; @@ -152,6 +155,14 @@ static void mmStop(SMnodeMgmt *pMgmt) { mndStop(pMgmt->pMnode); } +static int32_t mmSyncIsCatchUp(SMnodeMgmt *pMgmt) { + return mndIsCatchUp(pMgmt->pMnode); +} + +static ESyncRole mmSyncGetRole(SMnodeMgmt *pMgmt) { + return mndGetRole(pMgmt->pMnode); +} + SMgmtFunc mmGetMgmtFunc() { SMgmtFunc mgmtFunc = {0}; mgmtFunc.openFp = mmOpen; @@ -162,6 +173,8 @@ SMgmtFunc mmGetMgmtFunc() { mgmtFunc.dropFp = (NodeDropFp)mmProcessDropReq; mgmtFunc.requiredFp = mmRequire; mgmtFunc.getHandlesFp = mmGetMsgHandles; + mgmtFunc.isCatchUpFp = (NodeIsCatchUpFp)mmSyncIsCatchUp; + mgmtFunc.nodeRoleFp = (NodeRole)mmSyncGetRole; return mgmtFunc; } diff --git a/source/dnode/mgmt/mgmt_vnode/inc/vmInt.h b/source/dnode/mgmt/mgmt_vnode/inc/vmInt.h index 4374ae363c..83fb331dbd 100644 --- a/source/dnode/mgmt/mgmt_vnode/inc/vmInt.h +++ b/source/dnode/mgmt/mgmt_vnode/inc/vmInt.h @@ -90,6 +90,7 @@ int32_t vmProcessDropVnodeReq(SVnodeMgmt *pMgmt, SRpcMsg *pMsg); int32_t vmProcessAlterVnodeReplicaReq(SVnodeMgmt *pMgmt, SRpcMsg *pMsg); int32_t vmProcessDisableVnodeWriteReq(SVnodeMgmt *pMgmt, SRpcMsg *pMsg); int32_t vmProcessAlterHashRangeReq(SVnodeMgmt *pMgmt, SRpcMsg *pMsg); +int32_t vmProcessAlterVnodeTypeReq(SVnodeMgmt *pMgmt, SRpcMsg *pMsg); // vmFile.c int32_t vmGetVnodeListFromFile(SVnodeMgmt *pMgmt, SWrapperCfg **ppCfgs, int32_t *numOfVnodes); diff --git a/source/dnode/mgmt/mgmt_vnode/src/vmHandle.c b/source/dnode/mgmt/mgmt_vnode/src/vmHandle.c index 5edc183c84..282e694f56 100644 --- a/source/dnode/mgmt/mgmt_vnode/src/vmHandle.c +++ b/source/dnode/mgmt/mgmt_vnode/src/vmHandle.c @@ -131,15 +131,34 @@ static void vmGenerateVnodeCfg(SCreateVnodeReq *pCreate, SVnodeCfg *pCfg) { pCfg->tsdbPageSize = pCreate->tsdbPageSize * 1024; pCfg->standby = 0; - pCfg->syncCfg.myIndex = pCreate->selfIndex; - pCfg->syncCfg.replicaNum = pCreate->replica; + pCfg->syncCfg.replicaNum = 0; + pCfg->syncCfg.totalReplicaNum = 0; + memset(&pCfg->syncCfg.nodeInfo, 0, sizeof(pCfg->syncCfg.nodeInfo)); for (int32_t i = 0; i < pCreate->replica; ++i) { SNodeInfo *pNode = &pCfg->syncCfg.nodeInfo[i]; - pNode->nodeId = pCreate->replicas[i].id; - pNode->nodePort = pCreate->replicas[i].port; - tstrncpy(pNode->nodeFqdn, pCreate->replicas[i].fqdn, TSDB_FQDN_LEN); + pNode->nodeId = pCreate->replicas[pCfg->syncCfg.replicaNum].id; + pNode->nodePort = pCreate->replicas[pCfg->syncCfg.replicaNum].port; + pNode->nodeRole = TAOS_SYNC_ROLE_VOTER; + tstrncpy(pNode->nodeFqdn, pCreate->replicas[pCfg->syncCfg.replicaNum].fqdn, TSDB_FQDN_LEN); tmsgUpdateDnodeInfo(&pNode->nodeId, &pNode->clusterId, pNode->nodeFqdn, &pNode->nodePort); + pCfg->syncCfg.replicaNum++; + } + if(pCreate->selfIndex != -1){ + pCfg->syncCfg.myIndex = pCreate->selfIndex; + } + for (int32_t i = pCfg->syncCfg.replicaNum; i < pCreate->replica + pCreate->learnerReplica; ++i) { + SNodeInfo *pNode = &pCfg->syncCfg.nodeInfo[i]; + pNode->nodeId = pCreate->learnerReplicas[pCfg->syncCfg.totalReplicaNum].id; + pNode->nodePort = pCreate->learnerReplicas[pCfg->syncCfg.totalReplicaNum].port; + pNode->nodeRole = TAOS_SYNC_ROLE_LEARNER; + tstrncpy(pNode->nodeFqdn, pCreate->learnerReplicas[pCfg->syncCfg.totalReplicaNum].fqdn, TSDB_FQDN_LEN); + tmsgUpdateDnodeInfo(&pNode->nodeId, &pNode->clusterId, pNode->nodeFqdn, &pNode->nodePort); + pCfg->syncCfg.totalReplicaNum++; + } + pCfg->syncCfg.totalReplicaNum += pCfg->syncCfg.replicaNum; + if(pCreate->learnerSelfIndex != -1){ + pCfg->syncCfg.myIndex = pCreate->replica + pCreate->learnerSelfIndex; } } @@ -182,23 +201,40 @@ int32_t vmProcessCreateVnodeReq(SVnodeMgmt *pMgmt, SRpcMsg *pMsg) { return -1; } - dInfo("vgId:%d, start to create vnode, page:%d pageSize:%d buffer:%d szPage:%d szBuf:%" PRIu64 + if(req.learnerReplica == 0) + { + req.learnerSelfIndex = -1; + } + + dInfo("vgId:%d, vnode management handle msgType:%s, start to create vnode, page:%d pageSize:%d buffer:%d szPage:%d szBuf:%" PRIu64 ", cacheLast:%d cacheLastSize:%d sstTrigger:%d tsdbPageSize:%d %d dbname:%s dbId:%" PRId64 ", days:%d keep0:%d keep1:%d keep2:%d tsma:%d precision:%d compression:%d minRows:%d maxRows:%d" ", wal fsync:%d level:%d retentionPeriod:%d retentionSize:%" PRId64 " rollPeriod:%d segSize:%" PRId64 - ", hash method:%d begin:%u end:%u prefix:%d surfix:%d replica:%d selfIndex:%d strict:%d", - req.vgId, req.pages, req.pageSize, req.buffer, req.pageSize * 1024, (uint64_t)req.buffer * 1024 * 1024, + ", hash method:%d begin:%u end:%u prefix:%d surfix:%d replica:%d selfIndex:%d " + "learnerReplica:%d learnerSelfIndex:%d strict:%d", + req.vgId, TMSG_INFO(pMsg->msgType), req.pages, req.pageSize, req.buffer, req.pageSize * 1024, + (uint64_t)req.buffer * 1024 * 1024, req.cacheLast, req.cacheLastSize, req.sstTrigger, req.tsdbPageSize, req.tsdbPageSize * 1024, req.db, req.dbUid, req.daysPerFile, req.daysToKeep0, req.daysToKeep1, req.daysToKeep2, req.isTsma, req.precision, req.compression, req.minRows, req.maxRows, req.walFsyncPeriod, req.walLevel, req.walRetentionPeriod, req.walRetentionSize, req.walRollPeriod, req.walSegmentSize, req.hashMethod, req.hashBegin, req.hashEnd, req.hashPrefix, - req.hashSuffix, req.replica, req.selfIndex, req.strict); + req.hashSuffix, req.replica, req.selfIndex, req.learnerReplica, req.learnerSelfIndex, req.strict); for (int32_t i = 0; i < req.replica; ++i) { dInfo("vgId:%d, replica:%d ep:%s:%u dnode:%d", req.vgId, i, req.replicas[i].fqdn, req.replicas[i].port, req.replicas[i].id); } + for (int32_t i = 0; i < req.learnerReplica; ++i) { + dInfo("vgId:%d, learnerReplica:%d ep:%s:%u dnode:%d", req.vgId, i, req.learnerReplicas[i].fqdn, + req.learnerReplicas[i].port, req.replicas[i].id); + } - SReplica *pReplica = &req.replicas[req.selfIndex]; + SReplica *pReplica = NULL; + if(req.selfIndex != -1){ + pReplica = &req.replicas[req.selfIndex]; + } + else{ + pReplica = &req.learnerReplicas[req.learnerSelfIndex]; + } if (pReplica->id != pMgmt->pData->dnodeId || pReplica->port != tsServerPort || strcmp(pReplica->fqdn, tsLocalFqdn) != 0) { terrno = TSDB_CODE_INVALID_MSG; @@ -275,7 +311,8 @@ _OVER: vnodeClose(pImpl); vnodeDestroy(path, pMgmt->pTfs); } else { - dInfo("vgId:%d, vnode is created", req.vgId); + dInfo("vgId:%d, vnode management handle msgType:%s, end to create vnode, vnode is created", + req.vgId, TMSG_INFO(pMsg->msgType)); } tFreeSCreateVnodeReq(&req); @@ -283,6 +320,122 @@ _OVER: return code; } +int32_t vmProcessAlterVnodeTypeReq(SVnodeMgmt *pMgmt, SRpcMsg *pMsg) { + SAlterVnodeTypeReq req = {0}; + if (tDeserializeSAlterVnodeReplicaReq(pMsg->pCont, pMsg->contLen, &req) != 0) { + terrno = TSDB_CODE_INVALID_MSG; + return -1; + } + + if(req.learnerReplicas == 0){ + req.learnerSelfIndex = -1; + } + + dInfo("vgId:%d, vnode management handle msgType:%s, start to process alter-node-type-request", + req.vgId, TMSG_INFO(pMsg->msgType)); + + SVnodeObj *pVnode = vmAcquireVnode(pMgmt, req.vgId); + if (pVnode == NULL) { + dError("vgId:%d, failed to alter vnode type since %s", req.vgId, terrstr()); + terrno = TSDB_CODE_VND_NOT_EXIST; + return -1; + } + + ESyncRole role = vnodeGetRole(pVnode->pImpl); + dInfo("vgId:%d, checking node role:%d", req.vgId, role); + if(role == TAOS_SYNC_ROLE_VOTER){ + terrno = TSDB_CODE_VND_ALREADY_IS_VOTER; + vmReleaseVnode(pMgmt, pVnode); + return -1; + } + + dInfo("vgId:%d, checking node catch up", req.vgId); + if(vnodeIsCatchUp(pVnode->pImpl) != 1){ + terrno = TSDB_CODE_VND_NOT_CATCH_UP; + vmReleaseVnode(pMgmt, pVnode); + return -1; + } + + dInfo("node:%s, catched up leader, continue to process alter-node-type-request", pMgmt->name); + + int32_t vgId = req.vgId; + dInfo("vgId:%d, start to alter vnode type replica:%d selfIndex:%d strict:%d", vgId, req.replica, req.selfIndex, + req.strict); + for (int32_t i = 0; i < req.replica; ++i) { + SReplica *pReplica = &req.replicas[i]; + dInfo("vgId:%d, replica:%d ep:%s:%u dnode:%d", vgId, i, pReplica->fqdn, pReplica->port, pReplica->id); + } + for (int32_t i = 0; i < req.learnerReplica; ++i) { + SReplica *pReplica = &req.learnerReplicas[i]; + dInfo("vgId:%d, learnerReplicas:%d ep:%s:%u dnode:%d", vgId, i, pReplica->fqdn, pReplica->port, pReplica->id); + } + + if (req.replica <= 0 || + (req.selfIndex < 0 && req.learnerSelfIndex <0)|| + req.selfIndex >= req.replica || req.learnerSelfIndex >= req.learnerReplica) { + terrno = TSDB_CODE_INVALID_MSG; + dError("vgId:%d, failed to alter replica since invalid msg", vgId); + vmReleaseVnode(pMgmt, pVnode); + return -1; + } + + SReplica *pReplica = NULL; + if(req.selfIndex > 0){ + pReplica = &req.replicas[req.selfIndex]; + } + else{ + pReplica = &req.learnerReplicas[req.learnerSelfIndex]; + } + + if (pReplica->id != pMgmt->pData->dnodeId || pReplica->port != tsServerPort || + strcmp(pReplica->fqdn, tsLocalFqdn) != 0) { + terrno = TSDB_CODE_INVALID_MSG; + dError("vgId:%d, dnodeId:%d ep:%s:%u not matched with local dnode", vgId, pReplica->id, pReplica->fqdn, + pReplica->port); + vmReleaseVnode(pMgmt, pVnode); + return -1; + } + + dInfo("vgId:%d, start to close vnode", vgId); + SWrapperCfg wrapperCfg = { + .dropped = pVnode->dropped, + .vgId = pVnode->vgId, + .vgVersion = pVnode->vgVersion, + }; + tstrncpy(wrapperCfg.path, pVnode->path, sizeof(wrapperCfg.path)); + vmCloseVnode(pMgmt, pVnode, false); + + char path[TSDB_FILENAME_LEN] = {0}; + snprintf(path, TSDB_FILENAME_LEN, "vnode%svnode%d", TD_DIRSEP, vgId); + + dInfo("vgId:%d, start to alter vnode replica at %s", vgId, path); + if (vnodeAlterReplica(path, &req, pMgmt->pTfs) < 0) { + dError("vgId:%d, failed to alter vnode at %s since %s", vgId, path, terrstr()); + return -1; + } + + dInfo("vgId:%d, begin to open vnode", vgId); + SVnode *pImpl = vnodeOpen(path, pMgmt->pTfs, pMgmt->msgCb); + if (pImpl == NULL) { + dError("vgId:%d, failed to open vnode at %s since %s", vgId, path, terrstr()); + return -1; + } + + if (vmOpenVnode(pMgmt, &wrapperCfg, pImpl) != 0) { + dError("vgId:%d, failed to open vnode mgmt since %s", vgId, terrstr()); + return -1; + } + + if (vnodeStart(pImpl) != 0) { + dError("vgId:%d, failed to start sync since %s", vgId, terrstr()); + return -1; + } + + dInfo("vgId:%d, vnode management handle msgType:%s, end to process alter-node-type-request, vnode config is altered", + req.vgId, TMSG_INFO(pMsg->msgType)); + return 0; +} + int32_t vmProcessDisableVnodeWriteReq(SVnodeMgmt *pMgmt, SRpcMsg *pMsg) { SDisableVnodeWriteReq req = {0}; if (tDeserializeSDisableVnodeWriteReq(pMsg->pCont, pMsg->contLen, &req) != 0) { @@ -377,21 +530,40 @@ int32_t vmProcessAlterVnodeReplicaReq(SVnodeMgmt *pMgmt, SRpcMsg *pMsg) { return -1; } + if(alterReq.learnerReplica == 0){ + alterReq.learnerSelfIndex = -1; + } + int32_t vgId = alterReq.vgId; - dInfo("vgId:%d, start to alter vnode replica:%d selfIndex:%d strict:%d", vgId, alterReq.replica, alterReq.selfIndex, - alterReq.strict); + dInfo("vgId:%d,vnode management handle msgType:%s, start to alter vnode replica:%d selfIndex:%d leanerReplica:%d " + "learnerSelfIndex:%d strict:%d", + vgId, TMSG_INFO(pMsg->msgType), alterReq.replica, alterReq.selfIndex, alterReq.learnerReplica, + alterReq.learnerSelfIndex, alterReq.strict); for (int32_t i = 0; i < alterReq.replica; ++i) { SReplica *pReplica = &alterReq.replicas[i]; dInfo("vgId:%d, replica:%d ep:%s:%u dnode:%d", vgId, i, pReplica->fqdn, pReplica->port, pReplica->port); } - - if (alterReq.replica <= 0 || alterReq.selfIndex < 0 || alterReq.selfIndex >= alterReq.replica) { + for (int32_t i = 0; i < alterReq.learnerReplica; ++i) { + SReplica *pReplica = &alterReq.learnerReplicas[i]; + dInfo("vgId:%d, learnerReplicas:%d ep:%s:%u dnode:%d", vgId, i, pReplica->fqdn, pReplica->port, pReplica->port); + } + + if (alterReq.replica <= 0 || + (alterReq.selfIndex < 0 && alterReq.learnerSelfIndex <0)|| + alterReq.selfIndex >= alterReq.replica || alterReq.learnerSelfIndex >= alterReq.learnerReplica) { terrno = TSDB_CODE_INVALID_MSG; dError("vgId:%d, failed to alter replica since invalid msg", vgId); return -1; } - SReplica *pReplica = &alterReq.replicas[alterReq.selfIndex]; + SReplica *pReplica = NULL; + if(alterReq.selfIndex != -1){ + pReplica = &alterReq.replicas[alterReq.selfIndex]; + } + else{ + pReplica = &alterReq.learnerReplicas[alterReq.learnerSelfIndex]; + } + if (pReplica->id != pMgmt->pData->dnodeId || pReplica->port != tsServerPort || strcmp(pReplica->fqdn, tsLocalFqdn) != 0) { terrno = TSDB_CODE_INVALID_MSG; @@ -425,7 +597,7 @@ int32_t vmProcessAlterVnodeReplicaReq(SVnodeMgmt *pMgmt, SRpcMsg *pMsg) { return -1; } - dInfo("vgId:%d, close vnode", vgId); + dInfo("vgId:%d, begin to open vnode", vgId); SVnode *pImpl = vnodeOpen(path, pMgmt->pTfs, pMgmt->msgCb); if (pImpl == NULL) { dError("vgId:%d, failed to open vnode at %s since %s", vgId, path, terrstr()); @@ -442,7 +614,10 @@ int32_t vmProcessAlterVnodeReplicaReq(SVnodeMgmt *pMgmt, SRpcMsg *pMsg) { return -1; } - dInfo("vgId:%d, vnode config is altered", vgId); + dInfo("vgId:%d, vnode management handle msgType:%s, end to alter vnode replica:%d selfIndex:%d leanerReplica:%d " + "learnerSelfIndex:%d strict:%d", + vgId, TMSG_INFO(pMsg->msgType), alterReq.replica, alterReq.selfIndex, alterReq.learnerReplica, + alterReq.learnerSelfIndex, alterReq.strict); return 0; } @@ -517,9 +692,15 @@ SArray *vmGetMsgHandles() { if (dmSetMgmtHandle(pArray, TDMT_VND_TMQ_SUBSCRIBE, vmPutMsgToWriteQueue, 0) == NULL) goto _OVER; if (dmSetMgmtHandle(pArray, TDMT_VND_TMQ_DELETE_SUB, vmPutMsgToWriteQueue, 0) == NULL) goto _OVER; if (dmSetMgmtHandle(pArray, TDMT_VND_TMQ_COMMIT_OFFSET, vmPutMsgToWriteQueue, 0) == NULL) goto _OVER; + if (dmSetMgmtHandle(pArray, TDMT_VND_TMQ_SEEK_TO_OFFSET, vmPutMsgToWriteQueue, 0) == NULL) goto _OVER; if (dmSetMgmtHandle(pArray, TDMT_VND_TMQ_ADD_CHECKINFO, vmPutMsgToWriteQueue, 0) == NULL) goto _OVER; if (dmSetMgmtHandle(pArray, TDMT_VND_TMQ_DEL_CHECKINFO, vmPutMsgToWriteQueue, 0) == NULL) goto _OVER; +<<<<<<< HEAD if (dmSetMgmtHandle(pArray, TDMT_VND_TMQ_CONSUME, vmPutMsgToQueryQueue, 0) == NULL) goto _OVER; +======= + if (dmSetMgmtHandle(pArray, TDMT_VND_TMQ_CONSUME, vmPutMsgToFetchQueue, 0) == NULL) goto _OVER; + if (dmSetMgmtHandle(pArray, TDMT_VND_TMQ_VG_WALINFO, vmPutMsgToFetchQueue, 0) == NULL) goto _OVER; +>>>>>>> enh/3.0 if (dmSetMgmtHandle(pArray, TDMT_VND_DELETE, vmPutMsgToWriteQueue, 0) == NULL) goto _OVER; if (dmSetMgmtHandle(pArray, TDMT_VND_BATCH_DEL, vmPutMsgToWriteQueue, 0) == NULL) goto _OVER; if (dmSetMgmtHandle(pArray, TDMT_VND_COMMIT, vmPutMsgToWriteQueue, 0) == NULL) goto _OVER; @@ -550,6 +731,7 @@ SArray *vmGetMsgHandles() { if (dmSetMgmtHandle(pArray, TDMT_VND_TRIM, vmPutMsgToWriteQueue, 0) == NULL) goto _OVER; if (dmSetMgmtHandle(pArray, TDMT_DND_CREATE_VNODE, vmPutMsgToMgmtQueue, 0) == NULL) goto _OVER; if (dmSetMgmtHandle(pArray, TDMT_DND_DROP_VNODE, vmPutMsgToMgmtQueue, 0) == NULL) goto _OVER; + if (dmSetMgmtHandle(pArray, TDMT_DND_ALTER_VNODE_TYPE, vmPutMsgToMgmtQueue, 0) == NULL) goto _OVER; if (dmSetMgmtHandle(pArray, TDMT_SYNC_TIMEOUT_ELECTION, vmPutMsgToSyncQueue, 0) == NULL) goto _OVER; if (dmSetMgmtHandle(pArray, TDMT_SYNC_CLIENT_REQUEST, vmPutMsgToSyncQueue, 0) == NULL) goto _OVER; diff --git a/source/dnode/mgmt/mgmt_vnode/src/vmWorker.c b/source/dnode/mgmt/mgmt_vnode/src/vmWorker.c index a318b9886e..a5f53c8703 100644 --- a/source/dnode/mgmt/mgmt_vnode/src/vmWorker.c +++ b/source/dnode/mgmt/mgmt_vnode/src/vmWorker.c @@ -49,6 +49,9 @@ static void vmProcessMgmtQueue(SQueueInfo *pInfo, SRpcMsg *pMsg) { case TDMT_VND_ALTER_HASHRANGE: code = vmProcessAlterHashRangeReq(pMgmt, pMsg); break; + case TDMT_DND_ALTER_VNODE_TYPE: + code = vmProcessAlterVnodeTypeReq(pMgmt, pMsg); + break; default: terrno = TSDB_CODE_MSG_NOT_PROCESSED; dGError("msg:%p, not processed in vnode-mgmt queue", pMsg); @@ -57,7 +60,7 @@ static void vmProcessMgmtQueue(SQueueInfo *pInfo, SRpcMsg *pMsg) { if (IsReq(pMsg)) { if (code != 0) { if (terrno != 0) code = terrno; - dGError("msg:%p, failed to process since %s, type:%s", pMsg, terrstr(code), TMSG_INFO(pMsg->msgType)); + dGError("msg:%p, failed to process since %s, type:%s", pMsg, tstrerror(code), TMSG_INFO(pMsg->msgType)); } vmSendRsp(pMsg, code); } diff --git a/source/dnode/mgmt/node_mgmt/src/dmEnv.c b/source/dnode/mgmt/node_mgmt/src/dmEnv.c index acf96ad397..3459af1a3a 100644 --- a/source/dnode/mgmt/node_mgmt/src/dmEnv.c +++ b/source/dnode/mgmt/node_mgmt/src/dmEnv.c @@ -169,6 +169,8 @@ static int32_t dmProcessCreateNodeReq(EDndNodeType ntype, SRpcMsg *pMsg) { return -1; } + dInfo("start to process create-node-request"); + pWrapper = &pDnode->wrappers[ntype]; if (taosMkDir(pWrapper->path) != 0) { dmReleaseWrapper(pWrapper); @@ -198,6 +200,74 @@ static int32_t dmProcessCreateNodeReq(EDndNodeType ntype, SRpcMsg *pMsg) { return code; } +static int32_t dmProcessAlterNodeTypeReq(EDndNodeType ntype, SRpcMsg *pMsg) { + SDnode *pDnode = dmInstance(); + + SMgmtWrapper *pWrapper = dmAcquireWrapper(pDnode, ntype); + if (pWrapper == NULL) { + dError("fail to process alter node type since node not exist"); + return -1; + } + dmReleaseWrapper(pWrapper); + + dInfo("node:%s, start to process alter-node-type-request", pWrapper->name); + + pWrapper = &pDnode->wrappers[ntype]; + + if(pWrapper->func.nodeRoleFp != NULL){ + ESyncRole role = (*pWrapper->func.nodeRoleFp)(pWrapper->pMgmt); + dInfo("node:%s, checking node role:%d", pWrapper->name, role); + if(role == TAOS_SYNC_ROLE_VOTER){ + terrno = TSDB_CODE_MNODE_ALREADY_IS_VOTER; + return -1; + } + } + + if(pWrapper->func.isCatchUpFp != NULL){ + dInfo("node:%s, checking node catch up", pWrapper->name); + if((*pWrapper->func.isCatchUpFp)(pWrapper->pMgmt) != 1){ + terrno = TSDB_CODE_MNODE_NOT_CATCH_UP; + return -1; + } + } + + dInfo("node:%s, catched up leader, continue to process alter-node-type-request", pWrapper->name); + + taosThreadMutexLock(&pDnode->mutex); + + dInfo("node:%s, stopping node", pWrapper->name); + dmStopNode(pWrapper); + dInfo("node:%s, closing node", pWrapper->name); + dmCloseNode(pWrapper); + + pWrapper = &pDnode->wrappers[ntype]; + if (taosMkDir(pWrapper->path) != 0) { + dmReleaseWrapper(pWrapper); + terrno = TAOS_SYSTEM_ERROR(errno); + dError("failed to create dir:%s since %s", pWrapper->path, terrstr()); + return -1; + } + + SMgmtInputOpt input = dmBuildMgmtInputOpt(pWrapper); + + dInfo("node:%s, start to create", pWrapper->name); + int32_t code = (*pWrapper->func.createFp)(&input, pMsg); + if (code != 0) { + dError("node:%s, failed to create since %s", pWrapper->name, terrstr()); + } else { + dInfo("node:%s, has been created", pWrapper->name); + code = dmOpenNode(pWrapper); + if (code == 0) { + code = dmStartNode(pWrapper); + } + pWrapper->deployed = true; + pWrapper->required = true; + } + + taosThreadMutexUnlock(&pDnode->mutex); + return code; +} + static int32_t dmProcessDropNodeReq(EDndNodeType ntype, SRpcMsg *pMsg) { SDnode *pDnode = dmInstance(); @@ -251,6 +321,7 @@ SMgmtInputOpt dmBuildMgmtInputOpt(SMgmtWrapper *pWrapper) { .name = pWrapper->name, .pData = &pWrapper->pDnode->data, .processCreateNodeFp = dmProcessCreateNodeReq, + .processAlterNodeTypeFp = dmProcessAlterNodeTypeReq, .processDropNodeFp = dmProcessDropNodeReq, .sendMonitorReportFp = dmSendMonitorReport, .getVnodeLoadsFp = dmGetVnodeLoads, diff --git a/source/dnode/mgmt/node_util/inc/dmUtil.h b/source/dnode/mgmt/node_util/inc/dmUtil.h index cfdea40477..98ef8cd95b 100644 --- a/source/dnode/mgmt/node_util/inc/dmUtil.h +++ b/source/dnode/mgmt/node_util/inc/dmUtil.h @@ -89,6 +89,7 @@ typedef void (*SendMonitorReportFp)(); typedef void (*GetVnodeLoadsFp)(SMonVloadInfo *pInfo); typedef void (*GetMnodeLoadsFp)(SMonMloadInfo *pInfo); typedef void (*GetQnodeLoadsFp)(SQnodeLoad *pInfo); +typedef int32_t (*ProcessAlterNodeTypeFp)(EDndNodeType ntype, SRpcMsg *pMsg); typedef struct { int32_t dnodeId; @@ -112,6 +113,7 @@ typedef struct { SDnodeData *pData; SMsgCb msgCb; ProcessCreateNodeFp processCreateNodeFp; + ProcessAlterNodeTypeFp processAlterNodeTypeFp; ProcessDropNodeFp processDropNodeFp; SendMonitorReportFp sendMonitorReportFp; GetVnodeLoadsFp getVnodeLoadsFp; @@ -132,6 +134,8 @@ typedef int32_t (*NodeCreateFp)(const SMgmtInputOpt *pInput, SRpcMsg *pMsg); typedef int32_t (*NodeDropFp)(const SMgmtInputOpt *pInput, SRpcMsg *pMsg); typedef int32_t (*NodeRequireFp)(const SMgmtInputOpt *pInput, bool *required); typedef SArray *(*NodeGetHandlesFp)(); // array of SMgmtHandle +typedef bool (*NodeIsCatchUpFp)(void *pMgmt); +typedef bool (*NodeRole)(void *pMgmt); typedef struct { NodeOpenFp openFp; @@ -142,6 +146,8 @@ typedef struct { NodeDropFp dropFp; NodeRequireFp requiredFp; NodeGetHandlesFp getHandlesFp; + NodeIsCatchUpFp isCatchUpFp; + NodeRole nodeRoleFp; } SMgmtFunc; typedef struct { diff --git a/source/dnode/mnode/impl/inc/mndDef.h b/source/dnode/mnode/impl/inc/mndDef.h index 2579ff5231..4cbab214d2 100644 --- a/source/dnode/mnode/impl/inc/mndDef.h +++ b/source/dnode/mnode/impl/inc/mndDef.h @@ -118,7 +118,7 @@ typedef enum { } ETrnPolicy; typedef enum { - TRN_EXEC_PRARLLEL = 0, + TRN_EXEC_PARALLEL = 0, TRN_EXEC_SERIAL = 1, } ETrnExec; @@ -177,6 +177,7 @@ typedef struct { SArray* pRpcArray; SRWLatch lockRpcArray; int64_t mTraceId; + TdThreadMutex mutex; } STrans; typedef struct { @@ -215,6 +216,8 @@ typedef struct { bool syncRestore; int64_t stateStartTime; SDnodeObj* pDnode; + int32_t role; + SyncIndex lastIndex; } SMnodeObj; typedef struct { @@ -278,6 +281,7 @@ typedef struct { int8_t reserve; int32_t acctId; int32_t authVersion; + int32_t passVersion; SHashObj* readDbs; SHashObj* writeDbs; SHashObj* topics; @@ -341,6 +345,7 @@ typedef struct { ESyncState syncState; bool syncRestore; bool syncCanRead; + ESyncRole nodeRole; } SVnodeGid; typedef struct { @@ -361,7 +366,7 @@ typedef struct { int8_t compact; int8_t isTsma; int8_t replica; - SVnodeGid vnodeGid[TSDB_MAX_REPLICA]; + SVnodeGid vnodeGid[TSDB_MAX_REPLICA + TSDB_MAX_LEARNER_REPLICA]; void* pTsma; int32_t numOfCachedTables; } SVgObj; diff --git a/source/dnode/mnode/impl/inc/mndInt.h b/source/dnode/mnode/impl/inc/mndInt.h index ffb2443808..ec83f8a7e6 100644 --- a/source/dnode/mnode/impl/inc/mndInt.h +++ b/source/dnode/mnode/impl/inc/mndInt.h @@ -92,8 +92,11 @@ typedef struct { int64_t transSeq; TdThreadMutex lock; int8_t selfIndex; + int8_t numOfTotalReplicas; int8_t numOfReplicas; - SReplica replicas[TSDB_MAX_REPLICA]; + SReplica replicas[TSDB_MAX_REPLICA + TSDB_MAX_LEARNER_REPLICA]; + ESyncRole nodeRoles[TSDB_MAX_REPLICA + TSDB_MAX_LEARNER_REPLICA]; + SyncIndex lastIndex; } SSyncMgmt; typedef struct { diff --git a/source/dnode/mnode/impl/inc/mndTrans.h b/source/dnode/mnode/impl/inc/mndTrans.h index 057e3efbbc..03434573c4 100644 --- a/source/dnode/mnode/impl/inc/mndTrans.h +++ b/source/dnode/mnode/impl/inc/mndTrans.h @@ -76,6 +76,7 @@ void mndTransSetRpcRsp(STrans *pTrans, void *pCont, int32_t contLen); void mndTransSetCb(STrans *pTrans, ETrnFunc startFunc, ETrnFunc stopFunc, void *param, int32_t paramLen); void mndTransSetDbName(STrans *pTrans, const char *dbname, const char *stbname); void mndTransSetSerial(STrans *pTrans); +void mndTransSetParallel(STrans *pTrans); void mndTransSetOper(STrans *pTrans, EOperType oper); int32_t mndTrancCheckConflict(SMnode *pMnode, STrans *pTrans); diff --git a/source/dnode/mnode/impl/inc/mndUser.h b/source/dnode/mnode/impl/inc/mndUser.h index 95d15f6e5a..aa7f97f087 100644 --- a/source/dnode/mnode/impl/inc/mndUser.h +++ b/source/dnode/mnode/impl/inc/mndUser.h @@ -35,6 +35,8 @@ SHashObj *mndDupTableHash(SHashObj *pOld); SHashObj *mndDupTopicHash(SHashObj *pOld); int32_t mndValidateUserAuthInfo(SMnode *pMnode, SUserAuthVersion *pUsers, int32_t numOfUses, void **ppRsp, int32_t *pRspLen); +int32_t mndValidateUserPassInfo(SMnode *pMnode, SUserPassVersion *pUsers, int32_t numOfUses, void **ppRsp, + int32_t *pRspLen); int32_t mndUserRemoveDb(SMnode *pMnode, STrans *pTrans, char *db); int32_t mndUserRemoveTopic(SMnode *pMnode, STrans *pTrans, char *topic); diff --git a/source/dnode/mnode/impl/src/mndConsumer.c b/source/dnode/mnode/impl/src/mndConsumer.c index d144747e81..da4a42ccce 100644 --- a/source/dnode/mnode/impl/src/mndConsumer.c +++ b/source/dnode/mnode/impl/src/mndConsumer.c @@ -564,9 +564,14 @@ static int32_t mndProcessAskEpReq(SRpcMsg *pMsg) { return -1; } - ((SMqRspHead *)buf)->mqMsgType = TMQ_MSG_TYPE__EP_RSP; - ((SMqRspHead *)buf)->epoch = serverEpoch; - ((SMqRspHead *)buf)->consumerId = pConsumer->consumerId; + SMqRspHead* pHead = buf; + + pHead->mqMsgType = TMQ_MSG_TYPE__EP_RSP; + pHead->epoch = serverEpoch; + pHead->consumerId = pConsumer->consumerId; + pHead->walsver = 0; + pHead->walever = 0; + void *abuf = POINTER_SHIFT(buf, sizeof(SMqRspHead)); tEncodeSMqAskEpRsp(&abuf, &rsp); diff --git a/source/dnode/mnode/impl/src/mndDump.c b/source/dnode/mnode/impl/src/mndDump.c index 44a7d49fff..a991bddda8 100644 --- a/source/dnode/mnode/impl/src/mndDump.c +++ b/source/dnode/mnode/impl/src/mndDump.c @@ -421,6 +421,7 @@ void dumpUser(SSdb *pSdb, SJson *json) { tjsonAddStringToObject(item, "updateTime", i642str(pObj->updateTime)); tjsonAddStringToObject(item, "superUser", i642str(pObj->superUser)); tjsonAddStringToObject(item, "authVersion", i642str(pObj->authVersion)); + tjsonAddStringToObject(item, "passVersion", i642str(pObj->passVersion)); tjsonAddStringToObject(item, "numOfReadDbs", i642str(taosHashGetSize(pObj->readDbs))); tjsonAddStringToObject(item, "numOfWriteDbs", i642str(taosHashGetSize(pObj->writeDbs))); sdbRelease(pSdb, pObj); diff --git a/source/dnode/mnode/impl/src/mndFunc.c b/source/dnode/mnode/impl/src/mndFunc.c index 51a231daf4..af9bedd220 100644 --- a/source/dnode/mnode/impl/src/mndFunc.c +++ b/source/dnode/mnode/impl/src/mndFunc.c @@ -556,7 +556,7 @@ RETRIEVE_FUNC_OVER: return code; } -static void *mnodeGenTypeStr(char *buf, int32_t buflen, uint8_t type, int16_t len) { +static void *mnodeGenTypeStr(char *buf, int32_t buflen, uint8_t type, int32_t len) { char *msg = "unknown"; if (type >= sizeof(tDataTypes) / sizeof(tDataTypes[0])) { return msg; diff --git a/source/dnode/mnode/impl/src/mndMain.c b/source/dnode/mnode/impl/src/mndMain.c index 92ff550895..381b1e64ed 100644 --- a/source/dnode/mnode/impl/src/mndMain.c +++ b/source/dnode/mnode/impl/src/mndMain.c @@ -487,7 +487,10 @@ static void mndSetOptions(SMnode *pMnode, const SMnodeOpt *pOption) { pMnode->selfDnodeId = pOption->dnodeId; pMnode->syncMgmt.selfIndex = pOption->selfIndex; pMnode->syncMgmt.numOfReplicas = pOption->numOfReplicas; + pMnode->syncMgmt.numOfTotalReplicas = pOption->numOfTotalReplicas; + pMnode->syncMgmt.lastIndex = pOption->lastIndex; memcpy(pMnode->syncMgmt.replicas, pOption->replicas, sizeof(pOption->replicas)); + memcpy(pMnode->syncMgmt.nodeRoles, pOption->nodeRoles, sizeof(pOption->nodeRoles)); } SMnode *mndOpen(const char *path, const SMnodeOpt *pOption) { @@ -578,6 +581,16 @@ int32_t mndStart(SMnode *pMnode) { return mndInitTimer(pMnode); } +int32_t mndIsCatchUp(SMnode *pMnode) { + int64_t rid = pMnode->syncMgmt.sync; + return syncIsCatchUp(rid); +} + +ESyncRole mndGetRole(SMnode *pMnode){ + int64_t rid = pMnode->syncMgmt.sync; + return syncGetRole(rid); +} + void mndStop(SMnode *pMnode) { mndSetStop(pMnode); mndSyncStop(pMnode); diff --git a/source/dnode/mnode/impl/src/mndMnode.c b/source/dnode/mnode/impl/src/mndMnode.c index 50fab447e3..5e3476859a 100644 --- a/source/dnode/mnode/impl/src/mndMnode.c +++ b/source/dnode/mnode/impl/src/mndMnode.c @@ -23,7 +23,7 @@ #include "mndTrans.h" #include "tmisce.h" -#define MNODE_VER_NUMBER 1 +#define MNODE_VER_NUMBER 2 #define MNODE_RESERVE_SIZE 64 static int32_t mndCreateDefaultMnode(SMnode *pMnode); @@ -53,6 +53,7 @@ int32_t mndInitMnode(SMnode *pMnode) { mndSetMsgHandle(pMnode, TDMT_MND_CREATE_MNODE, mndProcessCreateMnodeReq); mndSetMsgHandle(pMnode, TDMT_DND_CREATE_MNODE_RSP, mndTransProcessRsp); + mndSetMsgHandle(pMnode, TDMT_DND_ALTER_MNODE_TYPE_RSP, mndTransProcessRsp); mndSetMsgHandle(pMnode, TDMT_MND_ALTER_MNODE, mndProcessAlterMnodeReq); mndSetMsgHandle(pMnode, TDMT_MND_ALTER_MNODE_RSP, mndTransProcessRsp); mndSetMsgHandle(pMnode, TDMT_MND_DROP_MNODE, mndProcessDropMnodeReq); @@ -126,6 +127,8 @@ static SSdbRaw *mndMnodeActionEncode(SMnodeObj *pObj) { SDB_SET_INT32(pRaw, dataPos, pObj->id, _OVER) SDB_SET_INT64(pRaw, dataPos, pObj->createdTime, _OVER) SDB_SET_INT64(pRaw, dataPos, pObj->updateTime, _OVER) + SDB_SET_INT32(pRaw, dataPos, pObj->role, _OVER) + SDB_SET_INT64(pRaw, dataPos, pObj->lastIndex, _OVER) SDB_SET_RESERVE(pRaw, dataPos, MNODE_RESERVE_SIZE, _OVER) terrno = 0; @@ -149,7 +152,7 @@ static SSdbRow *mndMnodeActionDecode(SSdbRaw *pRaw) { int8_t sver = 0; if (sdbGetRawSoftVer(pRaw, &sver) != 0) return NULL; - if (sver != MNODE_VER_NUMBER) { + if (sver != 1 && sver != 2) { terrno = TSDB_CODE_SDB_INVALID_DATA_VER; goto _OVER; } @@ -164,6 +167,10 @@ static SSdbRow *mndMnodeActionDecode(SSdbRaw *pRaw) { SDB_GET_INT32(pRaw, dataPos, &pObj->id, _OVER) SDB_GET_INT64(pRaw, dataPos, &pObj->createdTime, _OVER) SDB_GET_INT64(pRaw, dataPos, &pObj->updateTime, _OVER) + if(sver >=2){ + SDB_GET_INT32(pRaw, dataPos, &pObj->role, _OVER) + SDB_GET_INT64(pRaw, dataPos, &pObj->lastIndex, _OVER) + } SDB_GET_RESERVE(pRaw, dataPos, MNODE_RESERVE_SIZE, _OVER) terrno = 0; @@ -204,7 +211,9 @@ static int32_t mndMnodeActionDelete(SSdb *pSdb, SMnodeObj *pObj) { static int32_t mndMnodeActionUpdate(SSdb *pSdb, SMnodeObj *pOld, SMnodeObj *pNew) { mTrace("mnode:%d, perform update action, old row:%p new row:%p", pOld->id, pOld, pNew); + pOld->role = pNew->role; pOld->updateTime = pNew->updateTime; + pOld->lastIndex = pNew->lastIndex; mndReloadSyncConfig(pSdb->pMnode); return 0; @@ -302,6 +311,28 @@ static int32_t mndBuildCreateMnodeRedoAction(STrans *pTrans, SDCreateMnodeReq *p return 0; } +static int32_t mndBuildAlterMnodeTypeRedoAction(STrans *pTrans, + SDAlterMnodeTypeReq *pAlterMnodeTypeReq, SEpSet *pAlterMnodeTypeEpSet) { + int32_t contLen = tSerializeSDCreateMnodeReq(NULL, 0, pAlterMnodeTypeReq); + void *pReq = taosMemoryMalloc(contLen); + tSerializeSDCreateMnodeReq(pReq, contLen, pAlterMnodeTypeReq); + + STransAction action = { + .epSet = *pAlterMnodeTypeEpSet, + .pCont = pReq, + .contLen = contLen, + .msgType = TDMT_DND_ALTER_MNODE_TYPE, + .retryCode = TSDB_CODE_MNODE_NOT_CATCH_UP, + .acceptableCode = TSDB_CODE_MNODE_ALREADY_IS_VOTER, + }; + + if (mndTransAppendRedoAction(pTrans, &action) != 0) { + taosMemoryFree(pReq); + return -1; + } + return 0; +} + static int32_t mndBuildAlterMnodeRedoAction(STrans *pTrans, SDCreateMnodeReq *pAlterReq, SEpSet *pAlterEpSet) { int32_t contLen = tSerializeSDCreateMnodeReq(NULL, 0, pAlterReq); void *pReq = taosMemoryMalloc(contLen); @@ -347,6 +378,7 @@ static int32_t mndSetCreateMnodeRedoActions(SMnode *pMnode, STrans *pTrans, SDno SSdb *pSdb = pMnode->pSdb; void *pIter = NULL; int32_t numOfReplicas = 0; + int32_t numOfLearnerReplicas = 0; SDCreateMnodeReq createReq = {0}; SEpSet createEpset = {0}; @@ -355,18 +387,29 @@ static int32_t mndSetCreateMnodeRedoActions(SMnode *pMnode, STrans *pTrans, SDno pIter = sdbFetch(pSdb, SDB_MNODE, pIter, (void **)&pMObj); if (pIter == NULL) break; - createReq.replicas[numOfReplicas].id = pMObj->id; - createReq.replicas[numOfReplicas].port = pMObj->pDnode->port; - memcpy(createReq.replicas[numOfReplicas].fqdn, pMObj->pDnode->fqdn, TSDB_FQDN_LEN); + if(pMObj->role == TAOS_SYNC_ROLE_VOTER){ + createReq.replicas[numOfReplicas].id = pMObj->id; + createReq.replicas[numOfReplicas].port = pMObj->pDnode->port; + memcpy(createReq.replicas[numOfReplicas].fqdn, pMObj->pDnode->fqdn, TSDB_FQDN_LEN); + numOfReplicas++; + } + else{ + createReq.learnerReplicas[numOfLearnerReplicas].id = pMObj->id; + createReq.learnerReplicas[numOfLearnerReplicas].port = pMObj->pDnode->port; + memcpy(createReq.learnerReplicas[numOfLearnerReplicas].fqdn, pMObj->pDnode->fqdn, TSDB_FQDN_LEN); + numOfLearnerReplicas++; + } - numOfReplicas++; sdbRelease(pSdb, pMObj); } - createReq.replica = numOfReplicas + 1; - createReq.replicas[numOfReplicas].id = pDnode->id; - createReq.replicas[numOfReplicas].port = pDnode->port; - memcpy(createReq.replicas[numOfReplicas].fqdn, pDnode->fqdn, TSDB_FQDN_LEN); + createReq.replica = numOfReplicas; + createReq.learnerReplica = numOfLearnerReplicas + 1; + createReq.learnerReplicas[numOfLearnerReplicas].id = pDnode->id; + createReq.learnerReplicas[numOfLearnerReplicas].port = pDnode->port; + memcpy(createReq.learnerReplicas[numOfLearnerReplicas].fqdn, pDnode->fqdn, TSDB_FQDN_LEN); + + createReq.lastIndex = pObj->lastIndex; createEpset.inUse = 0; createEpset.numOfEps = 1; @@ -378,23 +421,78 @@ static int32_t mndSetCreateMnodeRedoActions(SMnode *pMnode, STrans *pTrans, SDno return 0; } +static int32_t mndSetAlterMnodeTypeRedoActions(SMnode *pMnode, STrans *pTrans, SDnodeObj *pDnode, SMnodeObj *pObj) { + SSdb *pSdb = pMnode->pSdb; + void *pIter = NULL; + SDAlterMnodeTypeReq alterReq = {0}; + SEpSet createEpset = {0}; + + while (1) { + SMnodeObj *pMObj = NULL; + pIter = sdbFetch(pSdb, SDB_MNODE, pIter, (void **)&pMObj); + if (pIter == NULL) break; + + if(pMObj->role == TAOS_SYNC_ROLE_VOTER){ + alterReq.replicas[alterReq.replica].id = pMObj->id; + alterReq.replicas[alterReq.replica].port = pMObj->pDnode->port; + memcpy(alterReq.replicas[alterReq.replica].fqdn, pMObj->pDnode->fqdn, TSDB_FQDN_LEN); + alterReq.replica++; + } + else{ + alterReq.learnerReplicas[alterReq.learnerReplica].id = pMObj->id; + alterReq.learnerReplicas[alterReq.learnerReplica].port = pMObj->pDnode->port; + memcpy(alterReq.learnerReplicas[alterReq.learnerReplica].fqdn, pMObj->pDnode->fqdn, TSDB_FQDN_LEN); + alterReq.learnerReplica++; + } + + sdbRelease(pSdb, pMObj); + } + + alterReq.replicas[alterReq.replica].id = pDnode->id; + alterReq.replicas[alterReq.replica].port = pDnode->port; + memcpy(alterReq.replicas[alterReq.replica].fqdn, pDnode->fqdn, TSDB_FQDN_LEN); + alterReq.replica++; + + alterReq.lastIndex = pObj->lastIndex; + + createEpset.inUse = 0; + createEpset.numOfEps = 1; + createEpset.eps[0].port = pDnode->port; + memcpy(createEpset.eps[0].fqdn, pDnode->fqdn, TSDB_FQDN_LEN); + + if (mndBuildAlterMnodeTypeRedoAction(pTrans, &alterReq, &createEpset) != 0) return -1; + + return 0; +} + static int32_t mndCreateMnode(SMnode *pMnode, SRpcMsg *pReq, SDnodeObj *pDnode, SMCreateMnodeReq *pCreate) { int32_t code = -1; - SMnodeObj mnodeObj = {0}; - mnodeObj.id = pDnode->id; - mnodeObj.createdTime = taosGetTimestampMs(); - mnodeObj.updateTime = mnodeObj.createdTime; - STrans *pTrans = mndTransCreate(pMnode, TRN_POLICY_RETRY, TRN_CONFLICT_GLOBAL, pReq, "create-mnode"); if (pTrans == NULL) goto _OVER; mndTransSetSerial(pTrans); mInfo("trans:%d, used to create mnode:%d", pTrans->id, pCreate->dnodeId); if (mndTrancCheckConflict(pMnode, pTrans) != 0) goto _OVER; + SMnodeObj mnodeObj = {0}; + mnodeObj.id = pDnode->id; + mnodeObj.createdTime = taosGetTimestampMs(); + mnodeObj.updateTime = mnodeObj.createdTime; + mnodeObj.role = TAOS_SYNC_ROLE_LEARNER; + mnodeObj.lastIndex = pMnode->applied; + if (mndSetCreateMnodeRedoActions(pMnode, pTrans, pDnode, &mnodeObj) != 0) goto _OVER; if (mndSetCreateMnodeRedoLogs(pMnode, pTrans, &mnodeObj) != 0) goto _OVER; - if (mndSetCreateMnodeCommitLogs(pMnode, pTrans, &mnodeObj) != 0) goto _OVER; + + SMnodeObj mnodeLeaderObj = {0}; + mnodeLeaderObj.id = pDnode->id; + mnodeLeaderObj.createdTime = taosGetTimestampMs(); + mnodeLeaderObj.updateTime = mnodeLeaderObj.createdTime; + mnodeLeaderObj.role = TAOS_SYNC_ROLE_VOTER; + mnodeLeaderObj.lastIndex = pMnode->applied + 1; + + if (mndSetAlterMnodeTypeRedoActions(pMnode, pTrans, pDnode, &mnodeLeaderObj) != 0) goto _OVER; + if (mndSetCreateMnodeCommitLogs(pMnode, pTrans, &mnodeLeaderObj) != 0) goto _OVER; if (mndTransPrepare(pMnode, pTrans) != 0) goto _OVER; code = 0; @@ -514,6 +612,7 @@ static int32_t mndSetDropMnodeRedoActions(SMnode *pMnode, STrans *pTrans, SDnode int32_t mndSetDropMnodeInfoToTrans(SMnode *pMnode, STrans *pTrans, SMnodeObj *pObj, bool force) { if (pObj == NULL) return 0; + pObj->lastIndex = pMnode->applied; if (mndSetDropMnodeRedoActions(pMnode, pTrans, pObj->pDnode, pObj, force) != 0) return -1; if (mndSetDropMnodeCommitLogs(pMnode, pTrans, pObj) != 0) return -1; return 0; @@ -730,7 +829,8 @@ static void mndReloadSyncConfig(SMnode *pMnode) { void *pIter = NULL; int32_t updatingMnodes = 0; int32_t readyMnodes = 0; - SSyncCfg cfg = {.myIndex = -1}; + SSyncCfg cfg = {.myIndex = -1, .lastIndex = 0,}; + SyncIndex maxIndex = 0; while (1) { pIter = sdbFetchAll(pSdb, SDB_MNODE, pIter, (void **)&pObj, &objStatus, false); @@ -745,26 +845,41 @@ static void mndReloadSyncConfig(SMnode *pMnode) { } if (objStatus == SDB_STATUS_READY || objStatus == SDB_STATUS_CREATING) { - SNodeInfo *pNode = &cfg.nodeInfo[cfg.replicaNum]; + SNodeInfo *pNode = &cfg.nodeInfo[cfg.totalReplicaNum]; pNode->nodeId = pObj->pDnode->id; pNode->clusterId = mndGetClusterId(pMnode); pNode->nodePort = pObj->pDnode->port; + pNode->nodeRole = pObj->role; tstrncpy(pNode->nodeFqdn, pObj->pDnode->fqdn, TSDB_FQDN_LEN); (void)tmsgUpdateDnodeInfo(&pNode->nodeId, &pNode->clusterId, pNode->nodeFqdn, &pNode->nodePort); mInfo("vgId:1, ep:%s:%u dnode:%d", pNode->nodeFqdn, pNode->nodePort, pNode->nodeId); if (pObj->pDnode->id == pMnode->selfDnodeId) { - cfg.myIndex = cfg.replicaNum; + cfg.myIndex = cfg.totalReplicaNum; + } + if(pNode->nodeRole == TAOS_SYNC_ROLE_VOTER){ + cfg.replicaNum++; + } + cfg.totalReplicaNum++; + if(pObj->lastIndex > cfg.lastIndex){ + cfg.lastIndex = pObj->lastIndex; } - cfg.replicaNum++; } + if (objStatus == SDB_STATUS_DROPPING) { + if(pObj->lastIndex > cfg.lastIndex){ + cfg.lastIndex = pObj->lastIndex; + } + } + + mInfo("vgId:1, mnode:%d, role:%d, lastIndex:%" PRId64, pObj->id, pObj->role, pObj->lastIndex); + sdbReleaseLock(pSdb, pObj, false); } - if (readyMnodes <= 0 || updatingMnodes <= 0) { - mInfo("vgId:1, mnode sync not reconfig since readyMnodes:%d updatingMnodes:%d", readyMnodes, updatingMnodes); - return; - } + //if (readyMnodes <= 0 || updatingMnodes <= 0) { + // mInfo("vgId:1, mnode sync not reconfig since readyMnodes:%d updatingMnodes:%d", readyMnodes, updatingMnodes); + // return; + //} if (cfg.myIndex == -1) { #if 1 @@ -777,12 +892,14 @@ static void mndReloadSyncConfig(SMnode *pMnode) { return; } - if (updatingMnodes > 0) { - mInfo("vgId:1, mnode sync reconfig, replica:%d myIndex:%d", cfg.replicaNum, cfg.myIndex); - for (int32_t i = 0; i < cfg.replicaNum; ++i) { + if (pMnode->syncMgmt.sync > 0) { + mInfo("vgId:1, mnode sync reconfig, totalReplica:%d replica:%d myIndex:%d", + cfg.totalReplicaNum, cfg.replicaNum, cfg.myIndex); + + for (int32_t i = 0; i < cfg.totalReplicaNum; ++i) { SNodeInfo *pNode = &cfg.nodeInfo[i]; - mInfo("vgId:1, index:%d, ep:%s:%u dnode:%d cluster:%" PRId64, i, pNode->nodeFqdn, pNode->nodePort, pNode->nodeId, - pNode->clusterId); + mInfo("vgId:1, index:%d, ep:%s:%u dnode:%d cluster:%" PRId64 " role:%d", i, pNode->nodeFqdn, pNode->nodePort, + pNode->nodeId, pNode->clusterId, pNode->nodeRole); } int32_t code = syncReconfig(pMnode->syncMgmt.sync, &cfg); diff --git a/source/dnode/mnode/impl/src/mndProfile.c b/source/dnode/mnode/impl/src/mndProfile.c index 50e502f4ab..2ebb5aeb99 100644 --- a/source/dnode/mnode/impl/src/mndProfile.c +++ b/source/dnode/mnode/impl/src/mndProfile.c @@ -283,6 +283,7 @@ static int32_t mndProcessConnectReq(SRpcMsg *pReq) { connectRsp.connType = connReq.connType; connectRsp.dnodeNum = mndGetDnodeSize(pMnode); connectRsp.svrTimestamp = taosGetTimestampSec(); + connectRsp.passVer = pUser->passVersion; strcpy(connectRsp.sVer, version); snprintf(connectRsp.sDetailVer, sizeof(connectRsp.sDetailVer), "ver:%s\nbuild:%s\ngitinfo:%s", version, buildinfo, @@ -547,6 +548,16 @@ static int32_t mndProcessQueryHeartBeat(SMnode *pMnode, SRpcMsg *pMsg, SClientHb } break; } + case HEARTBEAT_KEY_USER_PASSINFO: { + void *rspMsg = NULL; + int32_t rspLen = 0; + mndValidateUserPassInfo(pMnode, kv->value, kv->valueLen / sizeof(SUserPassVersion), &rspMsg, &rspLen); + if (rspMsg && rspLen > 0) { + SKv kv1 = {.key = HEARTBEAT_KEY_USER_PASSINFO, .valueLen = rspLen, .value = rspMsg}; + taosArrayPush(hbRsp.info, &kv1); + } + break; + } default: mError("invalid kv key:%d", kv->key); hbRsp.status = TSDB_CODE_APP_ERROR; diff --git a/source/dnode/mnode/impl/src/mndSync.c b/source/dnode/mnode/impl/src/mndSync.c index 4965d5c34a..0a6df02f5f 100644 --- a/source/dnode/mnode/impl/src/mndSync.c +++ b/source/dnode/mnode/impl/src/mndSync.c @@ -237,6 +237,23 @@ static void mndBecomeFollower(const SSyncFSM *pFsm) { taosThreadMutexUnlock(&pMgmt->lock); } +static void mndBecomeLearner(const SSyncFSM *pFsm) { + SMnode *pMnode = pFsm->data; + SSyncMgmt *pMgmt = &pMnode->syncMgmt; + mInfo("vgId:1, become learner"); + + taosThreadMutexLock(&pMgmt->lock); + if (pMgmt->transId != 0) { + mInfo("vgId:1, become learner and post sem, trans:%d, failed to propose since not leader", pMgmt->transId); + pMgmt->transId = 0; + pMgmt->transSec = 0; + pMgmt->transSeq = 0; + pMgmt->errCode = TSDB_CODE_SYN_NOT_LEADER; + tsem_post(&pMgmt->syncSem); + } + taosThreadMutexUnlock(&pMgmt->lock); +} + static void mndBecomeLeader(const SSyncFSM *pFsm) { mInfo("vgId:1, become leader"); SMnode *pMnode = pFsm->data; @@ -278,6 +295,7 @@ SSyncFSM *mndSyncMakeFsm(SMnode *pMnode) { pFsm->FpReConfigCb = NULL; pFsm->FpBecomeLeaderCb = mndBecomeLeader; pFsm->FpBecomeFollowerCb = mndBecomeFollower; + pFsm->FpBecomeLearnerCb = mndBecomeLearner; pFsm->FpGetSnapshot = mndSyncGetSnapshot; pFsm->FpGetSnapshotInfo = mndSyncGetSnapshotInfo; pFsm->FpSnapshotStartRead = mndSnapshotStartRead; @@ -317,13 +335,16 @@ int32_t mndInitSync(SMnode *pMnode) { mInfo("vgId:1, start to open sync, replica:%d selfIndex:%d", pMgmt->numOfReplicas, pMgmt->selfIndex); SSyncCfg *pCfg = &syncInfo.syncCfg; + pCfg->totalReplicaNum = pMgmt->numOfTotalReplicas; pCfg->replicaNum = pMgmt->numOfReplicas; pCfg->myIndex = pMgmt->selfIndex; - for (int32_t i = 0; i < pMgmt->numOfReplicas; ++i) { + pCfg->lastIndex = pMgmt->lastIndex; + for (int32_t i = 0; i < pMgmt->numOfTotalReplicas; ++i) { SNodeInfo *pNode = &pCfg->nodeInfo[i]; pNode->nodeId = pMgmt->replicas[i].id; pNode->nodePort = pMgmt->replicas[i].port; tstrncpy(pNode->nodeFqdn, pMgmt->replicas[i].fqdn, sizeof(pNode->nodeFqdn)); + pNode->nodeRole = pMgmt->nodeRoles[i]; (void)tmsgUpdateDnodeInfo(&pNode->nodeId, &pNode->clusterId, pNode->nodeFqdn, &pNode->nodePort); mInfo("vgId:1, index:%d ep:%s:%u dnode:%d cluster:%" PRId64, i, pNode->nodeFqdn, pNode->nodePort, pNode->nodeId, pNode->clusterId); diff --git a/source/dnode/mnode/impl/src/mndTrans.c b/source/dnode/mnode/impl/src/mndTrans.c index 106eea0313..cfb5bef9d0 100644 --- a/source/dnode/mnode/impl/src/mndTrans.c +++ b/source/dnode/mnode/impl/src/mndTrans.c @@ -546,6 +546,7 @@ static void mndTransDropData(STrans *pTrans) { pTrans->param = NULL; pTrans->paramLen = 0; } + (void)taosThreadMutexDestroy(&pTrans->mutex); } static int32_t mndTransActionDelete(SSdb *pSdb, STrans *pTrans, bool callFunc) { @@ -643,7 +644,7 @@ STrans *mndTransCreate(SMnode *pMnode, ETrnPolicy policy, ETrnConflct conflict, pTrans->stage = TRN_STAGE_PREPARE; pTrans->policy = policy; pTrans->conflict = conflict; - pTrans->exec = TRN_EXEC_PRARLLEL; + pTrans->exec = TRN_EXEC_PARALLEL; pTrans->createdTime = taosGetTimestampMs(); pTrans->redoActions = taosArrayInit(TRANS_ARRAY_SIZE, sizeof(STransAction)); pTrans->undoActions = taosArrayInit(TRANS_ARRAY_SIZE, sizeof(STransAction)); @@ -651,6 +652,7 @@ STrans *mndTransCreate(SMnode *pMnode, ETrnPolicy policy, ETrnConflct conflict, pTrans->pRpcArray = taosArrayInit(1, sizeof(SRpcHandleInfo)); pTrans->mTraceId = pReq ? TRACE_GET_ROOTID(&pReq->info.traceId) : 0; taosInitRWLatch(&pTrans->lockRpcArray); + taosThreadMutexInit(&pTrans->mutex, NULL); if (pTrans->redoActions == NULL || pTrans->undoActions == NULL || pTrans->commitActions == NULL || pTrans->pRpcArray == NULL) { @@ -793,6 +795,8 @@ void mndTransSetDbName(STrans *pTrans, const char *dbname, const char *stbname) void mndTransSetSerial(STrans *pTrans) { pTrans->exec = TRN_EXEC_SERIAL; } +void mndTransSetParallel(STrans *pTrans) { pTrans->exec = TRN_EXEC_PARALLEL; } + void mndTransSetOper(STrans *pTrans, EOperType oper) { pTrans->oper = oper; } static int32_t mndTransSync(SMnode *pMnode, STrans *pTrans) { @@ -1307,7 +1311,13 @@ static int32_t mndTransExecuteRedoActionsSerial(SMnode *pMnode, STrans *pTrans) int32_t code = 0; int32_t numOfActions = taosArrayGetSize(pTrans->redoActions); if (numOfActions == 0) return code; - if (pTrans->redoActionPos >= numOfActions) return code; + + taosThreadMutexLock(&pTrans->mutex); + + if (pTrans->redoActionPos >= numOfActions) { + taosThreadMutexUnlock(&pTrans->mutex); + return code; + } mInfo("trans:%d, execute %d actions serial, current redoAction:%d", pTrans->id, numOfActions, pTrans->redoActionPos); @@ -1377,6 +1387,8 @@ static int32_t mndTransExecuteRedoActionsSerial(SMnode *pMnode, STrans *pTrans) } } + taosThreadMutexUnlock(&pTrans->mutex); + return code; } diff --git a/source/dnode/mnode/impl/src/mndUser.c b/source/dnode/mnode/impl/src/mndUser.c index 2a0d753722..3da594109a 100644 --- a/source/dnode/mnode/impl/src/mndUser.c +++ b/source/dnode/mnode/impl/src/mndUser.c @@ -23,7 +23,7 @@ #include "mndTrans.h" #include "tbase64.h" -#define USER_VER_NUMBER 3 +#define USER_VER_NUMBER 4 #define USER_RESERVE_SIZE 64 static int32_t mndCreateDefaultUsers(SMnode *pMnode); @@ -174,6 +174,7 @@ SSdbRaw *mndUserActionEncode(SUserObj *pUser) { SDB_SET_INT8(pRaw, dataPos, pUser->enable, _OVER) SDB_SET_INT8(pRaw, dataPos, pUser->reserve, _OVER) SDB_SET_INT32(pRaw, dataPos, pUser->authVersion, _OVER) + SDB_SET_INT32(pRaw, dataPos, pUser->passVersion, _OVER) SDB_SET_INT32(pRaw, dataPos, numOfReadDbs, _OVER) SDB_SET_INT32(pRaw, dataPos, numOfWriteDbs, _OVER) SDB_SET_INT32(pRaw, dataPos, numOfTopics, _OVER) @@ -263,7 +264,7 @@ static SSdbRow *mndUserActionDecode(SSdbRaw *pRaw) { int8_t sver = 0; if (sdbGetRawSoftVer(pRaw, &sver) != 0) goto _OVER; - if (sver != 1 && sver != 2 && sver != 3) { + if (sver < 1 || sver > USER_VER_NUMBER) { terrno = TSDB_CODE_SDB_INVALID_DATA_VER; goto _OVER; } @@ -285,6 +286,9 @@ static SSdbRow *mndUserActionDecode(SSdbRaw *pRaw) { SDB_GET_INT8(pRaw, dataPos, &pUser->enable, _OVER) SDB_GET_INT8(pRaw, dataPos, &pUser->reserve, _OVER) SDB_GET_INT32(pRaw, dataPos, &pUser->authVersion, _OVER) + if (sver >= 4) { + SDB_GET_INT32(pRaw, dataPos, &pUser->passVersion, _OVER) + } int32_t numOfReadDbs = 0; int32_t numOfWriteDbs = 0; @@ -530,6 +534,7 @@ static int32_t mndUserActionUpdate(SSdb *pSdb, SUserObj *pOld, SUserObj *pNew) { taosWLockLatch(&pOld->lock); pOld->updateTime = pNew->updateTime; pOld->authVersion = pNew->authVersion; + pOld->passVersion = pNew->passVersion; pOld->sysInfo = pNew->sysInfo; pOld->enable = pNew->enable; memcpy(pOld->pass, pNew->pass, TSDB_PASSWORD_LEN); @@ -819,10 +824,14 @@ static int32_t mndProcessAlterUserReq(SRpcMsg *pReq) { if (mndUserDupObj(pUser, &newUser) != 0) goto _OVER; + newUser.passVersion = pUser->passVersion; if (alterReq.alterType == TSDB_ALTER_USER_PASSWD) { char pass[TSDB_PASSWORD_LEN + 1] = {0}; taosEncryptPass_c((uint8_t *)alterReq.pass, strlen(alterReq.pass), pass); memcpy(newUser.pass, pass, TSDB_PASSWORD_LEN); + if (0 != strncmp(pUser->pass, pass, TSDB_PASSWORD_LEN)) { + ++newUser.passVersion; + } } if (alterReq.alterType == TSDB_ALTER_USER_SUPERUSER) { @@ -1422,6 +1431,69 @@ _OVER: return code; } +int32_t mndValidateUserPassInfo(SMnode *pMnode, SUserPassVersion *pUsers, int32_t numOfUses, void **ppRsp, + int32_t *pRspLen) { + int32_t code = 0; + SUserPassBatchRsp batchRsp = {0}; + + for (int32_t i = 0; i < numOfUses; ++i) { + SUserObj *pUser = mndAcquireUser(pMnode, pUsers[i].user); + if (pUser == NULL) { + mError("user:%s, failed to validate user pass since %s", pUsers[i].user, terrstr()); + continue; + } + + pUsers[i].version = ntohl(pUsers[i].version); + if (pUser->passVersion <= pUsers[i].version) { + mTrace("user:%s, not update since mnd passVer %d <= client passVer %d", pUsers[i].user, pUser->passVersion, + pUsers[i].version); + mndReleaseUser(pMnode, pUser); + continue; + } + + SGetUserPassRsp rsp = {0}; + memcpy(rsp.user, pUser->user, TSDB_USER_LEN); + rsp.version = pUser->passVersion; + + if (!batchRsp.pArray && !(batchRsp.pArray = taosArrayInit(numOfUses, sizeof(SGetUserPassRsp)))) { + code = TSDB_CODE_OUT_OF_MEMORY; + mndReleaseUser(pMnode, pUser); + goto _OVER; + } + + taosArrayPush(batchRsp.pArray, &rsp); + mndReleaseUser(pMnode, pUser); + } + + if (taosArrayGetSize(batchRsp.pArray) <= 0) { + goto _OVER; + } + + int32_t rspLen = tSerializeSUserPassBatchRsp(NULL, 0, &batchRsp); + if (rspLen < 0) { + code = TSDB_CODE_OUT_OF_MEMORY; + goto _OVER; + } + void *pRsp = taosMemoryMalloc(rspLen); + if (pRsp == NULL) { + code = TSDB_CODE_OUT_OF_MEMORY; + goto _OVER; + } + tSerializeSUserPassBatchRsp(pRsp, rspLen, &batchRsp); + + *ppRsp = pRsp; + *pRspLen = rspLen; + +_OVER: + if (code) { + *ppRsp = NULL; + *pRspLen = 0; + } + + tFreeSUserPassBatchRsp(&batchRsp); + return code; +} + int32_t mndUserRemoveDb(SMnode *pMnode, STrans *pTrans, char *db) { int32_t code = 0; SSdb *pSdb = pMnode->pSdb; diff --git a/source/dnode/mnode/impl/src/mndVgroup.c b/source/dnode/mnode/impl/src/mndVgroup.c index 0003d07fd6..9497ab3870 100644 --- a/source/dnode/mnode/impl/src/mndVgroup.c +++ b/source/dnode/mnode/impl/src/mndVgroup.c @@ -62,6 +62,7 @@ int32_t mndInitVgroup(SMnode *pMnode) { mndSetMsgHandle(pMnode, TDMT_VND_COMPACT_RSP, mndTransProcessRsp); mndSetMsgHandle(pMnode, TDMT_VND_DISABLE_WRITE_RSP, mndTransProcessRsp); mndSetMsgHandle(pMnode, TDMT_SYNC_FORCE_FOLLOWER_RSP, mndTransProcessRsp); + mndSetMsgHandle(pMnode, TDMT_DND_ALTER_VNODE_TYPE_RSP, mndTransProcessRsp); mndSetMsgHandle(pMnode, TDMT_MND_REDISTRIBUTE_VGROUP, mndProcessRedistributeVgroupMsg); mndSetMsgHandle(pMnode, TDMT_MND_SPLIT_VGROUP, mndProcessSplitVgroupMsg); @@ -203,7 +204,7 @@ static int32_t mndVgroupActionUpdate(SSdb *pSdb, SVgObj *pOld, SVgObj *pNew) { pNew->compStorage = pOld->compStorage; pNew->pointsWritten = pOld->pointsWritten; pNew->compact = pOld->compact; - memcpy(pOld->vnodeGid, pNew->vnodeGid, TSDB_MAX_REPLICA * sizeof(SVnodeGid)); + memcpy(pOld->vnodeGid, pNew->vnodeGid, (TSDB_MAX_REPLICA + TSDB_MAX_LEARNER_REPLICA) * sizeof(SVnodeGid)); return 0; } @@ -244,8 +245,10 @@ void *mndBuildCreateVnodeReq(SMnode *pMnode, SDnodeObj *pDnode, SDbObj *pDb, SVg createReq.compression = pDb->cfg.compression; createReq.strict = pDb->cfg.strict; createReq.cacheLast = pDb->cfg.cacheLast; - createReq.replica = pVgroup->replica; + createReq.replica = 0; + createReq.learnerReplica = 0; createReq.selfIndex = -1; + createReq.learnerSelfIndex = -1; createReq.hashBegin = pVgroup->hashBegin; createReq.hashEnd = pVgroup->hashEnd; createReq.hashMethod = pDb->cfg.hashMethod; @@ -263,7 +266,15 @@ void *mndBuildCreateVnodeReq(SMnode *pMnode, SDnodeObj *pDnode, SDbObj *pDb, SVg createReq.tsdbPageSize = pDb->cfg.tsdbPageSize; for (int32_t v = 0; v < pVgroup->replica; ++v) { - SReplica *pReplica = &createReq.replicas[v]; + SReplica *pReplica = NULL; + + if(pVgroup->vnodeGid[v].nodeRole == TAOS_SYNC_ROLE_VOTER){ + pReplica = &createReq.replicas[createReq.replica]; + } + else{ + pReplica = &createReq.learnerReplicas[createReq.learnerReplica]; + } + SVnodeGid *pVgid = &pVgroup->vnodeGid[v]; SDnodeObj *pVgidDnode = mndAcquireDnode(pMnode, pVgid->dnodeId); if (pVgidDnode == NULL) { @@ -275,21 +286,40 @@ void *mndBuildCreateVnodeReq(SMnode *pMnode, SDnodeObj *pDnode, SDbObj *pDb, SVg memcpy(pReplica->fqdn, pVgidDnode->fqdn, TSDB_FQDN_LEN); mndReleaseDnode(pMnode, pVgidDnode); - if (pDnode->id == pVgid->dnodeId) { - createReq.selfIndex = v; + if(pVgroup->vnodeGid[v].nodeRole == TAOS_SYNC_ROLE_VOTER){ + if (pDnode->id == pVgid->dnodeId) { + createReq.selfIndex = createReq.replica; + } + } + else{ + if (pDnode->id == pVgid->dnodeId) { + createReq.learnerSelfIndex = createReq.learnerReplica; + } + } + + if(pVgroup->vnodeGid[v].nodeRole == TAOS_SYNC_ROLE_VOTER){ + createReq.replica++; + } + else{ + createReq.learnerReplica++; } } - if (createReq.selfIndex == -1) { + if (createReq.selfIndex == -1 && createReq.learnerSelfIndex == -1) { terrno = TSDB_CODE_APP_ERROR; return NULL; } - mInfo("vgId:%d, build create vnode req, replica:%d selfIndex:%d strict:%d", createReq.vgId, createReq.replica, - createReq.selfIndex, createReq.strict); + mInfo("vgId:%d, build create vnode req, replica:%d selfIndex:%d learnerReplica:%d learnerSelfIndex:%d strict:%d", + createReq.vgId, createReq.replica, createReq.selfIndex, createReq.learnerReplica, + createReq.learnerReplica, createReq.strict); for (int32_t i = 0; i < createReq.replica; ++i) { mInfo("vgId:%d, replica:%d ep:%s:%u", createReq.vgId, i, createReq.replicas[i].fqdn, createReq.replicas[i].port); } + for (int32_t i = 0; i < createReq.learnerReplica; ++i) { + mInfo("vgId:%d, replica:%d ep:%s:%u", createReq.vgId, i, createReq.learnerReplicas[i].fqdn, + createReq.learnerReplicas[i].port); + } int32_t contLen = tSerializeSCreateVnodeReq(NULL, 0, &createReq); if (contLen < 0) { @@ -356,12 +386,24 @@ static void *mndBuildAlterVnodeReplicaReq(SMnode *pMnode, SDbObj *pDb, SVgObj *p SAlterVnodeReplicaReq alterReq = { .vgId = pVgroup->vgId, .strict = pDb->cfg.strict, - .replica = pVgroup->replica, + .replica = 0, + .learnerReplica = 0, .selfIndex = -1, + .learnerSelfIndex = -1, }; for (int32_t v = 0; v < pVgroup->replica; ++v) { - SReplica *pReplica = &alterReq.replicas[v]; + SReplica *pReplica = NULL; + + if(pVgroup->vnodeGid[v].nodeRole == TAOS_SYNC_ROLE_VOTER){ + pReplica = &alterReq.replicas[alterReq.replica]; + alterReq.replica++; + } + else{ + pReplica = &alterReq.learnerReplicas[alterReq.learnerReplica]; + alterReq.learnerReplica++; + } + SVnodeGid *pVgid = &pVgroup->vnodeGid[v]; SDnodeObj *pVgidDnode = mndAcquireDnode(pMnode, pVgid->dnodeId); if (pVgidDnode == NULL) return NULL; @@ -371,18 +413,30 @@ static void *mndBuildAlterVnodeReplicaReq(SMnode *pMnode, SDbObj *pDb, SVgObj *p memcpy(pReplica->fqdn, pVgidDnode->fqdn, TSDB_FQDN_LEN); mndReleaseDnode(pMnode, pVgidDnode); - if (dnodeId == pVgid->dnodeId) { - alterReq.selfIndex = v; + if(pVgroup->vnodeGid[v].nodeRole == TAOS_SYNC_ROLE_VOTER){ + if (dnodeId == pVgid->dnodeId) { + alterReq.selfIndex = v; + } + } + else{ + if (dnodeId == pVgid->dnodeId) { + alterReq.learnerSelfIndex = v; + } } } - alterReq.replica = pVgroup->replica; - mInfo("vgId:%d, build alter vnode req, replica:%d selfIndex:%d strict:%d", alterReq.vgId, alterReq.replica, - alterReq.selfIndex, alterReq.strict); + + mInfo("vgId:%d, build alter vnode req, replica:%d selfIndex:%d learnerReplica:%d learnerSelfIndex:%d strict:%d", + alterReq.vgId, alterReq.replica, alterReq.selfIndex, alterReq.learnerReplica, + alterReq.learnerSelfIndex, alterReq.strict); for (int32_t i = 0; i < alterReq.replica; ++i) { mInfo("vgId:%d, replica:%d ep:%s:%u", alterReq.vgId, i, alterReq.replicas[i].fqdn, alterReq.replicas[i].port); } + for (int32_t i = 0; i < alterReq.learnerReplica; ++i) { + mInfo("vgId:%d, learnerReplica:%d ep:%s:%u", alterReq.vgId, i, + alterReq.learnerReplicas[i].fqdn, alterReq.learnerReplicas[i].port); + } - if (alterReq.selfIndex == -1) { + if (alterReq.selfIndex == -1 && alterReq.learnerSelfIndex == -1) { terrno = TSDB_CODE_APP_ERROR; return NULL; } @@ -1194,6 +1248,32 @@ int32_t mndAddAlterVnodeReplicaAction(SMnode *pMnode, STrans *pTrans, SDbObj *pD return 0; } +int32_t mndAddAlterVnodeTypeAction(SMnode *pMnode, STrans *pTrans, SDbObj *pDb, SVgObj *pVgroup, int32_t dnodeId) { + SDnodeObj *pDnode = mndAcquireDnode(pMnode, dnodeId); + if (pDnode == NULL) return -1; + + STransAction action = {0}; + action.epSet = mndGetDnodeEpset(pDnode); + mndReleaseDnode(pMnode, pDnode); + + int32_t contLen = 0; + void *pReq = mndBuildAlterVnodeReplicaReq(pMnode, pDb, pVgroup, dnodeId, &contLen); + if (pReq == NULL) return -1; + + action.pCont = pReq; + action.contLen = contLen; + action.msgType = TDMT_DND_ALTER_VNODE_TYPE; + action.acceptableCode = TSDB_CODE_VND_ALREADY_IS_VOTER; + action.retryCode = TSDB_CODE_VND_NOT_CATCH_UP; + + if (mndTransAppendRedoAction(pTrans, &action) != 0) { + taosMemoryFree(pReq); + return -1; + } + + return 0; +} + static int32_t mndAddDisableVnodeWriteAction(SMnode *pMnode, STrans *pTrans, SDbObj *pDb, SVgObj *pVgroup, int32_t dnodeId) { SDnodeObj *pDnode = mndAcquireDnode(pMnode, dnodeId); @@ -1837,7 +1917,7 @@ int32_t mndAddVgroupBalanceToTrans(SMnode *pMnode, SVgObj *pVgroup, STrans *pTra int32_t vgid = pVgroup->vgId; int8_t replica = pVgroup->replica; - if(pVgroup->replica <= 1) { + if(pVgroup->replica <= 1) { mInfo("trans:%d, vgid:%d no need to balance, replica:%d", pTrans->id, vgid, replica); return -1; } @@ -1871,6 +1951,19 @@ int32_t mndAddVgroupBalanceToTrans(SMnode *pMnode, SVgObj *pVgroup, STrans *pTra return -1; } + SDbObj *pDb = mndAcquireDb(pMnode, pVgroup->dbName); + if (pDb == NULL) { + mError("trans:%d, vgid:%d failed to be balanced to dnode:%d, because db not exist", pTrans->id, vgid, dnodeId); + return -1; + } + + if (mndAddAlterVnodeConfirmAction(pMnode, pTrans, pDb, pVgroup) != 0) { + mError("trans:%d, vgid:%d failed to be balanced to dnode:%d", pTrans->id, vgid, dnodeId); + return -1; + } + + mndReleaseDb(pMnode, pDb); + SSdbRaw *pRaw = mndVgroupActionEncode(pVgroup); if (pRaw == NULL) { mError("trans:%d, vgid:%d failed to encode action to dnode:%d", pTrans->id, vgid, dnodeId); @@ -1885,7 +1978,8 @@ int32_t mndAddVgroupBalanceToTrans(SMnode *pMnode, SVgObj *pVgroup, STrans *pTra } else { - mInfo("trans:%d, vgid:%d cant be balanced to dnode:%d, exist:%d, online:%d", pTrans->id, vgid, dnodeId, exist, online); + mInfo("trans:%d, vgid:%d cant be balanced to dnode:%d, exist:%d, online:%d", + pTrans->id, vgid, dnodeId, exist, online); } return 0; @@ -1953,18 +2047,33 @@ int32_t mndBuildAlterVgroupAction(SMnode *pMnode, STrans *pTrans, SDbObj *pOldDb mInfo("db:%s, vgId:%d, will add 2 vnodes, vn:0 dnode:%d", pVgroup->dbName, pVgroup->vgId, pVgroup->vnodeGid[0].dnodeId); + //add first if (mndAddVnodeToVgroup(pMnode, pTrans, &newVgroup, pArray) != 0) return -1; + + newVgroup.vnodeGid[0].nodeRole = TAOS_SYNC_ROLE_VOTER; + newVgroup.vnodeGid[1].nodeRole = TAOS_SYNC_ROLE_LEARNER; if (mndAddAlterVnodeReplicaAction(pMnode, pTrans, pNewDb, &newVgroup, newVgroup.vnodeGid[0].dnodeId) != 0) return -1; + if (mndAddCreateVnodeAction(pMnode, pTrans, pNewDb, &newVgroup, &newVgroup.vnodeGid[1]) != 0) return -1; + newVgroup.vnodeGid[1].nodeRole = TAOS_SYNC_ROLE_VOTER; + if (mndAddAlterVnodeTypeAction(pMnode, pTrans, pNewDb, &newVgroup, newVgroup.vnodeGid[1].dnodeId) != 0) + return -1; + if (mndAddAlterVnodeConfirmAction(pMnode, pTrans, pNewDb, &newVgroup) != 0) return -1; + //add second if (mndAddVnodeToVgroup(pMnode, pTrans, &newVgroup, pArray) != 0) return -1; + newVgroup.vnodeGid[0].nodeRole = TAOS_SYNC_ROLE_VOTER; + newVgroup.vnodeGid[1].nodeRole = TAOS_SYNC_ROLE_VOTER; + newVgroup.vnodeGid[2].nodeRole = TAOS_SYNC_ROLE_VOTER; + if (mndAddAlterVnodeReplicaAction(pMnode, pTrans, pNewDb, &newVgroup, newVgroup.vnodeGid[0].dnodeId) != 0) return -1; if (mndAddAlterVnodeReplicaAction(pMnode, pTrans, pNewDb, &newVgroup, newVgroup.vnodeGid[1].dnodeId) != 0) return -1; if (mndAddCreateVnodeAction(pMnode, pTrans, pNewDb, &newVgroup, &newVgroup.vnodeGid[2]) != 0) return -1; + if (mndAddAlterVnodeConfirmAction(pMnode, pTrans, pNewDb, &newVgroup) != 0) return -1; } else if (newVgroup.replica == 3 && pNewDb->cfg.replications == 1) { mInfo("db:%s, vgId:%d, will remove 2 vnodes, vn:0 dnode:%d vn:1 dnode:%d vn:2 dnode:%d", pVgroup->dbName, @@ -1989,6 +2098,8 @@ int32_t mndBuildAlterVgroupAction(SMnode *pMnode, STrans *pTrans, SDbObj *pOldDb return -1; } + mndSortVnodeGid(&newVgroup); + { SSdbRaw *pVgRaw = mndVgroupActionEncode(&newVgroup); if (pVgRaw == NULL) return -1; @@ -2006,6 +2117,21 @@ static int32_t mndAddAdjustVnodeHashRangeAction(SMnode *pMnode, STrans *pTrans, return 0; } +<<<<<<< HEAD +======= +static int32_t mndTransCommitVgStatus(STrans *pTrans, SVgObj *pVg, ESdbStatus vgStatus) { + SSdbRaw *pRaw = mndVgroupActionEncode(pVg); + if (pRaw == NULL) goto _err; + if (mndTransAppendCommitlog(pTrans, pRaw) != 0) goto _err; + (void)sdbSetRawStatus(pRaw, vgStatus); + pRaw = NULL; + return 0; +_err: + sdbFreeRaw(pRaw); + return -1; +} + +>>>>>>> enh/3.0 int32_t mndSplitVgroup(SMnode *pMnode, SRpcMsg *pReq, SDbObj *pDb, SVgObj *pVgroup) { int32_t code = -1; STrans *pTrans = NULL; @@ -2067,6 +2193,7 @@ int32_t mndSplitVgroup(SMnode *pMnode, SRpcMsg *pReq, SDbObj *pDb, SVgObj *pVgro mInfo("vgId:%d, vnode:%d dnode:%d", newVg2.vgId, i, newVg2.vnodeGid[i].dnodeId); } + // alter vgId and hash range int32_t maxVgId = sdbGetMaxId(pMnode->pSdb, SDB_VGROUP); if (mndAddAlterVnodeHashRangeAction(pMnode, pTrans, &newVg1, maxVgId) != 0) goto _OVER; newVg1.vgId = maxVgId; @@ -2075,31 +2202,24 @@ int32_t mndSplitVgroup(SMnode *pMnode, SRpcMsg *pReq, SDbObj *pDb, SVgObj *pVgro if (mndAddAlterVnodeHashRangeAction(pMnode, pTrans, &newVg2, maxVgId) != 0) goto _OVER; newVg2.vgId = maxVgId; + if (mndAddAlterVnodeConfirmAction(pMnode, pTrans, pDb, &newVg1) != 0) goto _OVER; + + if (mndAddAlterVnodeConfirmAction(pMnode, pTrans, pDb, &newVg2) != 0) goto _OVER; + // adjust vgroup replica if (pDb->cfg.replications != newVg1.replica) { if (mndBuildAlterVgroupAction(pMnode, pTrans, pDb, pDb, &newVg1, pArray) != 0) goto _OVER; + } else { + if (mndTransCommitVgStatus(pTrans, &newVg1, SDB_STATUS_READY) < 0) goto _OVER; } + if (pDb->cfg.replications != newVg2.replica) { if (mndBuildAlterVgroupAction(pMnode, pTrans, pDb, pDb, &newVg2, pArray) != 0) goto _OVER; + } else { + if (mndTransCommitVgStatus(pTrans, &newVg2, SDB_STATUS_READY) < 0) goto _OVER; } - pRaw = mndVgroupActionEncode(&newVg1); - if (pRaw == NULL) goto _OVER; - if (mndTransAppendCommitlog(pTrans, pRaw) != 0) goto _OVER; - (void)sdbSetRawStatus(pRaw, SDB_STATUS_READY); - pRaw = NULL; - - pRaw = mndVgroupActionEncode(&newVg2); - if (pRaw == NULL) goto _OVER; - if (mndTransAppendCommitlog(pTrans, pRaw) != 0) goto _OVER; - (void)sdbSetRawStatus(pRaw, SDB_STATUS_READY); - pRaw = NULL; - - pRaw = mndVgroupActionEncode(pVgroup); - if (pRaw == NULL) goto _OVER; - if (mndTransAppendCommitlog(pTrans, pRaw) != 0) goto _OVER; - (void)sdbSetRawStatus(pRaw, SDB_STATUS_DROPPED); - pRaw = NULL; + if (mndTransCommitVgStatus(pTrans, pVgroup, SDB_STATUS_DROPPED) < 0) goto _OVER; memcpy(&dbObj, pDb, sizeof(SDbObj)); if (dbObj.cfg.pRetensions != NULL) { @@ -2126,37 +2246,13 @@ _OVER: return code; } -static int32_t mndProcessSplitVgroupMsg(SRpcMsg *pReq) { - SMnode *pMnode = pReq->info.node; - int32_t code = -1; - SVgObj *pVgroup = NULL; - SDbObj *pDb = NULL; +extern int32_t mndProcessSplitVgroupMsgImp(SRpcMsg *pReq); - SSplitVgroupReq req = {0}; - if (tDeserializeSSplitVgroupReq(pReq->pCont, pReq->contLen, &req) != 0) { - terrno = TSDB_CODE_INVALID_MSG; - goto _OVER; - } +static int32_t mndProcessSplitVgroupMsg(SRpcMsg *pReq) { return mndProcessSplitVgroupMsgImp(pReq); } - mInfo("vgId:%d, start to split", req.vgId); - if (mndCheckOperPrivilege(pMnode, pReq->info.conn.user, MND_OPER_SPLIT_VGROUP) != 0) { - goto _OVER; - } - - pVgroup = mndAcquireVgroup(pMnode, req.vgId); - if (pVgroup == NULL) goto _OVER; - - pDb = mndAcquireDb(pMnode, pVgroup->dbName); - if (pDb == NULL) goto _OVER; - - code = mndSplitVgroup(pMnode, pReq, pDb, pVgroup); - if (code == 0) code = TSDB_CODE_ACTION_IN_PROGRESS; - -_OVER: - mndReleaseVgroup(pMnode, pVgroup); - mndReleaseDb(pMnode, pDb); - return code; -} +#ifndef TD_ENTERPRISE +int32_t mndProcessSplitVgroupMsgImp(SRpcMsg *pReq) { return 0; } +#endif static int32_t mndSetBalanceVgroupInfoToTrans(SMnode *pMnode, STrans *pTrans, SDbObj *pDb, SVgObj *pVgroup, SDnodeObj *pSrc, SDnodeObj *pDst) { diff --git a/source/dnode/vnode/CMakeLists.txt b/source/dnode/vnode/CMakeLists.txt index f3087e3662..e8660cd6ad 100644 --- a/source/dnode/vnode/CMakeLists.txt +++ b/source/dnode/vnode/CMakeLists.txt @@ -84,6 +84,7 @@ target_include_directories( PUBLIC "inc" PUBLIC "src/inc" PUBLIC "${TD_SOURCE_DIR}/include/libs/scalar" + PUBLIC "${TD_SOURCE_DIR}/contrib/rocksdb/include" ) target_link_libraries( vnode @@ -100,6 +101,7 @@ target_link_libraries( # PUBLIC bdb # PUBLIC scalar + PUBLIC rocksdb PUBLIC transport PUBLIC stream PUBLIC index diff --git a/source/dnode/vnode/inc/vnode.h b/source/dnode/vnode/inc/vnode.h index 88460cd3ca..eb70598201 100644 --- a/source/dnode/vnode/inc/vnode.h +++ b/source/dnode/vnode/inc/vnode.h @@ -61,13 +61,15 @@ void vnodeClose(SVnode *pVnode); int32_t vnodeSyncCommit(SVnode *pVnode); int32_t vnodeBegin(SVnode *pVnode); -int32_t vnodeStart(SVnode *pVnode); -void vnodeStop(SVnode *pVnode); -int64_t vnodeGetSyncHandle(SVnode *pVnode); -void vnodeGetSnapshot(SVnode *pVnode, SSnapshot *pSnapshot); -void vnodeGetInfo(SVnode *pVnode, const char **dbname, int32_t *vgId); -int32_t vnodeProcessCreateTSma(SVnode *pVnode, void *pCont, uint32_t contLen); -int32_t vnodeGetAllTableList(SVnode *pVnode, uint64_t uid, SArray *list); +int32_t vnodeStart(SVnode *pVnode); +void vnodeStop(SVnode *pVnode); +int64_t vnodeGetSyncHandle(SVnode *pVnode); +void vnodeGetSnapshot(SVnode *pVnode, SSnapshot *pSnapshot); +void vnodeGetInfo(SVnode *pVnode, const char **dbname, int32_t *vgId); +int32_t vnodeProcessCreateTSma(SVnode *pVnode, void *pCont, uint32_t contLen); +int32_t vnodeGetAllTableList(SVnode *pVnode, uint64_t uid, SArray *list); +int32_t vnodeIsCatchUp(SVnode *pVnode); +ESyncRole vnodeGetRole(SVnode *pVnode); int32_t vnodeGetCtbIdList(SVnode *pVnode, int64_t suid, SArray *list); int32_t vnodeGetCtbIdListByFilter(SVnode *pVnode, int64_t suid, SArray *list, bool (*filter)(void *arg), void *arg); @@ -175,8 +177,11 @@ typedef struct STsdbReader STsdbReader; #define CACHESCAN_RETRIEVE_LAST_ROW 0x4 #define CACHESCAN_RETRIEVE_LAST 0x8 -int32_t tsdbReaderOpen(SVnode *pVnode, SQueryTableDataCond *pCond, void *pTableList, int32_t numOfTables, - SSDataBlock *pResBlock, STsdbReader **ppReader, const char *idstr, bool countOnly); +int32_t tsdbSetTableList(STsdbReader *pReader, const void *pTableList, int32_t num); +int32_t tsdbReaderOpen(SVnode *pVnode, SQueryTableDataCond *pCond, void *pTableList, int32_t numOfTables, + SSDataBlock *pResBlock, STsdbReader **ppReader, const char *idstr, bool countOnly); + +void tsdbReaderSetId(STsdbReader *pReader, const char *idstr); void tsdbReaderClose(STsdbReader *pReader); int32_t tsdbNextDataBlock(STsdbReader *pReader, bool *hasNext); int32_t tsdbRetrieveDatablockSMA(STsdbReader *pReader, SSDataBlock *pDataBlock, bool *allHave); @@ -193,8 +198,9 @@ void tsdbReaderSetId(STsdbReader *pReader, const char *idstr); void tsdbReaderSetCloseFlag(STsdbReader *pReader); int32_t tsdbCacherowsReaderOpen(void *pVnode, int32_t type, void *pTableIdList, int32_t numOfTables, int32_t numOfCols, - uint64_t suid, void **pReader, const char *idstr); -int32_t tsdbRetrieveCacheRows(void *pReader, SSDataBlock *pResBlock, const int32_t *slotIds, SArray *pTableUids); + SArray *pCidList, int32_t *pSlotIds, uint64_t suid, void **pReader, const char *idstr); +int32_t tsdbRetrieveCacheRows(void *pReader, SSDataBlock *pResBlock, const int32_t *slotIds, const int32_t *dstSlotIds, + SArray *pTableUids); void *tsdbCacherowsReaderClose(void *pReader); int32_t tsdbGetTableSchema(SVnode *pVnode, int64_t uid, STSchema **pSchema, int64_t *suid); @@ -255,7 +261,7 @@ int32_t tqReaderAddTbUidList(STqReader *pReader, const SArray *pTableUidList); int32_t tqReaderRemoveTbUidList(STqReader *pReader, const SArray *tbUidList); int32_t tqSeekVer(STqReader *pReader, int64_t ver, const char *id); -int32_t tqNextBlockInWal(STqReader* pReader); +int32_t tqNextBlockInWal(STqReader *pReader); bool tqNextBlockImpl(STqReader *pReader); int32_t extractSubmitMsgFromWal(SWalReader *pReader, SPackedData *pPackedData); @@ -333,6 +339,7 @@ struct SVnodeCfg { SVnodeStats vndStats; uint32_t hashBegin; uint32_t hashEnd; + bool hashChange; int16_t sttTrigger; int16_t hashPrefix; int16_t hashSuffix; diff --git a/source/dnode/vnode/src/inc/tq.h b/source/dnode/vnode/src/inc/tq.h index d6e4b86097..8207b02c4c 100644 --- a/source/dnode/vnode/src/inc/tq.h +++ b/source/dnode/vnode/src/inc/tq.h @@ -107,6 +107,7 @@ typedef struct { SMqDataRsp* pDataRsp; char subKey[TSDB_SUBSCRIBE_KEY_LEN]; SRpcHandleInfo info; + STqHandle* pHandle; } STqPushEntry; struct STQ { @@ -146,8 +147,14 @@ int32_t tqFetchLog(STQ* pTq, STqHandle* pHandle, int64_t* fetchOffset, SWalCkHea // tqExec int32_t tqTaosxScanLog(STQ* pTq, STqHandle* pHandle, SPackedData submit, STaosxRsp* pRsp, int32_t* totalRows); int32_t tqAddBlockDataToRsp(const SSDataBlock* pBlock, SMqDataRsp* pRsp, int32_t numOfCols, int8_t precision); +<<<<<<< HEAD int32_t tqSendDataRsp(STQ* pTq, const SRpcMsg* pMsg, const SMqPollReq* pReq, const SMqDataRsp* pRsp, int32_t type); int32_t tqPushDataRsp(STQ* pTq, STqHandle* pHandle); +======= +int32_t tqSendDataRsp(STqHandle* pHandle, const SRpcMsg* pMsg, const SMqPollReq* pReq, const SMqDataRsp* pRsp, + int32_t type, int32_t vgId); +int32_t tqPushDataRsp(STqPushEntry* pPushEntry, int32_t vgId); +>>>>>>> enh/3.0 // tqMeta int32_t tqMetaOpen(STQ* pTq); @@ -183,7 +190,17 @@ int32_t tqStreamTasksScanWal(STQ* pTq); char* createStreamTaskIdStr(int64_t streamId, int32_t taskId); int32_t tqAddInputBlockNLaunchTask(SStreamTask* pTask, SStreamQueueItem* pQueueItem, int64_t ver); int32_t tqExtractDataForMq(STQ* pTq, STqHandle* pHandle, const SMqPollReq* pRequest, SRpcMsg* pMsg); +int32_t tqDoSendDataRsp(const SRpcHandleInfo* pRpcHandleInfo, const SMqDataRsp* pRsp, int32_t epoch, int64_t consumerId, + int32_t type, int64_t sver, int64_t ever); +<<<<<<< HEAD +======= +int32_t tqInitDataRsp(SMqDataRsp* pRsp, const SMqPollReq* pReq); +void doSaveTaskOffset(STqOffsetStore* pOffsetStore, const char* pKey, int64_t ver); +void saveOffsetForAllTasks(STQ* pTq, int64_t ver); +void initOffsetForAllRestoreTasks(STQ* pTq); + +>>>>>>> enh/3.0 #ifdef __cplusplus } #endif diff --git a/source/dnode/vnode/src/inc/tsdb.h b/source/dnode/vnode/src/inc/tsdb.h index 6102487400..b2bc9abf33 100644 --- a/source/dnode/vnode/src/inc/tsdb.h +++ b/source/dnode/vnode/src/inc/tsdb.h @@ -345,6 +345,16 @@ struct STsdbFS { SArray *aDFileSet; // SArray }; +typedef struct { + rocksdb_t *db; + rocksdb_options_t *options; + rocksdb_flushoptions_t *flushoptions; + rocksdb_writeoptions_t *writeoptions; + rocksdb_readoptions_t *readoptions; + rocksdb_writebatch_t *writebatch; + TdThreadMutex rMutex; +} SRocksCache; + struct STsdb { char *path; SVnode *pVnode; @@ -357,6 +367,7 @@ struct STsdb { TdThreadMutex lruMutex; SLRUCache *biCache; TdThreadMutex biMutex; + SRocksCache rCache; }; struct TSDBKEY { @@ -794,6 +805,8 @@ typedef struct SCacheRowsReader { uint64_t suid; char **transferBuf; // todo remove it soon int32_t numOfCols; + SArray *pCidList; + int32_t *pSlotIds; int32_t type; int32_t tableIndex; // currently returned result tables STableKeyInfo *pTableList; // table id list @@ -814,6 +827,10 @@ typedef struct { int32_t tsdbOpenCache(STsdb *pTsdb); void tsdbCloseCache(STsdb *pTsdb); +int32_t tsdbCacheUpdate(STsdb *pTsdb, tb_uid_t suid, tb_uid_t uid, TSDBROW *row); +int32_t tsdbCacheGet(STsdb *pTsdb, tb_uid_t uid, SArray *pLastArray, SCacheRowsReader *pr, int32_t ltype); +int32_t tsdbCacheDel(STsdb *pTsdb, tb_uid_t suid, tb_uid_t uid, TSKEY sKey, TSKEY eKey); + int32_t tsdbCacheInsertLast(SLRUCache *pCache, tb_uid_t uid, TSDBROW *row, STsdb *pTsdb); int32_t tsdbCacheInsertLastrow(SLRUCache *pCache, STsdb *pTsdb, tb_uid_t uid, TSDBROW *row, bool dup); int32_t tsdbCacheGetLastH(SLRUCache *pCache, tb_uid_t uid, SCacheRowsReader *pr, LRUHandle **h); diff --git a/source/dnode/vnode/src/inc/vnodeInt.h b/source/dnode/vnode/src/inc/vnodeInt.h index d58263ab86..24c8c5efa7 100644 --- a/source/dnode/vnode/src/inc/vnodeInt.h +++ b/source/dnode/vnode/src/inc/vnodeInt.h @@ -19,6 +19,7 @@ #include "executor.h" #include "filter.h" #include "qworker.h" +#include "rocksdb/c.h" #include "sync.h" #include "tRealloc.h" #include "tchecksum.h" @@ -177,6 +178,7 @@ int tsdbClose(STsdb** pTsdb); int32_t tsdbBegin(STsdb* pTsdb); int32_t tsdbPrepareCommit(STsdb* pTsdb); int32_t tsdbCommit(STsdb* pTsdb, SCommitInfo* pInfo); +int32_t tsdbCacheCommit(STsdb* pTsdb); int32_t tsdbCompact(STsdb* pTsdb, SCompactInfo* pInfo); int32_t tsdbFinishCommit(STsdb* pTsdb); int32_t tsdbRollbackCommit(STsdb* pTsdb); @@ -193,9 +195,16 @@ STQ* tqOpen(const char* path, SVnode* pVnode); void tqNotifyClose(STQ*); void tqClose(STQ*); int tqPushMsg(STQ*, void* msg, int32_t msgLen, tmsg_t msgType, int64_t ver); +<<<<<<< HEAD int tqRegisterPushHandle(STQ* pTq, void* handle, SRpcMsg* pMsg); int tqUnregisterPushHandle(STQ* pTq, void* pHandle); int tqStartStreamTasks(STQ* pTq); // restore all stream tasks after vnode launching completed. +======= +int tqRegisterPushHandle(STQ* pTq, void* pHandle, const SMqPollReq* pRequest, SRpcMsg* pRpcMsg, SMqDataRsp* pDataRsp, + int32_t type); +int tqUnregisterPushHandle(STQ* pTq, const char* pKey, int32_t keyLen, uint64_t consumerId, bool rspConsumer); +int tqStartStreamTasks(STQ* pTq); // restore all stream tasks after vnode launching completed. +>>>>>>> enh/3.0 int tqCommit(STQ*); int32_t tqUpdateTbUidList(STQ* pTq, const SArray* tbUidList, bool isAdd); @@ -206,7 +215,10 @@ int32_t tqProcessDelCheckInfoReq(STQ* pTq, int64_t version, char* msg, int32_t m int32_t tqProcessSubscribeReq(STQ* pTq, int64_t version, char* msg, int32_t msgLen); int32_t tqProcessDeleteSubReq(STQ* pTq, int64_t version, char* msg, int32_t msgLen); int32_t tqProcessOffsetCommitReq(STQ* pTq, int64_t version, char* msg, int32_t msgLen); +int32_t tqProcessSeekReq(STQ* pTq, int64_t sversion, char* msg, int32_t msgLen); int32_t tqProcessPollReq(STQ* pTq, SRpcMsg* pMsg); +int32_t tqProcessVgWalInfoReq(STQ* pTq, SRpcMsg* pMsg); + // tq-stream int32_t tqProcessTaskDeployReq(STQ* pTq, int64_t version, char* msg, int32_t msgLen); int32_t tqProcessTaskDropReq(STQ* pTq, int64_t version, char* msg, int32_t msgLen); diff --git a/source/dnode/vnode/src/tq/tq.c b/source/dnode/vnode/src/tq/tq.c index a0f913ec10..f4fcc7c02a 100644 --- a/source/dnode/vnode/src/tq/tq.c +++ b/source/dnode/vnode/src/tq/tq.c @@ -71,10 +71,21 @@ static void destroyTqHandle(void* data) { walCloseReader(pData->pWalReader); tqCloseReader(pData->execHandle.pTqReader); } +<<<<<<< HEAD if(pData->msg != NULL) { rpcFreeCont(pData->msg->pCont); taosMemoryFree(pData->msg); pData->msg = NULL; +======= +} + +static void tqPushEntryFree(void* data) { + STqPushEntry* p = *(void**)data; + if (p->pDataRsp->head.mqMsgType == TMQ_MSG_TYPE__POLL_RSP) { + tDeleteMqDataRsp(p->pDataRsp); + } else if (p->pDataRsp->head.mqMsgType == TMQ_MSG_TYPE__TAOSX_RSP) { + tDeleteSTaosxRsp((STaosxRsp*)p->pDataRsp); +>>>>>>> enh/3.0 } } @@ -151,6 +162,7 @@ void tqClose(STQ* pTq) { taosMemoryFree(pTq); } +<<<<<<< HEAD void tqNotifyClose(STQ* pTq) { if (pTq != NULL) { taosWLockLatch(&pTq->pStreamMeta->lock); @@ -231,18 +243,36 @@ int32_t tqPushDataRsp(STQ* pTq, STqHandle* pHandle) { dataRsp.head.epoch = pHandle->epoch; dataRsp.head.mqMsgType = TMQ_MSG_TYPE__POLL_RSP; doSendDataRsp(&pHandle->msg->info, &dataRsp, pHandle->epoch, pHandle->consumerId, TMQ_MSG_TYPE__POLL_RSP); +======= +int32_t tqPushDataRsp(STqPushEntry* pPushEntry, int32_t vgId) { + SMqDataRsp* pRsp = pPushEntry->pDataRsp; + SMqRspHead* pHeader = &pPushEntry->pDataRsp->head; + + int64_t sver = 0, ever = 0; + walReaderValidVersionRange(pPushEntry->pHandle->execHandle.pTqReader->pWalReader, &sver, &ever); + + tqDoSendDataRsp(&pPushEntry->info, pRsp, pHeader->epoch, pHeader->consumerId, pHeader->mqMsgType, sver, ever); +>>>>>>> enh/3.0 char buf1[80] = {0}; char buf2[80] = {0}; tFormatOffset(buf1, tListLen(buf1), &dataRsp.reqOffset); tFormatOffset(buf2, tListLen(buf2), &dataRsp.rspOffset); tqDebug("vgId:%d, from consumer:0x%" PRIx64 " (epoch %d) push rsp, block num: %d, req:%s, rsp:%s", +<<<<<<< HEAD TD_VID(pTq->pVnode), dataRsp.head.consumerId, dataRsp.head.epoch, dataRsp.blockNum, buf1, buf2); +======= + vgId, pRsp->head.consumerId, pRsp->head.epoch, pRsp->blockNum, buf1, buf2); +>>>>>>> enh/3.0 return 0; } -int32_t tqSendDataRsp(STQ* pTq, const SRpcMsg* pMsg, const SMqPollReq* pReq, const SMqDataRsp* pRsp, int32_t type) { - doSendDataRsp(&pMsg->info, pRsp, pReq->epoch, pReq->consumerId, type); +int32_t tqSendDataRsp(STqHandle* pHandle, const SRpcMsg* pMsg, const SMqPollReq* pReq, const SMqDataRsp* pRsp, + int32_t type, int32_t vgId) { + int64_t sver = 0, ever = 0; + walReaderValidVersionRange(pHandle->execHandle.pTqReader->pWalReader, &sver, &ever); + + tqDoSendDataRsp(&pMsg->info, pRsp, pReq->epoch, pReq->consumerId, type, sver, ever); char buf1[80] = {0}; char buf2[80] = {0}; @@ -250,50 +280,54 @@ int32_t tqSendDataRsp(STQ* pTq, const SRpcMsg* pMsg, const SMqPollReq* pReq, con tFormatOffset(buf2, 80, &pRsp->rspOffset); tqDebug("vgId:%d consumer:0x%" PRIx64 " (epoch %d) send rsp, block num:%d, req:%s, rsp:%s, reqId:0x%" PRIx64, - TD_VID(pTq->pVnode), pReq->consumerId, pReq->epoch, pRsp->blockNum, buf1, buf2, pReq->reqId); + vgId, pReq->consumerId, pReq->epoch, pRsp->blockNum, buf1, buf2, pReq->reqId); return 0; } int32_t tqProcessOffsetCommitReq(STQ* pTq, int64_t sversion, char* msg, int32_t msgLen) { - STqOffset offset = {0}; - int32_t vgId = TD_VID(pTq->pVnode); + SMqVgOffset vgOffset = {0}; + int32_t vgId = TD_VID(pTq->pVnode); SDecoder decoder; tDecoderInit(&decoder, (uint8_t*)msg, msgLen); - if (tDecodeSTqOffset(&decoder, &offset) < 0) { + if (tDecodeMqVgOffset(&decoder, &vgOffset) < 0) { return -1; } tDecoderClear(&decoder); - if (offset.val.type == TMQ_OFFSET__SNAPSHOT_DATA || offset.val.type == TMQ_OFFSET__SNAPSHOT_META) { + STqOffset* pOffset = &vgOffset.offset; + + if (pOffset->val.type == TMQ_OFFSET__SNAPSHOT_DATA || pOffset->val.type == TMQ_OFFSET__SNAPSHOT_META) { tqDebug("receive offset commit msg to %s on vgId:%d, offset(type:snapshot) uid:%" PRId64 ", ts:%" PRId64, - offset.subKey, vgId, offset.val.uid, offset.val.ts); - } else if (offset.val.type == TMQ_OFFSET__LOG) { - tqDebug("receive offset commit msg to %s on vgId:%d, offset(type:log) version:%" PRId64, offset.subKey, vgId, - offset.val.version); - if (offset.val.version + 1 == sversion) { - offset.val.version += 1; + pOffset->subKey, vgId, pOffset->val.uid, pOffset->val.ts); + } else if (pOffset->val.type == TMQ_OFFSET__LOG) { + tqDebug("receive offset commit msg to %s on vgId:%d, offset(type:log) version:%" PRId64, pOffset->subKey, vgId, + pOffset->val.version); + if (pOffset->val.version + 1 == sversion) { + pOffset->val.version += 1; } } else { - tqError("invalid commit offset type:%d", offset.val.type); + tqError("invalid commit offset type:%d", pOffset->val.type); return -1; } - STqOffset* pSavedOffset = tqOffsetRead(pTq->pOffsetStore, offset.subKey); - if (pSavedOffset != NULL && tqOffsetLessOrEqual(&offset, pSavedOffset)) { + STqOffset* pSavedOffset = tqOffsetRead(pTq->pOffsetStore, pOffset->subKey); + if (pSavedOffset != NULL && tqOffsetLessOrEqual(pOffset, pSavedOffset)) { + tqDebug("not update the offset, vgId:%d sub:%s since committed:%" PRId64 " less than/equal to existed:%" PRId64, + vgId, pOffset->subKey, pOffset->val.version, pSavedOffset->val.version); return 0; // no need to update the offset value } // save the new offset value - if (tqOffsetWrite(pTq->pOffsetStore, &offset) < 0) { + if (tqOffsetWrite(pTq->pOffsetStore, pOffset) < 0) { return -1; } - if (offset.val.type == TMQ_OFFSET__LOG) { - STqHandle* pHandle = taosHashGet(pTq->pHandle, offset.subKey, strlen(offset.subKey)); - if (pHandle && (walRefVer(pHandle->pRef, offset.val.version) < 0)) { + if (pOffset->val.type == TMQ_OFFSET__LOG) { + STqHandle* pHandle = taosHashGet(pTq->pHandle, pOffset->subKey, strlen(pOffset->subKey)); + if (pHandle && (walRefVer(pHandle->pRef, pOffset->val.version) < 0)) { return -1; } } @@ -301,6 +335,78 @@ int32_t tqProcessOffsetCommitReq(STQ* pTq, int64_t sversion, char* msg, int32_t return 0; } +int32_t tqProcessSeekReq(STQ* pTq, int64_t sversion, char* msg, int32_t msgLen) { + SMqVgOffset vgOffset = {0}; + int32_t vgId = TD_VID(pTq->pVnode); + + SDecoder decoder; + tDecoderInit(&decoder, (uint8_t*)msg, msgLen); + if (tDecodeMqVgOffset(&decoder, &vgOffset) < 0) { + return -1; + } + + tDecoderClear(&decoder); + + STqOffset* pOffset = &vgOffset.offset; + if (pOffset->val.type != TMQ_OFFSET__LOG) { + tqError("vgId:%d, subKey:%s invalid seek offset type:%d", vgId, pOffset->subKey, pOffset->val.type); + return -1; + } + + STqHandle* pHandle = taosHashGet(pTq->pHandle, pOffset->subKey, strlen(pOffset->subKey)); + if (pHandle == NULL) { + tqError("tmq seek: consumer:0x%" PRIx64 " vgId:%d subkey %s not found", vgOffset.consumerId, vgId, + pOffset->subKey); + terrno = TSDB_CODE_INVALID_MSG; + return -1; + } + + // 2. check consumer-vg assignment status + taosRLockLatch(&pTq->lock); + if (pHandle->consumerId != vgOffset.consumerId) { + tqDebug("ERROR tmq seek: consumer:0x%" PRIx64 " vgId:%d, subkey %s, mismatch for saved handle consumer:0x%" PRIx64, + vgOffset.consumerId, vgId, pOffset->subKey, pHandle->consumerId); + terrno = TSDB_CODE_TMQ_CONSUMER_MISMATCH; + taosRUnLockLatch(&pTq->lock); + return -1; + } + taosRUnLockLatch(&pTq->lock); + + //3. check the offset info + STqOffset* pSavedOffset = tqOffsetRead(pTq->pOffsetStore, pOffset->subKey); + if (pSavedOffset != NULL) { + if (pSavedOffset->val.type != TMQ_OFFSET__LOG) { + tqError("invalid saved offset type, vgId:%d sub:%s", vgId, pOffset->subKey); + return 0; // no need to update the offset value + } + + if (pSavedOffset->val.version == pOffset->val.version) { + tqDebug("vgId:%d subKey:%s no need to seek to %" PRId64 " prev offset:%" PRId64, vgId, pOffset->subKey, + pOffset->val.version, pSavedOffset->val.version); + return 0; + } + } + + int64_t sver = 0, ever = 0; + walReaderValidVersionRange(pHandle->execHandle.pTqReader->pWalReader, &sver, &ever); + if (pOffset->val.version < sver) { + pOffset->val.version = sver; + } else if (pOffset->val.version > ever) { + pOffset->val.version = ever; + } + + // save the new offset value + tqDebug("vgId:%d sub:%s seek to %" PRId64 " prev offset:%" PRId64, vgId, pOffset->subKey, pOffset->val.version, + pSavedOffset->val.version); + + if (tqOffsetWrite(pTq->pOffsetStore, pOffset) < 0) { + tqError("failed to save offset, vgId:%d sub:%s seek to %" PRId64, vgId, pOffset->subKey, pOffset->val.version); + return -1; + } + + return 0; +} + int32_t tqCheckColModifiable(STQ* pTq, int64_t tbUid, int32_t colId) { void* pIter = NULL; @@ -377,6 +483,81 @@ int32_t tqProcessPollReq(STQ* pTq, SRpcMsg* pMsg) { return tqExtractDataForMq(pTq, pHandle, &req, pMsg); } +int32_t tqProcessVgWalInfoReq(STQ* pTq, SRpcMsg* pMsg) { + SMqPollReq req = {0}; + if (tDeserializeSMqPollReq(pMsg->pCont, pMsg->contLen, &req) < 0) { + tqError("tDeserializeSMqPollReq %d failed", pMsg->contLen); + terrno = TSDB_CODE_INVALID_MSG; + return -1; + } + + int64_t consumerId = req.consumerId; + STqOffsetVal reqOffset = req.reqOffset; + int32_t vgId = TD_VID(pTq->pVnode); + + // 1. find handle + STqHandle* pHandle = taosHashGet(pTq->pHandle, req.subKey, strlen(req.subKey)); + if (pHandle == NULL) { + tqError("consumer:0x%" PRIx64 " vgId:%d subkey:%s not found", consumerId, vgId, req.subKey); + terrno = TSDB_CODE_INVALID_MSG; + return -1; + } + + // 2. check re-balance status + taosRLockLatch(&pTq->lock); + if (pHandle->consumerId != consumerId) { + tqDebug("ERROR consumer:0x%" PRIx64 " vgId:%d, subkey %s, mismatch for saved handle consumer:0x%" PRIx64, + consumerId, vgId, req.subKey, pHandle->consumerId); + terrno = TSDB_CODE_TMQ_CONSUMER_MISMATCH; + taosRUnLockLatch(&pTq->lock); + return -1; + } + taosRUnLockLatch(&pTq->lock); + + int64_t sver = 0, ever = 0; + walReaderValidVersionRange(pHandle->execHandle.pTqReader->pWalReader, &sver, &ever); + + SMqDataRsp dataRsp = {0}; + tqInitDataRsp(&dataRsp, &req); + + STqOffset* pOffset = tqOffsetRead(pTq->pOffsetStore, req.subKey); + if (pOffset != NULL) { + if (pOffset->val.type != TMQ_OFFSET__LOG) { + tqError("consumer:0x%" PRIx64 " vgId:%d subkey:%s use snapshot, no valid wal info", consumerId, vgId, req.subKey); + terrno = TSDB_CODE_INVALID_PARA; + tDeleteMqDataRsp(&dataRsp); + return -1; + } + + dataRsp.rspOffset.type = TMQ_OFFSET__LOG; + dataRsp.rspOffset.version = pOffset->val.version; + } else { + if (req.useSnapshot == true) { + tqError("consumer:0x%" PRIx64 " vgId:%d subkey:%s snapshot not support wal info", consumerId, vgId, req.subKey); + terrno = TSDB_CODE_INVALID_PARA; + tDeleteMqDataRsp(&dataRsp); + return -1; + } + + dataRsp.rspOffset.type = TMQ_OFFSET__LOG; + + if (reqOffset.type == TMQ_OFFSET__RESET_EARLIEAST) { + dataRsp.rspOffset.version = sver; + } else if (reqOffset.type == TMQ_OFFSET__RESET_LATEST) { + dataRsp.rspOffset.version = ever; + } else { + tqError("consumer:0x%" PRIx64 " vgId:%d subkey:%s invalid offset type:%d", consumerId, vgId, req.subKey, + reqOffset.type); + terrno = TSDB_CODE_INVALID_PARA; + tDeleteMqDataRsp(&dataRsp); + return -1; + } + } + + tqDoSendDataRsp(&pMsg->info, &dataRsp, req.epoch, req.consumerId, TMQ_MSG_TYPE__WALINFO_RSP, sver, ever); + return 0; +} + int32_t tqProcessDeleteSubReq(STQ* pTq, int64_t sversion, char* msg, int32_t msgLen) { SMqVDeleteReq* pReq = (SMqVDeleteReq*)msg; diff --git a/source/dnode/vnode/src/tq/tqPush.c b/source/dnode/vnode/src/tq/tqPush.c index 950c5ea96b..46414e198b 100644 --- a/source/dnode/vnode/src/tq/tqPush.c +++ b/source/dnode/vnode/src/tq/tqPush.c @@ -16,6 +16,261 @@ #include "tq.h" #include "vnd.h" +<<<<<<< HEAD +======= +#if 0 +void tqTmrRspFunc(void* param, void* tmrId) { + STqHandle* pHandle = (STqHandle*)param; + atomic_store_8(&pHandle->pushHandle.tmrStopped, 1); +} + +static int32_t tqLoopExecFromQueue(STQ* pTq, STqHandle* pHandle, SStreamDataSubmit** ppSubmit, SMqDataRsp* pRsp) { + SStreamDataSubmit* pSubmit = *ppSubmit; + while (pSubmit != NULL) { + if (tqLogScanExec(pTq, &pHandle->execHandle, pSubmit->data, pRsp, 0) < 0) { + } + // update processed + atomic_store_64(&pHandle->pushHandle.processedVer, pSubmit->ver); + streamQueueProcessSuccess(&pHandle->pushHandle.inputQ); + streamDataSubmitDestroy(pSubmit); + if (pRsp->blockNum > 0) { + *ppSubmit = pSubmit; + return 0; + } else { + pSubmit = streamQueueNextItem(&pHandle->pushHandle.inputQ); + } + } + *ppSubmit = pSubmit; + return -1; +} + +int32_t tqExecFromInputQ(STQ* pTq, STqHandle* pHandle) { + SMqDataRsp rsp = {0}; + // 1. guard and set status executing + int8_t execStatus = atomic_val_compare_exchange_8(&pHandle->pushHandle.execStatus, TASK_EXEC_STATUS__IDLE, + TASK_EXEC_STATUS__EXECUTING); + if (execStatus == TASK_EXEC_STATUS__IDLE) { + SStreamDataSubmit* pSubmit = NULL; + // 2. check processedVer + // 2.1. if not missed, get msg from queue + // 2.2. if missed, scan wal + pSubmit = streamQueueNextItem(&pHandle->pushHandle.inputQ); + while (pHandle->pushHandle.processedVer <= pSubmit->ver) { + // read from wal + } + while (pHandle->pushHandle.processedVer > pSubmit->ver + 1) { + streamQueueProcessSuccess(&pHandle->pushHandle.inputQ); + streamDataSubmitDestroy(pSubmit); + pSubmit = streamQueueNextItem(&pHandle->pushHandle.inputQ); + if (pSubmit == NULL) break; + } + // 3. exec, after each success, update processed ver + // first run + if (tqLoopExecFromQueue(pTq, pHandle, &pSubmit, &rsp) == 0) { + goto SEND_RSP; + } + // set exec status closing + atomic_store_8(&pHandle->pushHandle.execStatus, TASK_EXEC_STATUS__CLOSING); + // second run + if (tqLoopExecFromQueue(pTq, pHandle, &pSubmit, &rsp) == 0) { + goto SEND_RSP; + } + // set exec status idle + atomic_store_8(&pHandle->pushHandle.execStatus, TASK_EXEC_STATUS__IDLE); + } +SEND_RSP: + // 4. if get result + // 4.1 set exec input status blocked and exec status idle + atomic_store_8(&pHandle->pushHandle.execStatus, TASK_EXEC_STATUS__IDLE); + // 4.2 rpc send + rsp.rspOffset = pHandle->pushHandle.processedVer; + /*if (tqSendPollRsp(pTq, pMsg, pReq, &rsp) < 0) {*/ + /*return -1;*/ + /*}*/ + // 4.3 clear rpc info + memset(&pHandle->pushHandle.rpcInfo, 0, sizeof(SRpcHandleInfo)); + return 0; +} + +int32_t tqOpenPushHandle(STQ* pTq, STqHandle* pHandle) { + memset(&pHandle->pushHandle, 0, sizeof(STqPushHandle)); + pHandle->pushHandle.inputQ.queue = taosOpenQueue(); + pHandle->pushHandle.inputQ.qall = taosAllocateQall(); + if (pHandle->pushHandle.inputQ.queue == NULL || pHandle->pushHandle.inputQ.qall == NULL) { + if (pHandle->pushHandle.inputQ.queue) { + taosCloseQueue(pHandle->pushHandle.inputQ.queue); + } + if (pHandle->pushHandle.inputQ.qall) { + taosFreeQall(pHandle->pushHandle.inputQ.qall); + } + return -1; + } + return 0; +} + +int32_t tqPreparePush(STQ* pTq, STqHandle* pHandle, int64_t reqId, const SRpcHandleInfo* pInfo, int64_t processedVer, + int64_t timeout) { + memcpy(&pHandle->pushHandle.rpcInfo, pInfo, sizeof(SRpcHandleInfo)); + atomic_store_64(&pHandle->pushHandle.reqId, reqId); + atomic_store_64(&pHandle->pushHandle.processedVer, processedVer); + atomic_store_8(&pHandle->pushHandle.inputStatus, TASK_INPUT_STATUS__NORMAL); + atomic_store_8(&pHandle->pushHandle.tmrStopped, 0); + taosTmrReset(tqTmrRspFunc, (int32_t)timeout, pHandle, tqMgmt.timer, &pHandle->pushHandle.timerId); + return 0; +} + +int32_t tqEnqueue(STqHandle* pHandle, SStreamDataSubmit* pSubmit) { + int8_t inputStatus = atomic_load_8(&pHandle->pushHandle.inputStatus); + if (inputStatus == TASK_INPUT_STATUS__NORMAL) { + SStreamDataSubmit* pSubmitClone = streamSubmitBlockClone(pSubmit); + if (pSubmitClone == NULL) { + return -1; + } + taosWriteQitem(pHandle->pushHandle.inputQ.queue, pSubmitClone); + return 0; + } + return -1; +} + +int32_t tqSendExecReq(STQ* pTq, STqHandle* pHandle) { + // + return 0; +} + +int32_t tqPushMsgNew(STQ* pTq, void* msg, int32_t msgLen, tmsg_t msgType, int64_t ver, SRpcHandleInfo handleInfo) { + if (msgType != TDMT_VND_SUBMIT) return 0; + void* pIter = NULL; + STqHandle* pHandle = NULL; + SSubmitReq* pReq = (SSubmitReq*)msg; + int32_t workerId = 4; + int64_t fetchOffset = ver; + + while (1) { + pIter = taosHashIterate(pTq->pushMgr, pIter); + if (pIter == NULL) break; + pHandle = *(STqHandle**)pIter; + + taosWLockLatch(&pHandle->pushHandle.lock); + + SMqDataRsp rsp = {0}; + rsp.reqOffset = pHandle->pushHandle.reqOffset; + rsp.blockData = taosArrayInit(0, sizeof(void*)); + rsp.blockDataLen = taosArrayInit(0, sizeof(int32_t)); + + if (msgType == TDMT_VND_SUBMIT) { + tqLogScanExec(pTq, &pHandle->execHandle, pReq, &rsp, workerId); + } else { + tqError("tq push unexpected msg type %d", msgType); + } + + if (rsp.blockNum == 0) { + taosWUnLockLatch(&pHandle->pushHandle.lock); + continue; + } + + rsp.rspOffset = fetchOffset; + + int32_t tlen = sizeof(SMqRspHead) + tEncodeSMqDataBlkRsp(NULL, &rsp); + void* buf = rpcMallocCont(tlen); + if (buf == NULL) { + // todo free + return -1; + } + + ((SMqRspHead*)buf)->mqMsgType = TMQ_MSG_TYPE__POLL_RSP; + ((SMqRspHead*)buf)->epoch = pHandle->pushHandle.epoch; + ((SMqRspHead*)buf)->consumerId = pHandle->pushHandle.consumerId; + + void* abuf = POINTER_SHIFT(buf, sizeof(SMqRspHead)); + tEncodeSMqDataBlkRsp(&abuf, &rsp); + + SRpcMsg resp = { + .info = pHandle->pushHandle.rpcInfo, + .pCont = buf, + .contLen = tlen, + .code = 0, + }; + tmsgSendRsp(&resp); + + memset(&pHandle->pushHandle.rpcInfo, 0, sizeof(SRpcHandleInfo)); + taosWUnLockLatch(&pHandle->pushHandle.lock); + + tqDebug("vgId:%d offset %" PRId64 " from consumer:%" PRId64 ", (epoch %d) send rsp, block num: %d, req:%" PRId64 ", rsp:%" PRId64, + TD_VID(pTq->pVnode), fetchOffset, pHandle->pushHandle.consumerId, pHandle->pushHandle.epoch, rsp.blockNum, + rsp.reqOffset, rsp.rspOffset); + + // TODO destroy + taosArrayDestroy(rsp.blockData); + taosArrayDestroy(rsp.blockDataLen); + } + + return 0; +} +#endif + +typedef struct { + void* pKey; + int64_t keyLen; +} SItem; + +static void recordPushedEntry(SArray* cachedKey, void* pIter); +static void doRemovePushedEntry(SArray* pCachedKeys, STQ* pTq); + +static void freeItem(void* param) { + SItem* p = (SItem*)param; + taosMemoryFree(p->pKey); +} + +static void doPushDataForEntry(void* pIter, STqExecHandle* pExec, STQ* pTq, int64_t ver, int32_t vgId, char* pData, + int32_t dataLen, SArray* pCachedKey) { + STqPushEntry* pPushEntry = *(STqPushEntry**)pIter; + + SMqDataRsp* pRsp = pPushEntry->pDataRsp; + if (pRsp->reqOffset.version >= ver) { + tqDebug("vgId:%d, push entry req version %" PRId64 ", while push version %" PRId64 ", skip", vgId, + pRsp->reqOffset.version, ver); + return; + } + + qTaskInfo_t pTaskInfo = pExec->task; + + // prepare scan mem data + SPackedData submit = {.msgStr = pData, .msgLen = dataLen, .ver = ver}; + + if (qStreamSetScanMemData(pTaskInfo, submit) != 0) { + return; + } + qStreamSetOpen(pTaskInfo); + // here start to scan submit block to extract the subscribed data + int32_t totalRows = 0; + + while (1) { + SSDataBlock* pDataBlock = NULL; + uint64_t ts = 0; + if (qExecTask(pTaskInfo, &pDataBlock, &ts) < 0) { + tqDebug("vgId:%d, tq exec error since %s", vgId, terrstr()); + } + + if (pDataBlock == NULL) { + break; + } + + tqAddBlockDataToRsp(pDataBlock, pRsp, pExec->numOfCols, pTq->pVnode->config.tsdbCfg.precision); + pRsp->blockNum++; + totalRows += pDataBlock->info.rows; + } + + tqDebug("vgId:%d, tq handle push, subkey:%s, block num:%d, rows:%d", vgId, pPushEntry->subKey, pRsp->blockNum, + totalRows); + + if (pRsp->blockNum > 0) { + tqOffsetResetToLog(&pRsp->rspOffset, ver); + tqPushDataRsp(pPushEntry, vgId); + recordPushedEntry(pCachedKey, pIter); + } +} + +>>>>>>> enh/3.0 int32_t tqPushMsg(STQ* pTq, void* msg, int32_t msgLen, tmsg_t msgType, int64_t ver) { if (msgType == TDMT_VND_SUBMIT) { @@ -60,11 +315,36 @@ int32_t tqRegisterPushHandle(STQ* pTq, void* handle, SRpcMsg* pMsg) { pHandle->msg->pCont = tmp; } +<<<<<<< HEAD memcpy(pHandle->msg->pCont, pMsg->pCont, pMsg->contLen); pHandle->msg->contLen = pMsg->contLen; int32_t ret = taosHashPut(pTq->pPushMgr, pHandle->subKey, strlen(pHandle->subKey), &pHandle, POINTER_BYTES); tqDebug("vgId:%d data is over, ret:%d, consumerId:0x%" PRIx64 ", register to pHandle:%p, pCont:%p, len:%d", vgId, ret, pHandle->consumerId, pHandle, pHandle->msg->pCont, pHandle->msg->contLen); +======= + pPushEntry->pHandle = pHandle; + pPushEntry->info = pRpcMsg->info; + memcpy(pPushEntry->subKey, pTqHandle->subKey, TSDB_SUBSCRIBE_KEY_LEN); + + if (type == TMQ_MSG_TYPE__TAOSX_RSP) { + pPushEntry->pDataRsp = taosMemoryCalloc(1, sizeof(STaosxRsp)); + memcpy(pPushEntry->pDataRsp, pDataRsp, sizeof(STaosxRsp)); + } else if (type == TMQ_MSG_TYPE__POLL_RSP) { + pPushEntry->pDataRsp = taosMemoryCalloc(1, sizeof(SMqDataRsp)); + memcpy(pPushEntry->pDataRsp, pDataRsp, sizeof(SMqDataRsp)); + } + + SMqRspHead* pHead = &pPushEntry->pDataRsp->head; + + pHead->consumerId = consumerId; + pHead->epoch = pRequest->epoch; + pHead->mqMsgType = type; + + taosHashPut(pTq->pPushMgr, pTqHandle->subKey, strlen(pTqHandle->subKey), &pPushEntry, sizeof(void*)); + + tqDebug("tmq poll: consumer:0x%" PRIx64 ", subkey %s offset:%" PRId64 ", vgId:%d save handle to push mgr, total:%d", + consumerId, pTqHandle->subKey, pDataRsp->reqOffset.version, vgId, taosHashGetSize(pTq->pPushMgr)); +>>>>>>> enh/3.0 return 0; } @@ -78,9 +358,17 @@ int32_t tqUnregisterPushHandle(STQ* pTq, void *handle) { if(pHandle->msg != NULL) { tqPushDataRsp(pTq, pHandle); +<<<<<<< HEAD rpcFreeCont(pHandle->msg->pCont); taosMemoryFree(pHandle->msg); pHandle->msg = NULL; +======= + if (rspConsumer) { // rsp the old consumer with empty block. + tqPushDataRsp(*pEntry, vgId); + } + + taosHashRemove(pTq->pPushMgr, pKey, keyLen); +>>>>>>> enh/3.0 } return 0; diff --git a/source/dnode/vnode/src/tq/tqRestore.c b/source/dnode/vnode/src/tq/tqRestore.c index 800bcf2469..82c31f2f57 100644 --- a/source/dnode/vnode/src/tq/tqRestore.c +++ b/source/dnode/vnode/src/tq/tqRestore.c @@ -57,8 +57,12 @@ int32_t tqStreamTasksScanWal(STQ* pTq) { return 0; } +<<<<<<< HEAD static SArray* extractTaskIdList(SStreamMeta* pStreamMeta, int32_t numOfTasks) { SArray* pTaskIdList = taosArrayInit(numOfTasks, sizeof(int32_t)); +======= +int32_t streamTaskReplayWal(SStreamMeta* pStreamMeta, STqOffsetStore* pOffsetStore, bool* pScanIdle) { +>>>>>>> enh/3.0 void* pIter = NULL; taosWLockLatch(&pStreamMeta->lock); diff --git a/source/dnode/vnode/src/tq/tqUtil.c b/source/dnode/vnode/src/tq/tqUtil.c index a66d63a910..5dd14f69e7 100644 --- a/source/dnode/vnode/src/tq/tqUtil.c +++ b/source/dnode/vnode/src/tq/tqUtil.c @@ -17,7 +17,7 @@ #define IS_OFFSET_RESET_TYPE(_t) ((_t) < 0) -static int32_t tqSendMetaPollRsp(STQ* pTq, const SRpcMsg* pMsg, const SMqPollReq* pReq, const SMqMetaRsp* pRsp); +static int32_t tqSendMetaPollRsp(STqHandle* pHandle, const SRpcMsg* pMsg, const SMqPollReq* pReq, const SMqMetaRsp* pRsp, int32_t vgId); char* createStreamTaskIdStr(int64_t streamId, int32_t taskId) { char buf[128] = {0}; @@ -40,7 +40,80 @@ int32_t tqAddInputBlockNLaunchTask(SStreamTask* pTask, SStreamQueueItem* pQueueI return TSDB_CODE_SUCCESS; } +<<<<<<< HEAD static int32_t tqInitDataRsp(SMqDataRsp* pRsp, const SMqPollReq* pReq, int8_t subType) { +======= +void initOffsetForAllRestoreTasks(STQ* pTq) { + void* pIter = NULL; + + while(1) { + pIter = taosHashIterate(pTq->pStreamMeta->pTasks, pIter); + if (pIter == NULL) { + break; + } + + SStreamTask* pTask = *(SStreamTask**)pIter; + if (pTask->taskLevel != TASK_LEVEL__SOURCE) { + continue; + } + + if (pTask->status.taskStatus == TASK_STATUS__RECOVER_PREPARE || pTask->status.taskStatus == TASK_STATUS__WAIT_DOWNSTREAM) { + tqDebug("s-task:%s skip push data, since not ready, status %d", pTask->id.idStr, pTask->status.taskStatus); + continue; + } + + char key[128] = {0}; + createStreamTaskOffsetKey(key, pTask->id.streamId, pTask->id.taskId); + + STqOffset* pOffset = tqOffsetRead(pTq->pOffsetStore, key); + if (pOffset == NULL) { + doSaveTaskOffset(pTq->pOffsetStore, key, pTask->chkInfo.version); + } + } +} + +void saveOffsetForAllTasks(STQ* pTq, int64_t ver) { + void* pIter = NULL; + + while(1) { + pIter = taosHashIterate(pTq->pStreamMeta->pTasks, pIter); + if (pIter == NULL) { + break; + } + + SStreamTask* pTask = *(SStreamTask**)pIter; + if (pTask->taskLevel != TASK_LEVEL__SOURCE) { + continue; + } + + if (pTask->status.taskStatus == TASK_STATUS__RECOVER_PREPARE || pTask->status.taskStatus == TASK_STATUS__WAIT_DOWNSTREAM) { + tqDebug("s-task:%s skip push data, not ready for processing, status %d", pTask->id.idStr, + pTask->status.taskStatus); + continue; + } + + char key[128] = {0}; + createStreamTaskOffsetKey(key, pTask->id.streamId, pTask->id.taskId); + + STqOffset* pOffset = tqOffsetRead(pTq->pOffsetStore, key); + if (pOffset == NULL) { + doSaveTaskOffset(pTq->pOffsetStore, key, ver); + } + } +} + +void doSaveTaskOffset(STqOffsetStore* pOffsetStore, const char* pKey, int64_t ver) { + STqOffset offset = {0}; + tqOffsetResetToLog(&offset.val, ver); + + tstrncpy(offset.subKey, pKey, tListLen(offset.subKey)); + + // keep the offset info in the offset store + tqOffsetWrite(pOffsetStore, &offset); +} + +int32_t tqInitDataRsp(SMqDataRsp* pRsp, const SMqPollReq* pReq) { +>>>>>>> enh/3.0 pRsp->reqOffset = pReq->reqOffset; pRsp->blockData = taosArrayInit(0, sizeof(void*)); @@ -130,13 +203,13 @@ static int32_t extractResetOffsetVal(STqOffsetVal* pOffsetVal, STQ* pTq, STqHand } else if (reqOffset.type == TMQ_OFFSET__RESET_LATEST) { if (pHandle->execHandle.subType == TOPIC_SUB_TYPE__COLUMN) { SMqDataRsp dataRsp = {0}; - tqInitDataRsp(&dataRsp, pRequest, pHandle->execHandle.subType); + tqInitDataRsp(&dataRsp, pRequest); tqOffsetResetToLog(&dataRsp.rspOffset, walGetLastVer(pTq->pVnode->pWal)); tqDebug("tmq poll: consumer:0x%" PRIx64 ", subkey %s, vgId:%d, (latest) offset reset to %" PRId64, consumerId, pHandle->subKey, vgId, dataRsp.rspOffset.version); - int32_t code = tqSendDataRsp(pTq, pMsg, pRequest, &dataRsp, TMQ_MSG_TYPE__POLL_RSP); - tDeleteSMqDataRsp(&dataRsp); + int32_t code = tqSendDataRsp(pHandle, pMsg, pRequest, &dataRsp, TMQ_MSG_TYPE__POLL_RSP, vgId); + tDeleteMqDataRsp(&dataRsp); *pBlockReturned = true; return code; @@ -144,7 +217,7 @@ static int32_t extractResetOffsetVal(STqOffsetVal* pOffsetVal, STQ* pTq, STqHand STaosxRsp taosxRsp = {0}; tqInitTaosxRsp(&taosxRsp, pRequest); tqOffsetResetToLog(&taosxRsp.rspOffset, walGetLastVer(pTq->pVnode->pWal)); - int32_t code = tqSendDataRsp(pTq, pMsg, pRequest, (SMqDataRsp*)&taosxRsp, TMQ_MSG_TYPE__TAOSX_RSP); + int32_t code = tqSendDataRsp(pHandle, pMsg, pRequest, (SMqDataRsp*)&taosxRsp, TMQ_MSG_TYPE__TAOSX_RSP, vgId); tDeleteSTaosxRsp(&taosxRsp); *pBlockReturned = true; @@ -163,11 +236,13 @@ static int32_t extractResetOffsetVal(STqOffsetVal* pOffsetVal, STQ* pTq, STqHand static int32_t extractDataAndRspForNormalSubscribe(STQ* pTq, STqHandle* pHandle, const SMqPollReq* pRequest, SRpcMsg* pMsg, STqOffsetVal* pOffset) { + char buf[80] = {0}; uint64_t consumerId = pRequest->consumerId; int32_t vgId = TD_VID(pTq->pVnode); int code = 0; SMqDataRsp dataRsp = {0}; +<<<<<<< HEAD tqInitDataRsp(&dataRsp, pRequest, pHandle->execHandle.subType); qTaskInfo_t task = pHandle->execHandle.task; if(qTaskIsExecuting(task)){ @@ -207,10 +282,40 @@ static int32_t extractDataAndRspForNormalSubscribe(STQ* pTq, STqHandle* pHandle, // taosWUnLockLatch(&pTq->lock); tDeleteSMqDataRsp(&dataRsp); } +======= + tqInitDataRsp(&dataRsp, pRequest); + + // lock + taosWLockLatch(&pTq->lock); + + qSetTaskId(pHandle->execHandle.task, consumerId, pRequest->reqId); + + int code = tqScanData(pTq, pHandle, &dataRsp, pOffset); + if (code == 0) { + + // till now, all data has been transferred to consumer, new data needs to push client once arrived. + if (dataRsp.blockNum == 0 && dataRsp.reqOffset.type == TMQ_OFFSET__LOG && + dataRsp.reqOffset.version == dataRsp.rspOffset.version && pHandle->consumerId == pRequest->consumerId) { + code = tqRegisterPushHandle(pTq, pHandle, pRequest, pMsg, &dataRsp, TMQ_MSG_TYPE__POLL_RSP); + taosWUnLockLatch(&pTq->lock); + return code; + } + + // NOTE: this pHandle->consumerId may have been changed already. + code = tqSendDataRsp(pHandle, pMsg, pRequest, (SMqDataRsp*)&dataRsp, TMQ_MSG_TYPE__POLL_RSP, vgId); + } + + tFormatOffset(buf, 80, &dataRsp.rspOffset); + tqDebug("tmq poll: consumer:0x%" PRIx64 ", subkey %s, vgId:%d, rsp block:%d, rsp offset type:%s, reqId:0x%" PRIx64 + " code:%d", + consumerId, pHandle->subKey, vgId, dataRsp.blockNum, buf, pRequest->reqId, code); + taosWUnLockLatch(&pTq->lock); + tDeleteMqDataRsp(&dataRsp); + +>>>>>>> enh/3.0 return code; } - static int32_t extractDataAndRspForDbStbSubscribe(STQ* pTq, STqHandle* pHandle, const SMqPollReq* pRequest, SRpcMsg* pMsg, STqOffsetVal *offset) { int code = 0; int32_t vgId = TD_VID(pTq->pVnode); @@ -232,7 +337,7 @@ static int32_t extractDataAndRspForDbStbSubscribe(STQ* pTq, STqHandle* pHandle, } if (metaRsp.metaRspLen > 0) { - code = tqSendMetaPollRsp(pTq, pMsg, pRequest, &metaRsp); + code = tqSendMetaPollRsp(pHandle, pMsg, pRequest, &metaRsp, vgId); tqDebug("tmq poll: consumer:0x%" PRIx64 " subkey:%s vgId:%d, send meta offset type:%d,uid:%" PRId64 ",ts:%" PRId64, pRequest->consumerId, pHandle->subKey, vgId, metaRsp.rspOffset.type, metaRsp.rspOffset.uid, metaRsp.rspOffset.ts); taosMemoryFree(metaRsp.metaRsp); @@ -243,7 +348,7 @@ static int32_t extractDataAndRspForDbStbSubscribe(STQ* pTq, STqHandle* pHandle, tqDebug("taosx poll: consumer:0x%" PRIx64 " subkey:%s vgId:%d, send data blockNum:%d, offset type:%d,uid:%" PRId64 ",ts:%" PRId64,pRequest->consumerId, pHandle->subKey, vgId, taosxRsp.blockNum, taosxRsp.rspOffset.type, taosxRsp.rspOffset.uid,taosxRsp.rspOffset.ts); if (taosxRsp.blockNum > 0) { - code = tqSendDataRsp(pTq, pMsg, pRequest, (SMqDataRsp*)&taosxRsp, TMQ_MSG_TYPE__TAOSX_RSP); + code = tqSendDataRsp(pHandle, pMsg, pRequest, (SMqDataRsp*)&taosxRsp, TMQ_MSG_TYPE__TAOSX_RSP, vgId); tDeleteSTaosxRsp(&taosxRsp); return code; }else { @@ -251,7 +356,6 @@ static int32_t extractDataAndRspForDbStbSubscribe(STQ* pTq, STqHandle* pHandle, } } - if (offset->type == TMQ_OFFSET__LOG) { verifyOffset(pHandle->pWalReader, offset); int64_t fetchVer = offset->version + 1; @@ -267,13 +371,14 @@ static int32_t extractDataAndRspForDbStbSubscribe(STQ* pTq, STqHandle* pHandle, int32_t savedEpoch = atomic_load_32(&pHandle->epoch); if (savedEpoch > pRequest->epoch) { tqWarn("tmq poll: consumer:0x%" PRIx64 " (epoch %d), subkey:%s vgId:%d offset %" PRId64 - ", found new consumer epoch %d, discard req epoch %d", pRequest->consumerId, pRequest->epoch, pHandle->subKey, vgId, fetchVer, savedEpoch, pRequest->epoch); + ", found new consumer epoch %d, discard req epoch %d", + pRequest->consumerId, pRequest->epoch, pHandle->subKey, vgId, fetchVer, savedEpoch, pRequest->epoch); break; } if (tqFetchLog(pTq, pHandle, &fetchVer, &pCkHead, pRequest->reqId) < 0) { tqOffsetResetToLog(&taosxRsp.rspOffset, fetchVer); - code = tqSendDataRsp(pTq, pMsg, pRequest, (SMqDataRsp*)&taosxRsp, TMQ_MSG_TYPE__TAOSX_RSP); + code = tqSendDataRsp(pHandle, pMsg, pRequest, (SMqDataRsp*)&taosxRsp, TMQ_MSG_TYPE__TAOSX_RSP, vgId); tDeleteSTaosxRsp(&taosxRsp); taosMemoryFreeClear(pCkHead); return code; @@ -287,7 +392,7 @@ static int32_t extractDataAndRspForDbStbSubscribe(STQ* pTq, STqHandle* pHandle, if (pHead->msgType != TDMT_VND_SUBMIT) { if(totalRows > 0) { tqOffsetResetToLog(&taosxRsp.rspOffset, fetchVer - 1); - code = tqSendDataRsp(pTq, pMsg, pRequest, (SMqDataRsp*)&taosxRsp, TMQ_MSG_TYPE__TAOSX_RSP); + code = tqSendDataRsp(pHandle, pMsg, pRequest, (SMqDataRsp*)&taosxRsp, TMQ_MSG_TYPE__TAOSX_RSP, vgId); tDeleteSTaosxRsp(&taosxRsp); taosMemoryFreeClear(pCkHead); return code; @@ -298,7 +403,7 @@ static int32_t extractDataAndRspForDbStbSubscribe(STQ* pTq, STqHandle* pHandle, metaRsp.resMsgType = pHead->msgType; metaRsp.metaRspLen = pHead->bodyLen; metaRsp.metaRsp = pHead->body; - if (tqSendMetaPollRsp(pTq, pMsg, pRequest, &metaRsp) < 0) { + if (tqSendMetaPollRsp(pHandle, pMsg, pRequest, &metaRsp, vgId) < 0) { code = -1; taosMemoryFreeClear(pCkHead); tDeleteSTaosxRsp(&taosxRsp); @@ -328,7 +433,7 @@ static int32_t extractDataAndRspForDbStbSubscribe(STQ* pTq, STqHandle* pHandle, if (totalRows >= 4096 || taosxRsp.createTableNum > 0) { tqOffsetResetToLog(&taosxRsp.rspOffset, fetchVer); - code = tqSendDataRsp(pTq, pMsg, pRequest, (SMqDataRsp*)&taosxRsp, TMQ_MSG_TYPE__TAOSX_RSP); + code = tqSendDataRsp(pHandle, pMsg, pRequest, (SMqDataRsp*)&taosxRsp, TMQ_MSG_TYPE__TAOSX_RSP, vgId); tDeleteSTaosxRsp(&taosxRsp); taosMemoryFreeClear(pCkHead); return code; @@ -375,10 +480,19 @@ int32_t tqExtractDataForMq(STQ* pTq, STqHandle* pHandle, const SMqPollReq* pRequ } } -int32_t tqSendMetaPollRsp(STQ* pTq, const SRpcMsg* pMsg, const SMqPollReq* pReq, const SMqMetaRsp* pRsp) { +static void initMqRspHead(SMqRspHead* pMsgHead, int32_t type, int32_t epoch, int64_t consumerId, int64_t sver, + int64_t ever) { + pMsgHead->consumerId = consumerId; + pMsgHead->epoch = epoch; + pMsgHead->mqMsgType = type; + pMsgHead->walsver = sver; + pMsgHead->walever = ever; +} + +int32_t tqSendMetaPollRsp(STqHandle* pHandle, const SRpcMsg* pMsg, const SMqPollReq* pReq, const SMqMetaRsp* pRsp, int32_t vgId) { int32_t len = 0; int32_t code = 0; - tEncodeSize(tEncodeSMqMetaRsp, pRsp, len, code); + tEncodeSize(tEncodeMqMetaRsp, pRsp, len, code); if (code < 0) { return -1; } @@ -388,27 +502,64 @@ int32_t tqSendMetaPollRsp(STQ* pTq, const SRpcMsg* pMsg, const SMqPollReq* pReq, return -1; } - ((SMqRspHead*)buf)->mqMsgType = TMQ_MSG_TYPE__POLL_META_RSP; - ((SMqRspHead*)buf)->epoch = pReq->epoch; - ((SMqRspHead*)buf)->consumerId = pReq->consumerId; + int64_t sver = 0, ever = 0; + walReaderValidVersionRange(pHandle->execHandle.pTqReader->pWalReader, &sver, &ever); + initMqRspHead(buf, TMQ_MSG_TYPE__POLL_META_RSP, pReq->epoch, pReq->consumerId, sver, ever); void* abuf = POINTER_SHIFT(buf, sizeof(SMqRspHead)); SEncoder encoder = {0}; tEncoderInit(&encoder, abuf, len); - tEncodeSMqMetaRsp(&encoder, pRsp); + tEncodeMqMetaRsp(&encoder, pRsp); tEncoderClear(&encoder); - SRpcMsg resp = { - .info = pMsg->info, - .pCont = buf, - .contLen = tlen, - .code = 0, - }; - tmsgSendRsp(&resp); + SRpcMsg resp = { .info = pMsg->info, .pCont = buf, .contLen = tlen, .code = 0 }; - tqDebug("vgId:%d, from consumer:0x%" PRIx64 " (epoch %d) send rsp, res msg type %d, offset type:%d", - TD_VID(pTq->pVnode), pReq->consumerId, pReq->epoch, pRsp->resMsgType, pRsp->rspOffset.type); + tmsgSendRsp(&resp); + tqDebug("vgId:%d, from consumer:0x%" PRIx64 " (epoch %d) send rsp, res msg type %d, offset type:%d", vgId, + pReq->consumerId, pReq->epoch, pRsp->resMsgType, pRsp->rspOffset.type); return 0; } + +int32_t tqDoSendDataRsp(const SRpcHandleInfo* pRpcHandleInfo, const SMqDataRsp* pRsp, int32_t epoch, int64_t consumerId, + int32_t type, int64_t sver, int64_t ever) { + int32_t len = 0; + int32_t code = 0; + + if (type == TMQ_MSG_TYPE__POLL_RSP || type == TMQ_MSG_TYPE__WALINFO_RSP) { + tEncodeSize(tEncodeMqDataRsp, pRsp, len, code); + } else if (type == TMQ_MSG_TYPE__TAOSX_RSP) { + tEncodeSize(tEncodeSTaosxRsp, (STaosxRsp*)pRsp, len, code); + } + + if (code < 0) { + return -1; + } + + int32_t tlen = sizeof(SMqRspHead) + len; + void* buf = rpcMallocCont(tlen); + if (buf == NULL) { + return -1; + } + + SMqRspHead* pHead = (SMqRspHead*)buf; + initMqRspHead(pHead, type, epoch, consumerId, sver, ever); + + void* abuf = POINTER_SHIFT(buf, sizeof(SMqRspHead)); + + SEncoder encoder = {0}; + tEncoderInit(&encoder, abuf, len); + + if (type == TMQ_MSG_TYPE__POLL_RSP || type == TMQ_MSG_TYPE__WALINFO_RSP) { + tEncodeMqDataRsp(&encoder, pRsp); + } else if (type == TMQ_MSG_TYPE__TAOSX_RSP) { + tEncodeSTaosxRsp(&encoder, (STaosxRsp*)pRsp); + } + + tEncoderClear(&encoder); + SRpcMsg rsp = { .info = *pRpcHandleInfo, .pCont = buf, .contLen = tlen, .code = 0 }; + + tmsgSendRsp(&rsp); + return 0; +} \ No newline at end of file diff --git a/source/dnode/vnode/src/tsdb/tsdbCache.c b/source/dnode/vnode/src/tsdb/tsdbCache.c index 65fc086f8d..23b9f3db12 100644 --- a/source/dnode/vnode/src/tsdb/tsdbCache.c +++ b/source/dnode/vnode/src/tsdb/tsdbCache.c @@ -12,7 +12,6 @@ * You should have received a copy of the GNU Affero General Public License * along with this program. If not, see . */ - #include "tsdb.h" static int32_t tsdbOpenBICache(STsdb *pTsdb) { @@ -47,6 +46,502 @@ static void tsdbCloseBICache(STsdb *pTsdb) { } } +#define ROCKS_KEY_LEN 64 + +static void tsdbGetRocksPath(STsdb *pTsdb, char *path) { + SVnode *pVnode = pTsdb->pVnode; + if (pVnode->pTfs) { + if (path) { + snprintf(path, TSDB_FILENAME_LEN, "%s%s%s%scache.rdb", tfsGetPrimaryPath(pTsdb->pVnode->pTfs), TD_DIRSEP, + pTsdb->path, TD_DIRSEP); + } + } else { + if (path) { + snprintf(path, TSDB_FILENAME_LEN, "%s%scache.rdb", pTsdb->path, TD_DIRSEP); + } + } +} + +static int32_t tsdbOpenRocksCache(STsdb *pTsdb) { + int32_t code = 0; + + rocksdb_options_t *options = rocksdb_options_create(); + if (NULL == options) { + code = TSDB_CODE_OUT_OF_MEMORY; + goto _err; + } + + rocksdb_options_set_create_if_missing(options, 1); + // rocksdb_options_set_inplace_update_support(options, 1); + // rocksdb_options_set_allow_concurrent_memtable_write(options, 0); + + rocksdb_writeoptions_t *writeoptions = rocksdb_writeoptions_create(); + if (NULL == writeoptions) { + code = TSDB_CODE_OUT_OF_MEMORY; + goto _err2; + } + // rocksdb_writeoptions_disable_WAL(writeoptions, 1); + + rocksdb_readoptions_t *readoptions = rocksdb_readoptions_create(); + if (NULL == readoptions) { + code = TSDB_CODE_OUT_OF_MEMORY; + goto _err2; + } + + char *err = NULL; + char cachePath[TSDB_FILENAME_LEN] = {0}; + tsdbGetRocksPath(pTsdb, cachePath); + + rocksdb_t *db = rocksdb_open(options, cachePath, &err); + if (NULL == db) { + code = -1; + goto _err3; + } + + rocksdb_flushoptions_t *flushoptions = rocksdb_flushoptions_create(); + if (NULL == flushoptions) { + code = TSDB_CODE_OUT_OF_MEMORY; + goto _err4; + } + + rocksdb_writebatch_t *writebatch = rocksdb_writebatch_create(); + + pTsdb->rCache.writebatch = writebatch; + pTsdb->rCache.options = options; + pTsdb->rCache.writeoptions = writeoptions; + pTsdb->rCache.readoptions = readoptions; + pTsdb->rCache.flushoptions = flushoptions; + pTsdb->rCache.db = db; + + taosThreadMutexInit(&pTsdb->rCache.rMutex, NULL); + + return code; + +_err4: + rocksdb_readoptions_destroy(readoptions); +_err3: + rocksdb_writeoptions_destroy(writeoptions); +_err2: + rocksdb_options_destroy(options); +_err: + return code; +} + +static void tsdbCloseRocksCache(STsdb *pTsdb) { + rocksdb_close(pTsdb->rCache.db); + rocksdb_flushoptions_destroy(pTsdb->rCache.flushoptions); + rocksdb_writebatch_destroy(pTsdb->rCache.writebatch); + rocksdb_readoptions_destroy(pTsdb->rCache.readoptions); + rocksdb_writeoptions_destroy(pTsdb->rCache.writeoptions); + rocksdb_options_destroy(pTsdb->rCache.options); + taosThreadMutexDestroy(&pTsdb->rCache.rMutex); +} + +int32_t tsdbCacheCommit(STsdb *pTsdb) { + int32_t code = 0; + char *err = NULL; + + rocksdb_flush(pTsdb->rCache.db, pTsdb->rCache.flushoptions, &err); + if (NULL != err) { + tsdbError("vgId:%d, %s failed at line %d since %s", TD_VID(pTsdb->pVnode), __func__, __LINE__, err); + rocksdb_free(err); + code = -1; + } + + return code; +} + +SLastCol *tsdbCacheDeserialize(char const *value) { + if (!value) { + return NULL; + } + + SLastCol *pLastCol = (SLastCol *)value; + SColVal *pColVal = &pLastCol->colVal; + if (IS_VAR_DATA_TYPE(pColVal->type)) { + if (pColVal->value.nData > 0) { + pColVal->value.pData = (char *)value + sizeof(*pLastCol); + } else { + pColVal->value.pData = NULL; + } + } + + return pLastCol; +} + +void tsdbCacheSerialize(SLastCol *pLastCol, char **value, size_t *size) { + SColVal *pColVal = &pLastCol->colVal; + size_t length = sizeof(*pLastCol); + if (IS_VAR_DATA_TYPE(pColVal->type)) { + length += pColVal->value.nData; + } + *value = taosMemoryMalloc(length); + + *(SLastCol *)(*value) = *pLastCol; + if (IS_VAR_DATA_TYPE(pColVal->type)) { + uint8_t *pVal = pColVal->value.pData; + pColVal->value.pData = *value + sizeof(*pLastCol); + if (pColVal->value.nData > 0) { + memcpy(pColVal->value.pData, pVal, pColVal->value.nData); + } else { + pColVal->value.pData = NULL; + } + } + *size = length; +} + +static SLastCol *tsdbCacheLookup(STsdb *pTsdb, tb_uid_t uid, int16_t cid, char const *lstring) { + SLastCol *pLastCol = NULL; + + char *err = NULL; + size_t vlen = 0; + char key[ROCKS_KEY_LEN]; + size_t klen = snprintf(key, ROCKS_KEY_LEN, "%" PRIi64 ":%" PRIi16 ":%s", uid, cid, lstring); + char *value = NULL; + value = rocksdb_get(pTsdb->rCache.db, pTsdb->rCache.readoptions, key, klen, &vlen, &err); + if (NULL != err) { + tsdbError("vgId:%d, %s failed at line %d since %s", TD_VID(pTsdb->pVnode), __func__, __LINE__, err); + rocksdb_free(err); + } + + pLastCol = tsdbCacheDeserialize(value); + + return pLastCol; +} + +int32_t tsdbCacheUpdate(STsdb *pTsdb, tb_uid_t suid, tb_uid_t uid, TSDBROW *pRow) { + int32_t code = 0; + + // 1, fetch schema + STSchema *pTSchema = NULL; + int32_t sver = TSDBROW_SVERSION(pRow); + code = metaGetTbTSchemaEx(pTsdb->pVnode->pMeta, suid, uid, sver, &pTSchema); + if (code != TSDB_CODE_SUCCESS) { + terrno = code; + return -1; + } + + // 2, iterate col values into array + SArray *aColVal = taosArrayInit(32, sizeof(SColVal)); + + STSDBRowIter iter = {0}; + tsdbRowIterOpen(&iter, pRow, pTSchema); + + for (SColVal *pColVal = tsdbRowIterNext(&iter); pColVal; pColVal = tsdbRowIterNext(&iter)) { + /* + if (IS_VAR_DATA_TYPE(pColVal->type)) { + uint8_t *pVal = pColVal->value.pData; + + pColVal->value.pData = NULL; + code = tRealloc(&pColVal->value.pData, pColVal->value.nData); + if (code) { + code = TSDB_CODE_OUT_OF_MEMORY; + goto _exit; + } + + if (pColVal->value.nData) { + memcpy(pColVal->value.pData, pVal, pColVal->value.nData); + } + } + */ + taosArrayPush(aColVal, pColVal); + } + + tsdbRowClose(&iter); + + // 3, build keys & multi get from rocks + int num_keys = TARRAY_SIZE(aColVal); + char **keys_list = taosMemoryCalloc(num_keys * 2, sizeof(char *)); + size_t *keys_list_sizes = taosMemoryCalloc(num_keys * 2, sizeof(size_t)); + for (int i = 0; i < num_keys; ++i) { + SColVal *pColVal = (SColVal *)taosArrayGet(aColVal, i); + int16_t cid = pColVal->cid; + + char *keys = taosMemoryCalloc(2, ROCKS_KEY_LEN); + int last_key_len = snprintf(keys, ROCKS_KEY_LEN, "%" PRIi64 ":%" PRIi16 ":last", uid, cid); + if (last_key_len >= ROCKS_KEY_LEN) { + tsdbError("vgId:%d, %s failed at line %d since %s", TD_VID(pTsdb->pVnode), __func__, __LINE__, tstrerror(code)); + } + int lr_key_len = snprintf(keys + ROCKS_KEY_LEN, ROCKS_KEY_LEN, "%" PRIi64 ":%" PRIi16 ":last_row", uid, cid); + if (lr_key_len >= ROCKS_KEY_LEN) { + tsdbError("vgId:%d, %s failed at line %d since %s", TD_VID(pTsdb->pVnode), __func__, __LINE__, tstrerror(code)); + } + keys_list[i] = keys; + keys_list[num_keys + i] = keys + ROCKS_KEY_LEN; + keys_list_sizes[i] = last_key_len; + keys_list_sizes[num_keys + i] = lr_key_len; + } + char **values_list = taosMemoryCalloc(num_keys * 2, sizeof(char *)); + size_t *values_list_sizes = taosMemoryCalloc(num_keys * 2, sizeof(size_t)); + char **errs = taosMemoryCalloc(num_keys * 2, sizeof(char *)); + taosThreadMutexLock(&pTsdb->rCache.rMutex); + rocksdb_multi_get(pTsdb->rCache.db, pTsdb->rCache.readoptions, num_keys * 2, (const char *const *)keys_list, + keys_list_sizes, values_list, values_list_sizes, errs); + for (int i = 0; i < num_keys; ++i) { + taosMemoryFree(keys_list[i]); + } + for (int i = 0; i < num_keys * 2; ++i) { + rocksdb_free(errs[i]); + } + taosMemoryFree(keys_list); + taosMemoryFree(keys_list_sizes); + taosMemoryFree(errs); + + TSKEY keyTs = TSDBROW_TS(pRow); + rocksdb_writebatch_t *wb = pTsdb->rCache.writebatch; + for (int i = 0; i < num_keys; ++i) { + SColVal *pColVal = (SColVal *)taosArrayGet(aColVal, i); + if (COL_VAL_IS_VALUE(pColVal)) { + SLastCol *pLastCol = tsdbCacheDeserialize(values_list[i]); + + if (NULL == pLastCol || pLastCol->ts <= keyTs) { + char *value = NULL; + size_t vlen = 0; + tsdbCacheSerialize(&(SLastCol){.ts = keyTs, .colVal = *pColVal}, &value, &vlen); + char key[ROCKS_KEY_LEN]; + size_t klen = snprintf(key, ROCKS_KEY_LEN, "%" PRIi64 ":%" PRIi16 ":last", uid, pColVal->cid); + rocksdb_writebatch_put(wb, key, klen, value, vlen); + taosMemoryFree(value); + } + } + + if (!COL_VAL_IS_NONE(pColVal)) { + SLastCol *pLastCol = tsdbCacheDeserialize(values_list[i + num_keys]); + + if (NULL == pLastCol || pLastCol->ts <= keyTs) { + char *value = NULL; + size_t vlen = 0; + tsdbCacheSerialize(&(SLastCol){.ts = keyTs, .colVal = *pColVal}, &value, &vlen); + char key[ROCKS_KEY_LEN]; + size_t klen = snprintf(key, ROCKS_KEY_LEN, "%" PRIi64 ":%" PRIi16 ":last_row", uid, pColVal->cid); + rocksdb_writebatch_put(wb, key, klen, value, vlen); + taosMemoryFree(value); + } + } + + rocksdb_free(values_list[i]); + rocksdb_free(values_list[i + num_keys]); + } + taosMemoryFree(values_list); + taosMemoryFree(values_list_sizes); + + char *err = NULL; + rocksdb_write(pTsdb->rCache.db, pTsdb->rCache.writeoptions, wb, &err); + if (NULL != err) { + tsdbError("vgId:%d, %s failed at line %d since %s", TD_VID(pTsdb->pVnode), __func__, __LINE__, err); + rocksdb_free(err); + } + taosThreadMutexUnlock(&pTsdb->rCache.rMutex); + rocksdb_writebatch_clear(wb); + +_exit: + taosArrayDestroy(aColVal); + taosMemoryFree(pTSchema); + return code; +} + +static int32_t mergeLastCid(tb_uid_t uid, STsdb *pTsdb, SArray **ppLastArray, SCacheRowsReader *pr, int16_t *aCols, + int nCols, int16_t *slotIds); + +static int32_t mergeLastRowCid(tb_uid_t uid, STsdb *pTsdb, SArray **ppLastArray, SCacheRowsReader *pr, int16_t *aCols, + int nCols, int16_t *slotIds); + +int32_t tsdbCacheGet(STsdb *pTsdb, tb_uid_t uid, SArray *pLastArray, SCacheRowsReader *pr, int32_t ltype) { + static char const *alstring[2] = {"last_row", "last"}; + char const *lstring = alstring[ltype]; + int32_t code = 0; + + SArray *pCidList = pr->pCidList; + int num_keys = TARRAY_SIZE(pCidList); + char **keys_list = taosMemoryCalloc(num_keys, sizeof(char *)); + size_t *keys_list_sizes = taosMemoryCalloc(num_keys, sizeof(size_t)); + for (int i = 0; i < num_keys; ++i) { + int16_t cid = *(int16_t *)taosArrayGet(pCidList, i); + + char *keys = taosMemoryCalloc(2, ROCKS_KEY_LEN); + int last_key_len = snprintf(keys, ROCKS_KEY_LEN, "%" PRIi64 ":%" PRIi16 ":%s", uid, cid, lstring); + if (last_key_len >= ROCKS_KEY_LEN) { + tsdbError("vgId:%d, %s failed at line %d since %s", TD_VID(pTsdb->pVnode), __func__, __LINE__, tstrerror(code)); + } + + keys_list[i] = keys; + keys_list_sizes[i] = last_key_len; + } + char **values_list = taosMemoryCalloc(num_keys, sizeof(char *)); + size_t *values_list_sizes = taosMemoryCalloc(num_keys, sizeof(size_t)); + char **errs = taosMemoryCalloc(num_keys, sizeof(char *)); + rocksdb_multi_get(pTsdb->rCache.db, pTsdb->rCache.readoptions, num_keys, (const char *const *)keys_list, + keys_list_sizes, values_list, values_list_sizes, errs); + for (int i = 0; i < num_keys; ++i) { + taosMemoryFree(keys_list[i]); + rocksdb_free(errs[i]); + } + taosMemoryFree(keys_list); + taosMemoryFree(keys_list_sizes); + taosMemoryFree(errs); + + for (int i = 0; i < num_keys; ++i) { + bool freeCol = true; + SArray *pTmpColArray = NULL; + SLastCol *pLastCol = tsdbCacheDeserialize(values_list[i]); + int16_t cid = *(int16_t *)taosArrayGet(pCidList, i); + SLastCol noneCol = {.ts = TSKEY_MIN, .colVal = COL_VAL_NONE(cid, pr->pSchema->columns[pr->pSlotIds[i]].type)}; + if (pLastCol) { + SColVal *pColVal = &pLastCol->colVal; + if (IS_VAR_DATA_TYPE(pColVal->type)) { + uint8_t *pVal = pColVal->value.pData; + pColVal->value.pData = taosMemoryMalloc(pColVal->value.nData); + if (pColVal->value.nData) { + memcpy(pColVal->value.pData, pVal, pColVal->value.nData); + } + } + } else { + taosThreadMutexLock(&pTsdb->rCache.rMutex); + + pLastCol = tsdbCacheLookup(pTsdb, uid, cid, lstring); + if (!pLastCol) { + // recalc: load from tsdb + int16_t aCols[1] = {cid}; + int16_t slotIds[1] = {pr->pSlotIds[i]}; + pTmpColArray = NULL; + + if (ltype) { + mergeLastCid(uid, pTsdb, &pTmpColArray, pr, aCols, 1, slotIds); + } else { + mergeLastRowCid(uid, pTsdb, &pTmpColArray, pr, aCols, 1, slotIds); + } + + if (pTmpColArray && TARRAY_SIZE(pTmpColArray) >= 1) { + pLastCol = taosArrayGet(pTmpColArray, 0); + freeCol = false; + } + + // still null, then make up a none col value + if (!pLastCol) { + pLastCol = &noneCol; + freeCol = false; + } + + // store result back to rocks cache + rocksdb_writebatch_t *wb = pTsdb->rCache.writebatch; + char *value = NULL; + size_t vlen = 0; + tsdbCacheSerialize(pLastCol, &value, &vlen); + char key[ROCKS_KEY_LEN]; + size_t klen = snprintf(key, ROCKS_KEY_LEN, "%" PRIi64 ":%" PRIi16 ":%s", uid, pLastCol->colVal.cid, lstring); + rocksdb_writebatch_put(wb, key, klen, value, vlen); + + taosMemoryFree(value); + + char *err = NULL; + rocksdb_write(pTsdb->rCache.db, pTsdb->rCache.writeoptions, wb, &err); + if (NULL != err) { + tsdbError("vgId:%d, %s failed at line %d since %s", TD_VID(pTsdb->pVnode), __func__, __LINE__, err); + rocksdb_free(err); + } + } + + taosThreadMutexUnlock(&pTsdb->rCache.rMutex); + } + + taosArrayPush(pLastArray, pLastCol); + + taosArrayDestroy(pTmpColArray); + if (freeCol) { + taosMemoryFree(pLastCol); + } + } + taosMemoryFree(values_list); + taosMemoryFree(values_list_sizes); + + return code; +} + +int32_t tsdbCacheDel(STsdb *pTsdb, tb_uid_t suid, tb_uid_t uid, TSKEY sKey, TSKEY eKey) { + int32_t code = 0; + // 1, fetch schema + STSchema *pTSchema = NULL; + int32_t sver = -1; + code = metaGetTbTSchemaEx(pTsdb->pVnode->pMeta, suid, uid, sver, &pTSchema); + if (code != TSDB_CODE_SUCCESS) { + terrno = code; + return -1; + } + + // 3, build keys & multi get from rocks + int num_keys = pTSchema->numOfCols; + char **keys_list = taosMemoryCalloc(num_keys * 2, sizeof(char *)); + size_t *keys_list_sizes = taosMemoryCalloc(num_keys * 2, sizeof(size_t)); + for (int i = 0; i < num_keys; ++i) { + int16_t cid = pTSchema->columns[i].colId; + + char *keys = taosMemoryCalloc(2, ROCKS_KEY_LEN); + int last_key_len = snprintf(keys, ROCKS_KEY_LEN, "%" PRIi64 ":%" PRIi16 ":last", uid, cid); + if (last_key_len >= ROCKS_KEY_LEN) { + tsdbError("vgId:%d, %s failed at line %d since %s", TD_VID(pTsdb->pVnode), __func__, __LINE__, tstrerror(code)); + } + int lr_key_len = snprintf(keys + ROCKS_KEY_LEN, ROCKS_KEY_LEN, "%" PRIi64 ":%" PRIi16 ":last_row", uid, cid); + if (lr_key_len >= ROCKS_KEY_LEN) { + tsdbError("vgId:%d, %s failed at line %d since %s", TD_VID(pTsdb->pVnode), __func__, __LINE__, tstrerror(code)); + } + keys_list[i] = keys; + keys_list[num_keys + i] = keys + ROCKS_KEY_LEN; + keys_list_sizes[i] = last_key_len; + keys_list_sizes[num_keys + i] = lr_key_len; + } + char **values_list = taosMemoryCalloc(num_keys * 2, sizeof(char *)); + size_t *values_list_sizes = taosMemoryCalloc(num_keys * 2, sizeof(size_t)); + char **errs = taosMemoryCalloc(num_keys * 2, sizeof(char *)); + taosThreadMutexLock(&pTsdb->rCache.rMutex); + rocksdb_multi_get(pTsdb->rCache.db, pTsdb->rCache.readoptions, num_keys * 2, (const char *const *)keys_list, + keys_list_sizes, values_list, values_list_sizes, errs); + for (int i = 0; i < num_keys; ++i) { + taosMemoryFree(keys_list[i]); + } + for (int i = 0; i < num_keys * 2; ++i) { + rocksdb_free(errs[i]); + } + taosMemoryFree(keys_list); + taosMemoryFree(keys_list_sizes); + taosMemoryFree(errs); + + rocksdb_writebatch_t *wb = pTsdb->rCache.writebatch; + for (int i = 0; i < num_keys; ++i) { + SLastCol *pLastCol = tsdbCacheDeserialize(values_list[i]); + if (NULL != pLastCol && (pLastCol->ts <= eKey && pLastCol->ts >= sKey)) { + char key[ROCKS_KEY_LEN]; + size_t klen = snprintf(key, ROCKS_KEY_LEN, "%" PRIi64 ":%" PRIi16 ":last", uid, pLastCol->colVal.cid); + rocksdb_writebatch_delete(wb, key, klen); + } + + pLastCol = tsdbCacheDeserialize(values_list[i + num_keys]); + if (NULL != pLastCol && (pLastCol->ts <= eKey && pLastCol->ts >= sKey)) { + char key[ROCKS_KEY_LEN]; + size_t klen = snprintf(key, ROCKS_KEY_LEN, "%" PRIi64 ":%" PRIi16 ":last_row", uid, pLastCol->colVal.cid); + rocksdb_writebatch_delete(wb, key, klen); + } + + rocksdb_free(values_list[i]); + rocksdb_free(values_list[i + num_keys]); + } + taosMemoryFree(values_list); + taosMemoryFree(values_list_sizes); + + char *err = NULL; + rocksdb_write(pTsdb->rCache.db, pTsdb->rCache.writeoptions, wb, &err); + if (NULL != err) { + tsdbError("vgId:%d, %s failed at line %d since %s", TD_VID(pTsdb->pVnode), __func__, __LINE__, err); + rocksdb_free(err); + } + taosThreadMutexUnlock(&pTsdb->rCache.rMutex); + rocksdb_writebatch_clear(wb); + +_exit: + taosMemoryFree(pTSchema); + + return code; +} + int32_t tsdbOpenCache(STsdb *pTsdb) { int32_t code = 0; SLRUCache *pCache = NULL; @@ -64,6 +559,12 @@ int32_t tsdbOpenCache(STsdb *pTsdb) { goto _err; } + code = tsdbOpenRocksCache(pTsdb); + if (code != TSDB_CODE_SUCCESS) { + code = TSDB_CODE_OUT_OF_MEMORY; + goto _err; + } + taosLRUCacheSetStrictCapacity(pCache, false); taosThreadMutexInit(&pTsdb->lruMutex, NULL); @@ -84,6 +585,7 @@ void tsdbCloseCache(STsdb *pTsdb) { } tsdbCloseBICache(pTsdb); + tsdbCloseRocksCache(pTsdb); } static void getTableCacheKey(tb_uid_t uid, int cacheType, char *key, int *len) { @@ -170,7 +672,7 @@ int32_t tsdbCacheDeleteLast(SLRUCache *pCache, tb_uid_t uid, TSKEY eKey) { return code; } - +/* int32_t tsdbCacheDelete(SLRUCache *pCache, tb_uid_t uid, TSKEY eKey) { int32_t code = 0; char key[32] = {0}; @@ -225,7 +727,7 @@ int32_t tsdbCacheDelete(SLRUCache *pCache, tb_uid_t uid, TSKEY eKey) { return code; } - +*/ int32_t tsdbCacheInsertLastrow(SLRUCache *pCache, STsdb *pTsdb, tb_uid_t uid, TSDBROW *row, bool dup) { int32_t code = 0; STSRow *cacheRow = NULL; @@ -1425,6 +1927,21 @@ static int32_t initLastColArray(STSchema *pTSchema, SArray **ppColArray) { return TSDB_CODE_SUCCESS; } +static int32_t initLastColArrayPartial(STSchema *pTSchema, SArray **ppColArray, int16_t *slotIds, int nCols) { + SArray *pColArray = taosArrayInit(nCols, sizeof(SLastCol)); + if (NULL == pColArray) { + return TSDB_CODE_OUT_OF_MEMORY; + } + + for (int32_t i = 0; i < nCols; ++i) { + int16_t slotId = slotIds[i]; + SLastCol col = {.ts = 0, .colVal = COL_VAL_NULL(pTSchema->columns[slotId].colId, pTSchema->columns[slotId].type)}; + taosArrayPush(pColArray, &col); + } + *ppColArray = pColArray; + return TSDB_CODE_SUCCESS; +} + static int32_t cloneTSchema(STSchema *pSrc, STSchema **ppDst) { int32_t len = sizeof(STSchema) + sizeof(STColumn) * pSrc->numOfCols; *ppDst = taosMemoryMalloc(len); @@ -1763,6 +2280,342 @@ _err: return code; } +static int32_t mergeLastCid(tb_uid_t uid, STsdb *pTsdb, SArray **ppLastArray, SCacheRowsReader *pr, int16_t *aCols, + int nCols, int16_t *slotIds) { + STSchema *pTSchema = pr->pSchema; // metaGetTbTSchema(pTsdb->pVnode->pMeta, uid, -1, 1); + int16_t nLastCol = nCols; + int16_t noneCol = 0; + bool setNoneCol = false; + bool hasRow = false; + bool ignoreEarlierTs = false; + SArray *pColArray = NULL; + SColVal *pColVal = &(SColVal){0}; + + int32_t code = initLastColArrayPartial(pTSchema, &pColArray, slotIds, nCols); + if (TSDB_CODE_SUCCESS != code) { + return code; + } + SArray *aColArray = taosArrayInit(nCols, sizeof(int16_t)); + if (NULL == aColArray) { + taosArrayDestroy(pColArray); + + return TSDB_CODE_OUT_OF_MEMORY; + } + + for (int i = 0; i < nCols; ++i) { + taosArrayPush(aColArray, &aCols[i]); + } + + TSKEY lastRowTs = TSKEY_MAX; + + CacheNextRowIter iter = {0}; + nextRowIterOpen(&iter, uid, pTsdb, pTSchema, pr->suid, pr->pLoadInfo, pr->pReadSnap, &pr->pDataFReader, + &pr->pDataFReaderLast, pr->lastTs); + + do { + TSDBROW *pRow = NULL; + nextRowIterGet(&iter, &pRow, &ignoreEarlierTs, true, TARRAY_DATA(aColArray), TARRAY_SIZE(aColArray)); + + if (!pRow) { + break; + } + + hasRow = true; + + int32_t sversion = TSDBROW_SVERSION(pRow); + if (sversion != -1) { + code = updateTSchema(sversion, pr, uid); + if (TSDB_CODE_SUCCESS != code) { + goto _err; + } + pTSchema = pr->pCurrSchema; + } + // int16_t nCol = pTSchema->numOfCols; + + TSKEY rowTs = TSDBROW_TS(pRow); + + if (lastRowTs == TSKEY_MAX) { + lastRowTs = rowTs; + + for (int16_t iCol = noneCol; iCol < nCols; ++iCol) { + if (iCol >= nLastCol) { + break; + } + SLastCol *pCol = taosArrayGet(pColArray, iCol); + if (pCol->colVal.cid != pTSchema->columns[slotIds[iCol]].colId) { + continue; + } + if (slotIds[iCol] == 0) { + STColumn *pTColumn = &pTSchema->columns[0]; + + *pColVal = COL_VAL_VALUE(pTColumn->colId, pTColumn->type, (SValue){.val = rowTs}); + taosArraySet(pColArray, 0, &(SLastCol){.ts = rowTs, .colVal = *pColVal}); + continue; + } + tsdbRowGetColVal(pRow, pTSchema, slotIds[iCol], pColVal); + + *pCol = (SLastCol){.ts = rowTs, .colVal = *pColVal}; + if (IS_VAR_DATA_TYPE(pColVal->type) && pColVal->value.nData > 0) { + pCol->colVal.value.pData = taosMemoryMalloc(pCol->colVal.value.nData); + if (pCol->colVal.value.pData == NULL) { + terrno = TSDB_CODE_OUT_OF_MEMORY; + code = TSDB_CODE_OUT_OF_MEMORY; + goto _err; + } + memcpy(pCol->colVal.value.pData, pColVal->value.pData, pColVal->value.nData); + } + + if (!COL_VAL_IS_VALUE(pColVal)) { + if (!setNoneCol) { + noneCol = iCol; + setNoneCol = true; + } + } else { + int32_t aColIndex = taosArraySearchIdx(aColArray, &pColVal->cid, compareInt16Val, TD_EQ); + if (aColIndex >= 0) { + taosArrayRemove(aColArray, aColIndex); + } + } + } + if (!setNoneCol) { + // done, goto return pColArray + break; + } else { + continue; + } + } + + // merge into pColArray + setNoneCol = false; + for (int16_t iCol = noneCol; iCol < nCols; ++iCol) { + if (iCol >= nLastCol) { + break; + } + // high version's column value + SLastCol *lastColVal = (SLastCol *)taosArrayGet(pColArray, iCol); + if (lastColVal->colVal.cid != pTSchema->columns[slotIds[iCol]].colId) { + continue; + } + SColVal *tColVal = &lastColVal->colVal; + + tsdbRowGetColVal(pRow, pTSchema, slotIds[iCol], pColVal); + if (!COL_VAL_IS_VALUE(tColVal) && COL_VAL_IS_VALUE(pColVal)) { + SLastCol lastCol = {.ts = rowTs, .colVal = *pColVal}; + if (IS_VAR_DATA_TYPE(pColVal->type) && pColVal->value.nData > 0) { + SLastCol *pLastCol = (SLastCol *)taosArrayGet(pColArray, iCol); + taosMemoryFree(pLastCol->colVal.value.pData); + + lastCol.colVal.value.pData = taosMemoryMalloc(lastCol.colVal.value.nData); + if (lastCol.colVal.value.pData == NULL) { + terrno = TSDB_CODE_OUT_OF_MEMORY; + code = TSDB_CODE_OUT_OF_MEMORY; + goto _err; + } + memcpy(lastCol.colVal.value.pData, pColVal->value.pData, pColVal->value.nData); + } + + taosArraySet(pColArray, iCol, &lastCol); + int32_t aColIndex = taosArraySearchIdx(aColArray, &lastCol.colVal.cid, compareInt16Val, TD_EQ); + taosArrayRemove(aColArray, aColIndex); + } else if (!COL_VAL_IS_VALUE(tColVal) && !COL_VAL_IS_VALUE(pColVal) && !setNoneCol) { + noneCol = iCol; + setNoneCol = true; + } + } + } while (setNoneCol); + + if (!hasRow) { + if (ignoreEarlierTs) { + taosArrayDestroy(pColArray); + pColArray = NULL; + } else { + taosArrayClear(pColArray); + } + } + *ppLastArray = pColArray; + + nextRowIterClose(&iter); + taosArrayDestroy(aColArray); + + return code; + +_err: + nextRowIterClose(&iter); + // taosMemoryFreeClear(pTSchema); + *ppLastArray = NULL; + taosArrayDestroy(pColArray); + taosArrayDestroy(aColArray); + return code; +} + +static int32_t mergeLastRowCid(tb_uid_t uid, STsdb *pTsdb, SArray **ppLastArray, SCacheRowsReader *pr, int16_t *aCols, + int nCols, int16_t *slotIds) { + STSchema *pTSchema = pr->pSchema; // metaGetTbTSchema(pTsdb->pVnode->pMeta, uid, -1, 1); + int16_t nLastCol = nCols; + int16_t noneCol = 0; + bool setNoneCol = false; + bool hasRow = false; + bool ignoreEarlierTs = false; + SArray *pColArray = NULL; + SColVal *pColVal = &(SColVal){0}; + + int32_t code = initLastColArrayPartial(pTSchema, &pColArray, slotIds, nCols); + if (TSDB_CODE_SUCCESS != code) { + return code; + } + SArray *aColArray = taosArrayInit(nCols, sizeof(int16_t)); + if (NULL == aColArray) { + taosArrayDestroy(pColArray); + + return TSDB_CODE_OUT_OF_MEMORY; + } + + for (int i = 0; i < nCols; ++i) { + taosArrayPush(aColArray, &aCols[i]); + } + + TSKEY lastRowTs = TSKEY_MAX; + + CacheNextRowIter iter = {0}; + nextRowIterOpen(&iter, uid, pTsdb, pTSchema, pr->suid, pr->pLoadInfo, pr->pReadSnap, &pr->pDataFReader, + &pr->pDataFReaderLast, pr->lastTs); + + do { + TSDBROW *pRow = NULL; + nextRowIterGet(&iter, &pRow, &ignoreEarlierTs, true, TARRAY_DATA(aColArray), TARRAY_SIZE(aColArray)); + + if (!pRow) { + break; + } + + hasRow = true; + + int32_t sversion = TSDBROW_SVERSION(pRow); + if (sversion != -1) { + code = updateTSchema(sversion, pr, uid); + if (TSDB_CODE_SUCCESS != code) { + goto _err; + } + pTSchema = pr->pCurrSchema; + } + // int16_t nCol = pTSchema->numOfCols; + + TSKEY rowTs = TSDBROW_TS(pRow); + + if (lastRowTs == TSKEY_MAX) { + lastRowTs = rowTs; + + for (int16_t iCol = noneCol; iCol < nCols; ++iCol) { + if (iCol >= nLastCol) { + break; + } + SLastCol *pCol = taosArrayGet(pColArray, iCol); + if (pCol->colVal.cid != pTSchema->columns[slotIds[iCol]].colId) { + continue; + } + if (slotIds[iCol] == 0) { + STColumn *pTColumn = &pTSchema->columns[0]; + + *pColVal = COL_VAL_VALUE(pTColumn->colId, pTColumn->type, (SValue){.val = rowTs}); + taosArraySet(pColArray, 0, &(SLastCol){.ts = rowTs, .colVal = *pColVal}); + continue; + } + tsdbRowGetColVal(pRow, pTSchema, slotIds[iCol], pColVal); + + *pCol = (SLastCol){.ts = rowTs, .colVal = *pColVal}; + if (IS_VAR_DATA_TYPE(pColVal->type) && pColVal->value.nData > 0) { + pCol->colVal.value.pData = taosMemoryMalloc(pCol->colVal.value.nData); + if (pCol->colVal.value.pData == NULL) { + terrno = TSDB_CODE_OUT_OF_MEMORY; + code = TSDB_CODE_OUT_OF_MEMORY; + goto _err; + } + memcpy(pCol->colVal.value.pData, pColVal->value.pData, pColVal->value.nData); + } + + if (COL_VAL_IS_NONE(pColVal)) { + if (!setNoneCol) { + noneCol = iCol; + setNoneCol = true; + } + } else { + int32_t aColIndex = taosArraySearchIdx(aColArray, &pColVal->cid, compareInt16Val, TD_EQ); + if (aColIndex >= 0) { + taosArrayRemove(aColArray, aColIndex); + } + } + } + if (!setNoneCol) { + // done, goto return pColArray + break; + } else { + continue; + } + } + + // merge into pColArray + setNoneCol = false; + for (int16_t iCol = noneCol; iCol < nCols; ++iCol) { + if (iCol >= nLastCol) { + break; + } + // high version's column value + SLastCol *lastColVal = (SLastCol *)taosArrayGet(pColArray, iCol); + if (lastColVal->colVal.cid != pTSchema->columns[slotIds[iCol]].colId) { + continue; + } + SColVal *tColVal = &lastColVal->colVal; + + tsdbRowGetColVal(pRow, pTSchema, slotIds[iCol], pColVal); + if (COL_VAL_IS_NONE(tColVal) && !COL_VAL_IS_NONE(pColVal)) { + SLastCol lastCol = {.ts = rowTs, .colVal = *pColVal}; + if (IS_VAR_DATA_TYPE(pColVal->type) && pColVal->value.nData > 0) { + SLastCol *pLastCol = (SLastCol *)taosArrayGet(pColArray, iCol); + taosMemoryFree(pLastCol->colVal.value.pData); + + lastCol.colVal.value.pData = taosMemoryMalloc(lastCol.colVal.value.nData); + if (lastCol.colVal.value.pData == NULL) { + terrno = TSDB_CODE_OUT_OF_MEMORY; + code = TSDB_CODE_OUT_OF_MEMORY; + goto _err; + } + memcpy(lastCol.colVal.value.pData, pColVal->value.pData, pColVal->value.nData); + } + + taosArraySet(pColArray, iCol, &lastCol); + int32_t aColIndex = taosArraySearchIdx(aColArray, &lastCol.colVal.cid, compareInt16Val, TD_EQ); + taosArrayRemove(aColArray, aColIndex); + } else if (COL_VAL_IS_NONE(tColVal) && !COL_VAL_IS_NONE(pColVal) && !setNoneCol) { + noneCol = iCol; + setNoneCol = true; + } + } + } while (setNoneCol); + + if (!hasRow) { + if (ignoreEarlierTs) { + taosArrayDestroy(pColArray); + pColArray = NULL; + } else { + taosArrayClear(pColArray); + } + } + *ppLastArray = pColArray; + + nextRowIterClose(&iter); + taosArrayDestroy(aColArray); + + return code; + +_err: + nextRowIterClose(&iter); + + *ppLastArray = NULL; + taosArrayDestroy(pColArray); + taosArrayDestroy(aColArray); + return code; +} + int32_t tsdbCacheGetLastrowH(SLRUCache *pCache, tb_uid_t uid, SCacheRowsReader *pr, LRUHandle **handle) { int32_t code = 0; char key[32] = {0}; diff --git a/source/dnode/vnode/src/tsdb/tsdbCacheRead.c b/source/dnode/vnode/src/tsdb/tsdbCacheRead.c index 061cbc17f3..53103e9fbb 100644 --- a/source/dnode/vnode/src/tsdb/tsdbCacheRead.c +++ b/source/dnode/vnode/src/tsdb/tsdbCacheRead.c @@ -21,47 +21,30 @@ #define HASTYPE(_type, _t) (((_type) & (_t)) == (_t)) static int32_t saveOneRow(SArray* pRow, SSDataBlock* pBlock, SCacheRowsReader* pReader, const int32_t* slotIds, - void** pRes, const char* idStr) { + const int32_t* dstSlotIds, void** pRes, const char* idStr) { int32_t numOfRows = pBlock->info.rows; + bool allNullRow = true; if (HASTYPE(pReader->type, CACHESCAN_RETRIEVE_LAST)) { - bool allNullRow = true; - for (int32_t i = 0; i < pReader->numOfCols; ++i) { - SColumnInfoData* pColInfoData = taosArrayGet(pBlock->pDataBlock, i); + SColumnInfoData* pColInfoData = taosArrayGet(pBlock->pDataBlock, dstSlotIds[i]); SFirstLastRes* p = (SFirstLastRes*)varDataVal(pRes[i]); + int32_t slotId = slotIds[i]; + SLastCol* pColVal = (SLastCol*)taosArrayGet(pRow, i); - if (slotIds[i] == -1) { // the primary timestamp - SLastCol* pColVal = (SLastCol*)taosArrayGet(pRow, 0); - p->ts = pColVal->ts; - p->bytes = TSDB_KEYSIZE; - *(int64_t*)p->buf = pColVal->ts; - allNullRow = false; - } else { - int32_t slotId = slotIds[i]; - // add check for null value, caused by the modification of table schema (new column added). - if (slotId >= taosArrayGetSize(pRow)) { - p->ts = 0; - p->isNull = true; - colDataSetNULL(pColInfoData, numOfRows); - continue; - } + p->ts = pColVal->ts; + p->isNull = !COL_VAL_IS_VALUE(&pColVal->colVal); + allNullRow = p->isNull & allNullRow; - SLastCol* pColVal = (SLastCol*)taosArrayGet(pRow, slotId); + if (!p->isNull) { + if (IS_VAR_DATA_TYPE(pColVal->colVal.type)) { + varDataSetLen(p->buf, pColVal->colVal.value.nData); - p->ts = pColVal->ts; - p->isNull = !COL_VAL_IS_VALUE(&pColVal->colVal); - allNullRow = p->isNull & allNullRow; - - if (!p->isNull) { - if (IS_VAR_DATA_TYPE(pColVal->colVal.type)) { - varDataSetLen(p->buf, pColVal->colVal.value.nData); - memcpy(varDataVal(p->buf), pColVal->colVal.value.pData, pColVal->colVal.value.nData); - p->bytes = pColVal->colVal.value.nData + VARSTR_HEADER_SIZE; // binary needs to plus the header size - } else { - memcpy(p->buf, &pColVal->colVal.value, pReader->pSchema->columns[slotId].bytes); - p->bytes = pReader->pSchema->columns[slotId].bytes; - } + memcpy(varDataVal(p->buf), pColVal->colVal.value.pData, pColVal->colVal.value.nData); + p->bytes = pColVal->colVal.value.nData + VARSTR_HEADER_SIZE; // binary needs to plus the header size + } else { + memcpy(p->buf, &pColVal->colVal.value, pReader->pSchema->columns[slotId].bytes); + p->bytes = pReader->pSchema->columns[slotId].bytes; } } @@ -74,36 +57,31 @@ static int32_t saveOneRow(SArray* pRow, SSDataBlock* pBlock, SCacheRowsReader* p pBlock->info.rows += allNullRow ? 0 : 1; } else if (HASTYPE(pReader->type, CACHESCAN_RETRIEVE_LAST_ROW)) { for (int32_t i = 0; i < pReader->numOfCols; ++i) { - SColumnInfoData* pColInfoData = taosArrayGet(pBlock->pDataBlock, i); + SColumnInfoData* pColInfoData = taosArrayGet(pBlock->pDataBlock, dstSlotIds[i]); - if (slotIds[i] == -1) { - SLastCol* pColVal = (SLastCol*)taosArrayGet(pRow, 0); - colDataSetVal(pColInfoData, numOfRows, (const char*)&pColVal->ts, false); - } else { - int32_t slotId = slotIds[i]; - // add check for null value, caused by the modification of table schema (new column added). - if (slotId >= taosArrayGetSize(pRow)) { + int32_t slotId = slotIds[i]; + SLastCol* pColVal = (SLastCol*)taosArrayGet(pRow, i); + SColVal* pVal = &pColVal->colVal; + + if (COL_VAL_IS_NONE(&pColVal->colVal)) { + continue; + } + allNullRow = false; + if (IS_VAR_DATA_TYPE(pColVal->colVal.type)) { + if (!COL_VAL_IS_VALUE(&pColVal->colVal)) { colDataSetNULL(pColInfoData, numOfRows); - continue; - } - SLastCol* pColVal = (SLastCol*)taosArrayGet(pRow, slotId); - SColVal* pVal = &pColVal->colVal; - - if (IS_VAR_DATA_TYPE(pColVal->colVal.type)) { - if (!COL_VAL_IS_VALUE(&pColVal->colVal)) { - colDataSetNULL(pColInfoData, numOfRows); - } else { - varDataSetLen(pReader->transferBuf[slotId], pVal->value.nData); - memcpy(varDataVal(pReader->transferBuf[slotId]), pVal->value.pData, pVal->value.nData); - colDataSetVal(pColInfoData, numOfRows, pReader->transferBuf[slotId], false); - } } else { - colDataSetVal(pColInfoData, numOfRows, (const char*)&pVal->value.val, !COL_VAL_IS_VALUE(pVal)); + varDataSetLen(pReader->transferBuf[slotId], pVal->value.nData); + + memcpy(varDataVal(pReader->transferBuf[slotId]), pVal->value.pData, pVal->value.nData); + colDataSetVal(pColInfoData, numOfRows, pReader->transferBuf[slotId], false); } + } else { + colDataSetVal(pColInfoData, numOfRows, (const char*)&pVal->value.val, !COL_VAL_IS_VALUE(pVal)); } } - pBlock->info.rows += 1; + pBlock->info.rows += allNullRow ? 0 : 1; } else { tsdbError("invalid retrieve type:%d, %s", pReader->type, idStr); return TSDB_CODE_INVALID_PARA; @@ -143,7 +121,7 @@ static int32_t setTableSchema(SCacheRowsReader* p, uint64_t suid, const char* id } int32_t tsdbCacherowsReaderOpen(void* pVnode, int32_t type, void* pTableIdList, int32_t numOfTables, int32_t numOfCols, - uint64_t suid, void** pReader, const char* idstr) { + SArray* pCidList, int32_t* pSlotIds, uint64_t suid, void** pReader, const char* idstr) { *pReader = NULL; SCacheRowsReader* p = taosMemoryCalloc(1, sizeof(SCacheRowsReader)); if (p == NULL) { @@ -155,6 +133,8 @@ int32_t tsdbCacherowsReaderOpen(void* pVnode, int32_t type, void* pTableIdList, p->pTsdb = p->pVnode->pTsdb; p->verRange = (SVersionRange){.minVer = 0, .maxVer = UINT64_MAX}; p->numOfCols = numOfCols; + p->pCidList = pCidList; + p->pSlotIds = pSlotIds; p->suid = suid; if (numOfTables == 0) { @@ -235,32 +215,9 @@ void* tsdbCacherowsReaderClose(void* pReader) { return NULL; } -static int32_t doExtractCacheRow(SCacheRowsReader* pr, SLRUCache* lruCache, uint64_t uid, SArray** pRow, - LRUHandle** h) { - int32_t code = TSDB_CODE_SUCCESS; - *pRow = NULL; - - if (HASTYPE(pr->type, CACHESCAN_RETRIEVE_LAST_ROW)) { - code = tsdbCacheGetLastrowH(lruCache, uid, pr, h); - } else { - code = tsdbCacheGetLastH(lruCache, uid, pr, h); - } - - if (code != TSDB_CODE_SUCCESS) { - return code; - } - - // no data in the table of Uid - if (*h != NULL) { - *pRow = (SArray*)taosLRUCacheValue(lruCache, *h); - } - - return code; -} - static void freeItem(void* pItem) { SLastCol* pCol = (SLastCol*)pItem; - if (IS_VAR_DATA_TYPE(pCol->colVal.type)) { + if (IS_VAR_DATA_TYPE(pCol->colVal.type) && pCol->colVal.value.pData) { taosMemoryFree(pCol->colVal.value.pData); } } @@ -286,19 +243,17 @@ static int32_t tsdbCacheQueryReseek(void* pQHandle) { } } -int32_t tsdbRetrieveCacheRows(void* pReader, SSDataBlock* pResBlock, const int32_t* slotIds, SArray* pTableUidList) { +int32_t tsdbRetrieveCacheRows(void* pReader, SSDataBlock* pResBlock, const int32_t* slotIds, const int32_t* dstSlotIds, + SArray* pTableUidList) { if (pReader == NULL || pResBlock == NULL) { return TSDB_CODE_INVALID_PARA; } SCacheRowsReader* pr = pReader; - - int32_t code = TSDB_CODE_SUCCESS; - SLRUCache* lruCache = pr->pVnode->pTsdb->lruCache; - LRUHandle* h = NULL; - SArray* pRow = NULL; - bool hasRes = false; - SArray* pLastCols = NULL; + int32_t code = TSDB_CODE_SUCCESS; + SArray* pRow = taosArrayInit(TARRAY_SIZE(pr->pCidList), sizeof(SLastCol)); + bool hasRes = false; + SArray* pLastCols = NULL; void** pRes = taosMemoryCalloc(pr->numOfCols, POINTER_BYTES); if (pRes == NULL) { @@ -307,20 +262,22 @@ int32_t tsdbRetrieveCacheRows(void* pReader, SSDataBlock* pResBlock, const int32 } for (int32_t j = 0; j < pr->numOfCols; ++j) { - pRes[j] = taosMemoryCalloc( - 1, sizeof(SFirstLastRes) + pr->pSchema->columns[-1 == slotIds[j] ? 0 : slotIds[j]].bytes + VARSTR_HEADER_SIZE); + pRes[j] = + taosMemoryCalloc(1, sizeof(SFirstLastRes) + pr->pSchema->columns[/*-1 == slotIds[j] ? 0 : */ slotIds[j]].bytes + + VARSTR_HEADER_SIZE); SFirstLastRes* p = (SFirstLastRes*)varDataVal(pRes[j]); p->ts = INT64_MIN; } - pLastCols = taosArrayInit(pr->pSchema->numOfCols, sizeof(SLastCol)); + pLastCols = taosArrayInit(pr->numOfCols, sizeof(SLastCol)); if (pLastCols == NULL) { code = TSDB_CODE_OUT_OF_MEMORY; goto _end; } - for (int32_t i = 0; i < pr->pSchema->numOfCols; ++i) { - struct STColumn* pCol = &pr->pSchema->columns[i]; + for (int32_t i = 0; i < pr->numOfCols; ++i) { + int32_t slotId = slotIds[i]; + struct STColumn* pCol = &pr->pSchema->columns[slotId]; SLastCol p = {.ts = INT64_MIN, .colVal.type = pCol->type, .colVal.flag = CV_FLAG_NULL}; if (IS_VAR_DATA_TYPE(pCol->type)) { @@ -337,6 +294,8 @@ int32_t tsdbRetrieveCacheRows(void* pReader, SSDataBlock* pResBlock, const int32 pr->pDataFReader = NULL; pr->pDataFReaderLast = NULL; + int32_t ltype = (pr->type & CACHESCAN_RETRIEVE_LAST) >> 3; + // retrieve the only one last row of all tables in the uid list. if (HASTYPE(pr->type, CACHESCAN_RETRIEVE_TYPE_SINGLE)) { int64_t st = taosGetTimestampUs(); @@ -344,16 +303,14 @@ int32_t tsdbRetrieveCacheRows(void* pReader, SSDataBlock* pResBlock, const int32 for (int32_t i = 0; i < pr->numOfTables; ++i) { STableKeyInfo* pKeyInfo = &pr->pTableList[i]; - code = doExtractCacheRow(pr, lruCache, pKeyInfo->uid, &pRow, &h); - if (code != TSDB_CODE_SUCCESS) { - goto _end; - } - - if (h == NULL) { + tsdbCacheGet(pr->pTsdb, pKeyInfo->uid, pRow, pr, ltype); + if (TARRAY_SIZE(pRow) <= 0) { + taosArrayClearEx(pRow, freeItem); continue; } - if (taosArrayGetSize(pRow) <= 0) { - tsdbCacheRelease(lruCache, h); + SLastCol* pColVal = (SLastCol*)taosArrayGet(pRow, 0); + if (COL_VAL_IS_NONE(&pColVal->colVal)) { + taosArrayClearEx(pRow, freeItem); continue; } @@ -361,47 +318,34 @@ int32_t tsdbRetrieveCacheRows(void* pReader, SSDataBlock* pResBlock, const int32 bool hasNotNullRow = true; int64_t singleTableLastTs = INT64_MAX; for (int32_t k = 0; k < pr->numOfCols; ++k) { - int32_t slotId = slotIds[k]; + SLastCol* p = taosArrayGet(pLastCols, k); + SLastCol* pColVal = (SLastCol*)taosArrayGet(pRow, k); - if (slotId == -1) { // the primary timestamp - SLastCol* p = taosArrayGet(pLastCols, 0); - SLastCol* pCol = (SLastCol*)taosArrayGet(pRow, 0); - if (pCol->ts > p->ts) { - hasRes = true; - p->ts = pCol->ts; - p->colVal = pCol->colVal; - singleTableLastTs = pCol->ts; + if (pColVal->ts > p->ts) { + if (!COL_VAL_IS_VALUE(&pColVal->colVal) && HASTYPE(pr->type, CACHESCAN_RETRIEVE_LAST)) { + if (!COL_VAL_IS_VALUE(&p->colVal)) { + hasNotNullRow = false; + } + continue; } - } else { - SLastCol* p = taosArrayGet(pLastCols, slotId); - SLastCol* pColVal = (SLastCol*)taosArrayGet(pRow, slotId); - if (pColVal->ts > p->ts) { - if (!COL_VAL_IS_VALUE(&pColVal->colVal) && HASTYPE(pr->type, CACHESCAN_RETRIEVE_LAST)) { - if (!COL_VAL_IS_VALUE(&p->colVal)) { - hasNotNullRow = false; - } - continue; + hasRes = true; + p->ts = pColVal->ts; + if (pColVal->ts < singleTableLastTs && HASTYPE(pr->type, CACHESCAN_RETRIEVE_LAST)) { + singleTableLastTs = pColVal->ts; + } + + if (!IS_VAR_DATA_TYPE(pColVal->colVal.type)) { + p->colVal = pColVal->colVal; + } else { + if (COL_VAL_IS_VALUE(&pColVal->colVal)) { + memcpy(p->colVal.value.pData, pColVal->colVal.value.pData, pColVal->colVal.value.nData); } - hasRes = true; - p->ts = pColVal->ts; - if (pColVal->ts < singleTableLastTs && HASTYPE(pr->type, CACHESCAN_RETRIEVE_LAST)) { - singleTableLastTs = pColVal->ts; - } - - if (!IS_VAR_DATA_TYPE(pColVal->colVal.type)) { - p->colVal = pColVal->colVal; - } else { - if (COL_VAL_IS_VALUE(&pColVal->colVal)) { - memcpy(p->colVal.value.pData, pColVal->colVal.value.pData, pColVal->colVal.value.nData); - } - - p->colVal.value.nData = pColVal->colVal.value.nData; - p->colVal.type = pColVal->colVal.type; - p->colVal.flag = pColVal->colVal.flag; - p->colVal.cid = pColVal->colVal.cid; - } + p->colVal.value.nData = pColVal->colVal.value.nData; + p->colVal.type = pColVal->colVal.type; + p->colVal.flag = pColVal->colVal.flag; + p->colVal.cid = pColVal->colVal.cid; } } } @@ -423,33 +367,31 @@ int32_t tsdbRetrieveCacheRows(void* pReader, SSDataBlock* pResBlock, const int32 taosArraySet(pTableUidList, 0, &pKeyInfo->uid); } - tsdbCacheRelease(lruCache, h); + taosArrayClearEx(pRow, freeItem); } if (hasRes) { - saveOneRow(pLastCols, pResBlock, pr, slotIds, pRes, pr->idstr); + saveOneRow(pLastCols, pResBlock, pr, slotIds, dstSlotIds, pRes, pr->idstr); } } else if (HASTYPE(pr->type, CACHESCAN_RETRIEVE_TYPE_ALL)) { for (int32_t i = pr->tableIndex; i < pr->numOfTables; ++i) { STableKeyInfo* pKeyInfo = &pr->pTableList[i]; - code = doExtractCacheRow(pr, lruCache, pKeyInfo->uid, &pRow, &h); - if (code != TSDB_CODE_SUCCESS) { - goto _end; - } - if (h == NULL) { + tsdbCacheGet(pr->pTsdb, pKeyInfo->uid, pRow, pr, ltype); + if (TARRAY_SIZE(pRow) <= 0) { + taosArrayClearEx(pRow, freeItem); continue; } - if (taosArrayGetSize(pRow) <= 0) { - tsdbCacheRelease(lruCache, h); + SLastCol* pColVal = (SLastCol*)taosArrayGet(pRow, 0); + if (COL_VAL_IS_NONE(&pColVal->colVal)) { + taosArrayClearEx(pRow, freeItem); continue; } - saveOneRow(pRow, pResBlock, pr, slotIds, pRes, pr->idstr); - // TODO reset the pRes + saveOneRow(pRow, pResBlock, pr, slotIds, dstSlotIds, pRes, pr->idstr); + taosArrayClearEx(pRow, freeItem); taosArrayPush(pTableUidList, &pKeyInfo->uid); - tsdbCacheRelease(lruCache, h); pr->tableIndex += 1; if (pResBlock->info.rows >= pResBlock->info.capacity) { @@ -475,6 +417,7 @@ _end: } taosMemoryFree(pRes); + taosArrayDestroyEx(pRow, freeItem); taosArrayDestroyEx(pLastCols, freeItem); return code; } diff --git a/source/dnode/vnode/src/tsdb/tsdbMemTable.c b/source/dnode/vnode/src/tsdb/tsdbMemTable.c index f27a28acb3..cd511e0d23 100644 --- a/source/dnode/vnode/src/tsdb/tsdbMemTable.c +++ b/source/dnode/vnode/src/tsdb/tsdbMemTable.c @@ -13,8 +13,8 @@ * along with this program. If not, see . */ -#include #include "tsdb.h" +#include "util/tsimplehash.h" #define MEM_MIN_HASH 1024 #define SL_MAX_LEVEL 5 @@ -141,7 +141,6 @@ int32_t tsdbDeleteTableData(STsdb *pTsdb, int64_t version, tb_uid_t suid, tb_uid SMemTable *pMemTable = pTsdb->mem; STbData *pTbData = NULL; SVBufPool *pPool = pTsdb->pVnode->inUse; - TSDBKEY lastKey = {.version = version, .ts = eKey}; // check if table exists SMetaInfo info; @@ -182,7 +181,7 @@ int32_t tsdbDeleteTableData(STsdb *pTsdb, int64_t version, tb_uid_t suid, tb_uid pMemTable->nDel++; pMemTable->minVer = TMIN(pMemTable->minVer, version); pMemTable->maxVer = TMIN(pMemTable->maxVer, version); - + /* if (TSDB_CACHE_LAST_ROW(pMemTable->pTsdb->pVnode->config) && tsdbKeyCmprFn(&lastKey, &pTbData->maxKey) >= 0) { tsdbCacheDeleteLastrow(pTsdb->lruCache, pTbData->uid, eKey); } @@ -190,6 +189,10 @@ int32_t tsdbDeleteTableData(STsdb *pTsdb, int64_t version, tb_uid_t suid, tb_uid if (TSDB_CACHE_LAST(pMemTable->pTsdb->pVnode->config)) { tsdbCacheDeleteLast(pTsdb->lruCache, pTbData->uid, eKey); } + */ + if (eKey >= pTbData->maxKey && sKey <= pTbData->maxKey) { + tsdbCacheDel(pTsdb, suid, uid, sKey, eKey); + } tsdbTrace("vgId:%d, delete data from table suid:%" PRId64 " uid:%" PRId64 " skey:%" PRId64 " eKey:%" PRId64 " at version %" PRId64, @@ -285,8 +288,8 @@ bool tsdbTbDataIterNext(STbDataIter *pIter) { int64_t tsdbCountTbDataRows(STbData *pTbData) { SMemSkipListNode *pNode = pTbData->sl.pHead; - int64_t rowsNum = 0; - + int64_t rowsNum = 0; + while (NULL != pNode) { pNode = SL_GET_NODE_FORWARD(pNode, 0); if (pNode == pTbData->sl.pTail) { @@ -299,17 +302,17 @@ int64_t tsdbCountTbDataRows(STbData *pTbData) { return rowsNum; } -void tsdbMemTableCountRows(SMemTable *pMemTable, SSHashObj* pTableMap, int64_t *rowsNum) { +void tsdbMemTableCountRows(SMemTable *pMemTable, SSHashObj *pTableMap, int64_t *rowsNum) { taosRLockLatch(&pMemTable->latch); for (int32_t i = 0; i < pMemTable->nBucket; ++i) { STbData *pTbData = pMemTable->aBucket[i]; while (pTbData) { - void* p = tSimpleHashGet(pTableMap, &pTbData->uid, sizeof(pTbData->uid)); + void *p = tSimpleHashGet(pTableMap, &pTbData->uid, sizeof(pTbData->uid)); if (p == NULL) { pTbData = pTbData->next; continue; } - + *rowsNum += tsdbCountTbDataRows(pTbData); pTbData = pTbData->next; } @@ -669,15 +672,8 @@ static int32_t tsdbInsertColDataToTable(SMemTable *pMemTable, STbData *pTbData, if (key.ts >= pTbData->maxKey) { pTbData->maxKey = key.ts; - - if (TSDB_CACHE_LAST_ROW(pMemTable->pTsdb->pVnode->config)) { - tsdbCacheInsertLastrow(pMemTable->pTsdb->lruCache, pMemTable->pTsdb, pTbData->uid, &lRow, true); - } - } - - if (TSDB_CACHE_LAST(pMemTable->pTsdb->pVnode->config)) { - tsdbCacheInsertLast(pMemTable->pTsdb->lruCache, pTbData->uid, &lRow, pMemTable->pTsdb); } + tsdbCacheUpdate(pMemTable->pTsdb, pTbData->suid, pTbData->uid, &lRow); // SMemTable pMemTable->minKey = TMIN(pMemTable->minKey, pTbData->minKey); @@ -737,15 +733,8 @@ static int32_t tsdbInsertRowDataToTable(SMemTable *pMemTable, STbData *pTbData, if (key.ts >= pTbData->maxKey) { pTbData->maxKey = key.ts; - - if (TSDB_CACHE_LAST_ROW(pMemTable->pTsdb->pVnode->config)) { - tsdbCacheInsertLastrow(pMemTable->pTsdb->lruCache, pMemTable->pTsdb, pTbData->uid, &lRow, true); - } - } - - if (TSDB_CACHE_LAST(pMemTable->pTsdb->pVnode->config)) { - tsdbCacheInsertLast(pMemTable->pTsdb->lruCache, pTbData->uid, &lRow, pMemTable->pTsdb); } + tsdbCacheUpdate(pMemTable->pTsdb, pTbData->suid, pTbData->uid, &lRow); // SMemTable pMemTable->minKey = TMIN(pMemTable->minKey, pTbData->minKey); diff --git a/source/dnode/vnode/src/vnd/vnodeCfg.c b/source/dnode/vnode/src/vnd/vnodeCfg.c index c326c8bfac..faa4d2fc57 100644 --- a/source/dnode/vnode/src/vnd/vnodeCfg.c +++ b/source/dnode/vnode/src/vnd/vnodeCfg.c @@ -57,6 +57,28 @@ int vnodeCheckCfg(const SVnodeCfg *pCfg) { return 0; } +const char* vnodeRoleToStr(ESyncRole role) { + switch (role) { + case TAOS_SYNC_ROLE_VOTER: + return "true"; + case TAOS_SYNC_ROLE_LEARNER: + return "false"; + default: + return "unknown"; + } +} + +const ESyncRole vnodeStrToRole(char* str) { + if(strcmp(str, "true") == 0){ + return TAOS_SYNC_ROLE_VOTER; + } + if(strcmp(str, "false") == 0){ + return TAOS_SYNC_ROLE_LEARNER; + } + + return TAOS_SYNC_ROLE_ERROR; +} + int vnodeEncodeConfig(const void *pObj, SJson *pJson) { const SVnodeCfg *pCfg = (SVnodeCfg *)pObj; @@ -112,6 +134,7 @@ int vnodeEncodeConfig(const void *pObj, SJson *pJson) { if (tjsonAddIntegerToObject(pJson, "sstTrigger", pCfg->sttTrigger) < 0) return -1; if (tjsonAddIntegerToObject(pJson, "hashBegin", pCfg->hashBegin) < 0) return -1; if (tjsonAddIntegerToObject(pJson, "hashEnd", pCfg->hashEnd) < 0) return -1; + if (tjsonAddIntegerToObject(pJson, "hashChange", pCfg->hashChange) < 0) return -1; if (tjsonAddIntegerToObject(pJson, "hashMethod", pCfg->hashMethod) < 0) return -1; if (tjsonAddIntegerToObject(pJson, "hashPrefix", pCfg->hashPrefix) < 0) return -1; if (tjsonAddIntegerToObject(pJson, "hashSuffix", pCfg->hashSuffix) < 0) return -1; @@ -128,9 +151,9 @@ int vnodeEncodeConfig(const void *pObj, SJson *pJson) { SJson *nodeInfo = tjsonCreateArray(); if (nodeInfo == NULL) return -1; if (tjsonAddItemToObject(pJson, "syncCfg.nodeInfo", nodeInfo) < 0) return -1; - vDebug("vgId:%d, encode config, replicas:%d selfIndex:%d", pCfg->vgId, pCfg->syncCfg.replicaNum, - pCfg->syncCfg.myIndex); - for (int i = 0; i < pCfg->syncCfg.replicaNum; ++i) { + vDebug("vgId:%d, encode config, replicas:%d totalReplicas:%d selfIndex:%d", pCfg->vgId, pCfg->syncCfg.replicaNum, + pCfg->syncCfg.totalReplicaNum, pCfg->syncCfg.myIndex); + for (int i = 0; i < pCfg->syncCfg.totalReplicaNum; ++i) { SJson *info = tjsonCreateObject(); SNodeInfo *pNode = (SNodeInfo *)&pCfg->syncCfg.nodeInfo[i]; if (info == NULL) return -1; @@ -138,6 +161,7 @@ int vnodeEncodeConfig(const void *pObj, SJson *pJson) { if (tjsonAddStringToObject(info, "nodeFqdn", pNode->nodeFqdn) < 0) return -1; if (tjsonAddIntegerToObject(info, "nodeId", pNode->nodeId) < 0) return -1; if (tjsonAddIntegerToObject(info, "clusterId", pNode->clusterId) < 0) return -1; + if (tjsonAddStringToObject(info, "isReplica", vnodeRoleToStr(pNode->nodeRole)) < 0) return -1; if (tjsonAddItemToArray(nodeInfo, info) < 0) return -1; vDebug("vgId:%d, encode config, replica:%d ep:%s:%u dnode:%d", pCfg->vgId, i, pNode->nodeFqdn, pNode->nodePort, pNode->nodeId); @@ -226,6 +250,8 @@ int vnodeDecodeConfig(const SJson *pJson, void *pObj) { if (code < 0) return -1; tjsonGetNumberValue(pJson, "hashEnd", pCfg->hashEnd, code); if (code < 0) return -1; + tjsonGetNumberValue(pJson, "hashChange", pCfg->hashChange, code); + if (code < 0) return -1; tjsonGetNumberValue(pJson, "hashMethod", pCfg->hashMethod, code); if (code < 0) return -1; tjsonGetNumberValue(pJson, "hashPrefix", pCfg->hashPrefix, code); @@ -251,10 +277,10 @@ int vnodeDecodeConfig(const SJson *pJson, void *pObj) { SJson *nodeInfo = tjsonGetObjectItem(pJson, "syncCfg.nodeInfo"); int arraySize = tjsonGetArraySize(nodeInfo); - if (arraySize != pCfg->syncCfg.replicaNum) return -1; + pCfg->syncCfg.totalReplicaNum = arraySize; - vDebug("vgId:%d, decode config, replicas:%d selfIndex:%d", pCfg->vgId, pCfg->syncCfg.replicaNum, - pCfg->syncCfg.myIndex); + vDebug("vgId:%d, decode config, replicas:%d totalReplicas:%d selfIndex:%d", pCfg->vgId, pCfg->syncCfg.replicaNum, + pCfg->syncCfg.totalReplicaNum, pCfg->syncCfg.myIndex); for (int i = 0; i < arraySize; ++i) { SJson *info = tjsonGetArrayItem(nodeInfo, i); SNodeInfo *pNode = &pCfg->syncCfg.nodeInfo[i]; @@ -266,6 +292,15 @@ int vnodeDecodeConfig(const SJson *pJson, void *pObj) { if (code < 0) return -1; tjsonGetNumberValue(info, "clusterId", pNode->clusterId, code); if (code < 0) return -1; + char role[10] = {0}; + code = tjsonGetStringValue(info, "isReplica", role); + if (code < 0) return -1; + if(strlen(role) != 0){ + pNode->nodeRole = vnodeStrToRole(role); + } + else{ + pNode->nodeRole = TAOS_SYNC_ROLE_VOTER; + } vDebug("vgId:%d, decode config, replica:%d ep:%s:%u dnode:%d", pCfg->vgId, i, pNode->nodeFqdn, pNode->nodePort, pNode->nodeId); } diff --git a/source/dnode/vnode/src/vnd/vnodeCommit.c b/source/dnode/vnode/src/vnd/vnodeCommit.c index 847125018c..74168591d2 100644 --- a/source/dnode/vnode/src/vnd/vnodeCommit.c +++ b/source/dnode/vnode/src/vnd/vnodeCommit.c @@ -144,8 +144,8 @@ _exit: } int vnodeShouldCommit(SVnode *pVnode, bool atExit) { - bool diskAvail = osDataSpaceAvailable(); - bool needCommit = false; + bool diskAvail = osDataSpaceAvailable(); + bool needCommit = false; taosThreadMutexLock(&pVnode->mutex); if (pVnode->inUse && diskAvail) { @@ -439,6 +439,9 @@ static int vnodeCommitImpl(SCommitInfo *pInfo) { code = tsdbCommit(pVnode->pTsdb, pInfo); TSDB_CHECK_CODE(code, lino, _exit); + code = tsdbCacheCommit(pVnode->pTsdb); + TSDB_CHECK_CODE(code, lino, _exit); + if (VND_IS_RSMA(pVnode)) { code = smaCommit(pVnode->pSma, pInfo); TSDB_CHECK_CODE(code, lino, _exit); diff --git a/source/dnode/vnode/src/vnd/vnodeOpen.c b/source/dnode/vnode/src/vnd/vnodeOpen.c index e8f34d27ac..b5e7c6875b 100644 --- a/source/dnode/vnode/src/vnd/vnodeOpen.c +++ b/source/dnode/vnode/src/vnd/vnodeOpen.c @@ -76,19 +76,41 @@ int32_t vnodeAlterReplica(const char *path, SAlterVnodeReplicaReq *pReq, STfs *p } SSyncCfg *pCfg = &info.config.syncCfg; - pCfg->myIndex = pReq->selfIndex; - pCfg->replicaNum = pReq->replica; + + pCfg->replicaNum = 0; + pCfg->totalReplicaNum = 0; memset(&pCfg->nodeInfo, 0, sizeof(pCfg->nodeInfo)); - vInfo("vgId:%d, save config while alter, replicas:%d selfIndex:%d", pReq->vgId, pCfg->replicaNum, pCfg->myIndex); for (int i = 0; i < pReq->replica; ++i) { SNodeInfo *pNode = &pCfg->nodeInfo[i]; pNode->nodeId = pReq->replicas[i].id; pNode->nodePort = pReq->replicas[i].port; tstrncpy(pNode->nodeFqdn, pReq->replicas[i].fqdn, sizeof(pNode->nodeFqdn)); + pNode->nodeRole = TAOS_SYNC_ROLE_VOTER; (void)tmsgUpdateDnodeInfo(&pNode->nodeId, &pNode->clusterId, pNode->nodeFqdn, &pNode->nodePort); vInfo("vgId:%d, replica:%d ep:%s:%u dnode:%d", pReq->vgId, i, pNode->nodeFqdn, pNode->nodePort, pNode->nodeId); + pCfg->replicaNum++; } + if(pReq->selfIndex != -1){ + pCfg->myIndex = pReq->selfIndex; + } + for (int i = pCfg->replicaNum; i < pReq->replica + pReq->learnerReplica; ++i) { + SNodeInfo *pNode = &pCfg->nodeInfo[i]; + pNode->nodeId = pReq->learnerReplicas[pCfg->totalReplicaNum].id; + pNode->nodePort = pReq->learnerReplicas[pCfg->totalReplicaNum].port; + pNode->nodeRole = TAOS_SYNC_ROLE_LEARNER; + tstrncpy(pNode->nodeFqdn, pReq->learnerReplicas[pCfg->totalReplicaNum].fqdn, sizeof(pNode->nodeFqdn)); + (void)tmsgUpdateDnodeInfo(&pNode->nodeId, &pNode->clusterId, pNode->nodeFqdn, &pNode->nodePort); + vInfo("vgId:%d, replica:%d ep:%s:%u dnode:%d", pReq->vgId, i, pNode->nodeFqdn, pNode->nodePort, pNode->nodeId); + pCfg->totalReplicaNum++; + } + pCfg->totalReplicaNum += pReq->replica; + if(pReq->learnerSelfIndex != -1){ + pCfg->myIndex = pReq->replica + pReq->learnerSelfIndex; + } + + vInfo("vgId:%d, save config while alter, replicas:%d totalReplicas:%d selfIndex:%d", + pReq->vgId, pCfg->replicaNum, pCfg->totalReplicaNum, pCfg->myIndex); info.config.syncCfg = *pCfg; ret = vnodeSaveInfo(dir, &info); @@ -176,11 +198,13 @@ int32_t vnodeAlterHashRange(const char *srcPath, const char *dstPath, SAlterVnod info.config.vgId = pReq->dstVgId; info.config.hashBegin = pReq->hashBegin; info.config.hashEnd = pReq->hashEnd; + info.config.hashChange = true; info.config.walCfg.vgId = pReq->dstVgId; SSyncCfg *pCfg = &info.config.syncCfg; pCfg->myIndex = 0; pCfg->replicaNum = 1; + pCfg->totalReplicaNum = 1; memset(&pCfg->nodeInfo, 0, sizeof(pCfg->nodeInfo)); vInfo("vgId:%d, alter vnode replicas to 1", pReq->srcVgId); @@ -212,8 +236,6 @@ int32_t vnodeAlterHashRange(const char *srcPath, const char *dstPath, SAlterVnod return -1; } - // todo vnode compact here - vInfo("vgId:%d, vnode hashrange is altered", info.config.vgId); return 0; } @@ -248,7 +270,7 @@ SVnode *vnodeOpen(const char *path, STfs *pTfs, SMsgCb msgCb) { // save vnode info on dnode ep changed bool updated = false; SSyncCfg *pCfg = &info.config.syncCfg; - for (int32_t i = 0; i < pCfg->replicaNum; ++i) { + for (int32_t i = 0; i < pCfg->totalReplicaNum; ++i) { SNodeInfo *pNode = &pCfg->nodeInfo[i]; if (tmsgUpdateDnodeInfo(&pNode->nodeId, &pNode->clusterId, pNode->nodeFqdn, &pNode->nodePort)) { updated = true; @@ -404,6 +426,14 @@ void vnodeClose(SVnode *pVnode) { // start the sync timer after the queue is ready int32_t vnodeStart(SVnode *pVnode) { return vnodeSyncStart(pVnode); } +int32_t vnodeIsCatchUp(SVnode *pVnode){ + return syncIsCatchUp(pVnode->sync); +} + +ESyncRole vnodeGetRole(SVnode *pVnode){ + return syncGetRole(pVnode->sync); +} + void vnodeStop(SVnode *pVnode) {} int64_t vnodeGetSyncHandle(SVnode *pVnode) { return pVnode->sync; } diff --git a/source/dnode/vnode/src/vnd/vnodeSvr.c b/source/dnode/vnode/src/vnd/vnodeSvr.c index bd4a20b07e..fe1ccf90c8 100644 --- a/source/dnode/vnode/src/vnd/vnodeSvr.c +++ b/source/dnode/vnode/src/vnd/vnodeSvr.c @@ -389,6 +389,11 @@ int32_t vnodeProcessWriteMsg(SVnode *pVnode, SRpcMsg *pMsg, int64_t version, SRp goto _err; } break; + case TDMT_VND_TMQ_SEEK_TO_OFFSET: + if (tqProcessSeekReq(pVnode->pTq, version, pReq, pMsg->contLen - sizeof(SMsgHead)) < 0) { + goto _err; + } + break; case TDMT_VND_TMQ_ADD_CHECKINFO: if (tqProcessAddCheckInfoReq(pVnode->pTq, version, pReq, len) < 0) { goto _err; @@ -430,7 +435,10 @@ int32_t vnodeProcessWriteMsg(SVnode *pVnode, SRpcMsg *pMsg, int64_t version, SRp } } break; case TDMT_VND_ALTER_CONFIRM: - vnodeProcessAlterConfirmReq(pVnode, version, pReq, len, pRsp); + needCommit = pVnode->config.hashChange; + if (vnodeProcessAlterConfirmReq(pVnode, version, pReq, len, pRsp) < 0) { + goto _err; + } break; case TDMT_VND_ALTER_CONFIG: vnodeProcessAlterConfigReq(pVnode, version, pReq, len, pRsp); @@ -548,6 +556,10 @@ int32_t vnodeProcessFetchMsg(SVnode *pVnode, SRpcMsg *pMsg, SQueueInfo *pInfo) { return vnodeGetTableCfg(pVnode, pMsg, true); case TDMT_VND_BATCH_META: return vnodeGetBatchMeta(pVnode, pMsg); + case TDMT_VND_TMQ_CONSUME: + return tqProcessPollReq(pVnode->pTq, pMsg); + case TDMT_VND_TMQ_VG_WALINFO: + return tqProcessVgWalInfoReq(pVnode->pTq, pMsg); case TDMT_STREAM_TASK_RUN: return tqProcessTaskRunReq(pVnode->pTq, pMsg); case TDMT_STREAM_TASK_DISPATCH: @@ -1066,7 +1078,7 @@ static int32_t vnodeCellValConvertToColVal(STColumn *pCol, SCellVal *pCellVal, S if (IS_VAR_DATA_TYPE(pCol->type)) { pColVal->value.nData = varDataLen(pCellVal->val); - pColVal->value.pData = varDataVal(pCellVal->val); + pColVal->value.pData = (uint8_t *)varDataVal(pCellVal->val); } else if (TSDB_DATA_TYPE_FLOAT == pCol->type) { float f = GET_FLOAT_VAL(pCellVal->val); memcpy(&pColVal->value.val, &f, sizeof(f)); @@ -1105,7 +1117,7 @@ static int32_t vnodeDecodeCreateTbReq(SSubmitReqConvertCxt *pCxt) { } SDecoder decoder = {0}; - tDecoderInit(&decoder, pCxt->pBlock->data, pCxt->msgIter.schemaLen); + tDecoderInit(&decoder, (uint8_t *)pCxt->pBlock->data, pCxt->msgIter.schemaLen); int32_t code = tDecodeSVCreateTbReq(&decoder, pCxt->pTbData->pCreateTbReq); tDecoderClear(&decoder); @@ -1167,7 +1179,7 @@ static int32_t vnodeRebuildSubmitReqMsg(SSubmitReq2 *pSubmitReq, void **ppMsg) { } if (TSDB_CODE_SUCCESS == code) { SEncoder encoder; - tEncoderInit(&encoder, pMsg, msglen); + tEncoderInit(&encoder, (uint8_t *)pMsg, msglen); code = tEncodeSubmitReq(&encoder, pSubmitReq); tEncoderClear(&encoder); } @@ -1456,14 +1468,39 @@ int32_t vnodeProcessCreateTSma(SVnode *pVnode, void *pCont, uint32_t contLen) { return vnodeProcessCreateTSmaReq(pVnode, 1, pCont, contLen, NULL); } +static int32_t vnodeConsolidateAlterHashRange(SVnode *pVnode, int64_t version) { + int32_t code = TSDB_CODE_SUCCESS; + + vInfo("vgId:%d, trim meta of tables per hash range [%" PRIu32 ", %" PRIu32 "]. apply-index:%" PRId64, TD_VID(pVnode), + pVnode->config.hashBegin, pVnode->config.hashEnd, version); + + // TODO: trim meta of tables from TDB per hash range [pVnode->config.hashBegin, pVnode->config.hashEnd] + + return code; +} + static int32_t vnodeProcessAlterConfirmReq(SVnode *pVnode, int64_t version, void *pReq, int32_t len, SRpcMsg *pRsp) { - vInfo("vgId:%d, alter replica confim msg is processed", TD_VID(pVnode)); + vInfo("vgId:%d, vnode handle msgType:alter-confirm, alter confim msg is processed", TD_VID(pVnode)); + int32_t code = TSDB_CODE_SUCCESS; + if (!pVnode->config.hashChange) { + goto _exit; + } + + code = vnodeConsolidateAlterHashRange(pVnode, version); + if (code < 0) { + vError("vgId:%d, failed to consolidate alter hashrange since %s. version:%" PRId64, TD_VID(pVnode), terrstr(), + version); + goto _exit; + } + pVnode->config.hashChange = false; + +_exit: pRsp->msgType = TDMT_VND_ALTER_CONFIRM_RSP; - pRsp->code = TSDB_CODE_SUCCESS; + pRsp->code = code; pRsp->pCont = NULL; pRsp->contLen = 0; - return 0; + return code; } static int32_t vnodeProcessAlterConfigReq(SVnode *pVnode, int64_t version, void *pReq, int32_t len, SRpcMsg *pRsp) { diff --git a/source/dnode/vnode/src/vnd/vnodeSync.c b/source/dnode/vnode/src/vnd/vnodeSync.c index f2ee5358cc..4ea5e3c6ec 100644 --- a/source/dnode/vnode/src/vnd/vnodeSync.c +++ b/source/dnode/vnode/src/vnd/vnodeSync.c @@ -571,6 +571,19 @@ static void vnodeBecomeFollower(const SSyncFSM *pFsm) { taosThreadMutexUnlock(&pVnode->lock); } +static void vnodeBecomeLearner(const SSyncFSM *pFsm) { + SVnode *pVnode = pFsm->data; + vInfo("vgId:%d, become learner", pVnode->config.vgId); + + taosThreadMutexLock(&pVnode->lock); + if (pVnode->blocked) { + pVnode->blocked = false; + vDebug("vgId:%d, become learner and post block", pVnode->config.vgId); + tsem_post(&pVnode->syncSem); + } + taosThreadMutexUnlock(&pVnode->lock); +} + static void vnodeBecomeLeader(const SSyncFSM *pFsm) { SVnode *pVnode = pFsm->data; vDebug("vgId:%d, become leader", pVnode->config.vgId); @@ -612,6 +625,7 @@ static SSyncFSM *vnodeSyncMakeFsm(SVnode *pVnode) { pFsm->FpApplyQueueItems = vnodeApplyQueueItems; pFsm->FpBecomeLeaderCb = vnodeBecomeLeader; pFsm->FpBecomeFollowerCb = vnodeBecomeFollower; + pFsm->FpBecomeLearnerCb = vnodeBecomeLearner; pFsm->FpReConfigCb = NULL; pFsm->FpSnapshotStartRead = vnodeSnapshotStartRead; pFsm->FpSnapshotStopRead = vnodeSnapshotStopRead; @@ -644,7 +658,7 @@ int32_t vnodeSyncOpen(SVnode *pVnode, char *path) { SSyncCfg *pCfg = &syncInfo.syncCfg; vInfo("vgId:%d, start to open sync, replica:%d selfIndex:%d", pVnode->config.vgId, pCfg->replicaNum, pCfg->myIndex); - for (int32_t i = 0; i < pCfg->replicaNum; ++i) { + for (int32_t i = 0; i < pCfg->totalReplicaNum; ++i) { SNodeInfo *pNode = &pCfg->nodeInfo[i]; vInfo("vgId:%d, index:%d ep:%s:%u dnode:%d cluster:%" PRId64, pVnode->config.vgId, i, pNode->nodeFqdn, pNode->nodePort, pNode->nodeId, pNode->clusterId); diff --git a/source/libs/command/src/explain.c b/source/libs/command/src/explain.c index 4302302d7a..c5b9eb7475 100644 --- a/source/libs/command/src/explain.c +++ b/source/libs/command/src/explain.c @@ -1128,6 +1128,11 @@ int32_t qExplainResNodeToRowsImpl(SExplainResNode *pResNode, SExplainCtx *ctx, i EXPLAIN_ROW_END(); QRY_ERR_RET(qExplainResAppendRow(ctx, tbuf, tlen, level + 1)); + EXPLAIN_ROW_NEW(level + 1, EXPLAIN_OUTPUT_FORMAT); + EXPLAIN_ROW_APPEND(EXPLAIN_IGNORE_GROUPID_FORMAT, pMergeNode->ignoreGroupId ? "true" : "false"); + EXPLAIN_ROW_END(); + QRY_ERR_RET(qExplainResAppendRow(ctx, tbuf, tlen, level + 1)); + EXPLAIN_ROW_NEW(level + 1, EXPLAIN_MERGE_KEYS_FORMAT); if (pMergeNode->groupSort) { EXPLAIN_ROW_APPEND(EXPLAIN_STRING_TYPE_FORMAT, "_group_id asc"); diff --git a/source/libs/executor/src/aggregateoperator.c b/source/libs/executor/src/aggregateoperator.c index 9b463a3dee..5cd95d3311 100644 --- a/source/libs/executor/src/aggregateoperator.c +++ b/source/libs/executor/src/aggregateoperator.c @@ -464,8 +464,8 @@ int32_t doInitAggInfoSup(SAggSupporter* pAggSup, SqlFunctionCtx* pCtx, int32_t n getBufferPgSize(pAggSup->resultRowSize, &defaultPgsz, &defaultBufsz); if (!osTempSpaceAvailable()) { - code = TSDB_CODE_NO_AVAIL_DISK; - qError("Init stream agg supporter failed since %s, %s", terrstr(code), pKey); + code = TSDB_CODE_NO_DISKSPACE; + qError("Init stream agg supporter failed since %s, key:%s, tempDir:%s", terrstr(code), pKey, tsTempDir); return code; } diff --git a/source/libs/executor/src/cachescanoperator.c b/source/libs/executor/src/cachescanoperator.c index 00eb042478..430b0b11ed 100644 --- a/source/libs/executor/src/cachescanoperator.c +++ b/source/libs/executor/src/cachescanoperator.c @@ -33,18 +33,21 @@ typedef struct SCacheRowsScanInfo { void* pLastrowReader; SColMatchInfo matchInfo; int32_t* pSlotIds; + int32_t* pDstSlotIds; SExprSupp pseudoExprSup; int32_t retrieveType; int32_t currentGroupIndex; SSDataBlock* pBufferredRes; SArray* pUidList; + SArray* pCidList; int32_t indexOfBufferedRes; STableListInfo* pTableList; } SCacheRowsScanInfo; static SSDataBlock* doScanCache(SOperatorInfo* pOperator); static void destroyCacheScanOperator(void* param); -static int32_t extractCacheScanSlotId(const SArray* pColMatchInfo, SExecTaskInfo* pTaskInfo, int32_t** pSlotIds); +static int32_t extractCacheScanSlotId(const SArray* pColMatchInfo, SExecTaskInfo* pTaskInfo, int32_t** pSlotIds, + int32_t** pDstSlotIds); static int32_t removeRedundantTsCol(SLastRowScanPhysiNode* pScanNode, SColMatchInfo* pColMatchInfo); #define SCAN_ROW_TYPE(_t) ((_t) ? CACHESCAN_RETRIEVE_LAST : CACHESCAN_RETRIEVE_LAST_ROW) @@ -73,9 +76,16 @@ SOperatorInfo* createCacherowsScanOperator(SLastRowScanPhysiNode* pScanNode, SRe goto _error; } + SArray* pCidList = taosArrayInit(numOfCols, sizeof(int16_t)); + for (int i = 0; i < TARRAY_SIZE(pInfo->matchInfo.pList); ++i) { + SColMatchItem* pColInfo = taosArrayGet(pInfo->matchInfo.pList, i); + taosArrayPush(pCidList, &pColInfo->colId); + } + pInfo->pCidList = pCidList; + removeRedundantTsCol(pScanNode, &pInfo->matchInfo); - code = extractCacheScanSlotId(pInfo->matchInfo.pList, pTaskInfo, &pInfo->pSlotIds); + code = extractCacheScanSlotId(pInfo->matchInfo.pList, pTaskInfo, &pInfo->pSlotIds, &pInfo->pDstSlotIds); if (code != TSDB_CODE_SUCCESS) { goto _error; } @@ -93,8 +103,8 @@ SOperatorInfo* createCacherowsScanOperator(SLastRowScanPhysiNode* pScanNode, SRe uint64_t suid = tableListGetSuid(pTableListInfo); code = tsdbCacherowsReaderOpen(pInfo->readHandle.vnode, pInfo->retrieveType, pList, totalTables, - taosArrayGetSize(pInfo->matchInfo.pList), suid, &pInfo->pLastrowReader, - pTaskInfo->id.str); + taosArrayGetSize(pInfo->matchInfo.pList), pCidList, pInfo->pSlotIds, suid, + &pInfo->pLastrowReader, pTaskInfo->id.str); if (code != TSDB_CODE_SUCCESS) { goto _error; } @@ -162,8 +172,8 @@ SSDataBlock* doScanCache(SOperatorInfo* pOperator) { blockDataCleanup(pInfo->pBufferredRes); taosArrayClear(pInfo->pUidList); - int32_t code = - tsdbRetrieveCacheRows(pInfo->pLastrowReader, pInfo->pBufferredRes, pInfo->pSlotIds, pInfo->pUidList); + int32_t code = tsdbRetrieveCacheRows(pInfo->pLastrowReader, pInfo->pBufferredRes, pInfo->pSlotIds, + pInfo->pDstSlotIds, pInfo->pUidList); if (code != TSDB_CODE_SUCCESS) { T_LONG_JMP(pTaskInfo->env, code); } @@ -229,8 +239,8 @@ SSDataBlock* doScanCache(SOperatorInfo* pOperator) { } code = tsdbCacherowsReaderOpen(pInfo->readHandle.vnode, pInfo->retrieveType, pList, num, - taosArrayGetSize(pInfo->matchInfo.pList), suid, &pInfo->pLastrowReader, - pTaskInfo->id.str); + taosArrayGetSize(pInfo->matchInfo.pList), pInfo->pCidList, pInfo->pSlotIds, suid, + &pInfo->pLastrowReader, pTaskInfo->id.str); if (code != TSDB_CODE_SUCCESS) { pInfo->currentGroupIndex += 1; taosArrayClear(pInfo->pUidList); @@ -239,7 +249,8 @@ SSDataBlock* doScanCache(SOperatorInfo* pOperator) { taosArrayClear(pInfo->pUidList); - code = tsdbRetrieveCacheRows(pInfo->pLastrowReader, pInfo->pRes, pInfo->pSlotIds, pInfo->pUidList); + code = tsdbRetrieveCacheRows(pInfo->pLastrowReader, pInfo->pRes, pInfo->pSlotIds, pInfo->pDstSlotIds, + pInfo->pUidList); if (code != TSDB_CODE_SUCCESS) { T_LONG_JMP(pTaskInfo->env, code); } @@ -282,6 +293,8 @@ void destroyCacheScanOperator(void* param) { blockDataDestroy(pInfo->pRes); blockDataDestroy(pInfo->pBufferredRes); taosMemoryFree(pInfo->pSlotIds); + taosMemoryFree(pInfo->pDstSlotIds); + taosArrayDestroy(pInfo->pCidList); taosArrayDestroy(pInfo->pUidList); taosArrayDestroy(pInfo->matchInfo.pList); tableListDestroy(pInfo->pTableList); @@ -294,7 +307,8 @@ void destroyCacheScanOperator(void* param) { taosMemoryFreeClear(param); } -int32_t extractCacheScanSlotId(const SArray* pColMatchInfo, SExecTaskInfo* pTaskInfo, int32_t** pSlotIds) { +int32_t extractCacheScanSlotId(const SArray* pColMatchInfo, SExecTaskInfo* pTaskInfo, int32_t** pSlotIds, + int32_t** pDstSlotIds) { size_t numOfCols = taosArrayGetSize(pColMatchInfo); *pSlotIds = taosMemoryMalloc(numOfCols * sizeof(int32_t)); @@ -302,18 +316,25 @@ int32_t extractCacheScanSlotId(const SArray* pColMatchInfo, SExecTaskInfo* pTask return TSDB_CODE_OUT_OF_MEMORY; } + *pDstSlotIds = taosMemoryMalloc(numOfCols * sizeof(int32_t)); + if (*pDstSlotIds == NULL) { + taosMemoryFree(*pSlotIds); + return TSDB_CODE_OUT_OF_MEMORY; + } + SSchemaWrapper* pWrapper = pTaskInfo->schemaInfo.sw; for (int32_t i = 0; i < numOfCols; ++i) { SColMatchItem* pColMatch = taosArrayGet(pColMatchInfo, i); for (int32_t j = 0; j < pWrapper->nCols; ++j) { - if (pColMatch->colId == pWrapper->pSchema[j].colId && pColMatch->colId == PRIMARYKEY_TIMESTAMP_COL_ID) { + /* if (pColMatch->colId == pWrapper->pSchema[j].colId && pColMatch->colId == PRIMARYKEY_TIMESTAMP_COL_ID) { (*pSlotIds)[pColMatch->dstSlotId] = -1; break; - } + }*/ if (pColMatch->colId == pWrapper->pSchema[j].colId) { - (*pSlotIds)[pColMatch->dstSlotId] = j; + (*pSlotIds)[i] = j; + (*pDstSlotIds)[i] = pColMatch->dstSlotId; break; } } diff --git a/source/libs/executor/src/groupoperator.c b/source/libs/executor/src/groupoperator.c index 280edf8b53..7ad8821ff9 100644 --- a/source/libs/executor/src/groupoperator.c +++ b/source/libs/executor/src/groupoperator.c @@ -872,9 +872,9 @@ SOperatorInfo* createPartitionOperatorInfo(SOperatorInfo* downstream, SPartition getBufferPgSize(pInfo->binfo.pRes->info.rowSize, &defaultPgsz, &defaultBufsz); if (!osTempSpaceAvailable()) { - terrno = TSDB_CODE_NO_AVAIL_DISK; + terrno = TSDB_CODE_NO_DISKSPACE; pTaskInfo->code = terrno; - qError("Create partition operator info failed since %s", terrstr(terrno)); + qError("Create partition operator info failed since %s, tempDir:%s", terrstr(), tsTempDir); goto _error; } diff --git a/source/libs/executor/src/scanoperator.c b/source/libs/executor/src/scanoperator.c index e1fa7a282b..af310305af 100644 --- a/source/libs/executor/src/scanoperator.c +++ b/source/libs/executor/src/scanoperator.c @@ -2515,6 +2515,51 @@ _error: return NULL; } +static void doTagScanOneTable(SOperatorInfo* pOperator, const SSDataBlock* pRes, int32_t count, SMetaReader* mr) { + SExecTaskInfo* pTaskInfo = pOperator->pTaskInfo; + STagScanInfo* pInfo = pOperator->info; + SExprInfo* pExprInfo = &pOperator->exprSupp.pExprInfo[0]; + + STableKeyInfo* item = tableListGetInfo(pInfo->pTableListInfo, pInfo->curPos); + int32_t code = metaGetTableEntryByUid(mr, item->uid); + tDecoderClear(&(*mr).coder); + if (code != TSDB_CODE_SUCCESS) { + qError("failed to get table meta, uid:0x%" PRIx64 ", code:%s, %s", item->uid, tstrerror(terrno), + GET_TASKID(pTaskInfo)); + metaReaderClear(mr); + T_LONG_JMP(pTaskInfo->env, terrno); + } + + char str[512]; + for (int32_t j = 0; j < pOperator->exprSupp.numOfExprs; ++j) { + SColumnInfoData* pDst = taosArrayGet(pRes->pDataBlock, pExprInfo[j].base.resSchema.slotId); + + // refactor later + if (fmIsScanPseudoColumnFunc(pExprInfo[j].pExpr->_function.functionId)) { + STR_TO_VARSTR(str, (*mr).me.name); + colDataSetVal(pDst, (count), str, false); + } else { // it is a tag value + STagVal val = {0}; + val.cid = pExprInfo[j].base.pParam[0].pCol->colId; + const char* p = metaGetTableTagVal((*mr).me.ctbEntry.pTags, pDst->info.type, &val); + + char* data = NULL; + if (pDst->info.type != TSDB_DATA_TYPE_JSON && p != NULL) { + data = tTagValToData((const STagVal*)p, false); + } else { + data = (char*)p; + } + colDataSetVal(pDst, (count), data, + (data == NULL) || (pDst->info.type == TSDB_DATA_TYPE_JSON && tTagIsJsonNull(data))); + + if (pDst->info.type != TSDB_DATA_TYPE_JSON && p != NULL && IS_VAR_DATA_TYPE(((const STagVal*)p)->type) && + data != NULL) { + taosMemoryFree(data); + } + } + } +} + static SSDataBlock* doTagScan(SOperatorInfo* pOperator) { if (pOperator->status == OP_EXEC_DONE) { return NULL; @@ -2539,48 +2584,22 @@ static SSDataBlock* doTagScan(SOperatorInfo* pOperator) { metaReaderInit(&mr, pInfo->readHandle.meta, 0); while (pInfo->curPos < size && count < pOperator->resultInfo.capacity) { - STableKeyInfo* item = tableListGetInfo(pInfo->pTableListInfo, pInfo->curPos); - int32_t code = metaGetTableEntryByUid(&mr, item->uid); - tDecoderClear(&mr.coder); - if (code != TSDB_CODE_SUCCESS) { - qError("failed to get table meta, uid:0x%" PRIx64 ", code:%s, %s", item->uid, tstrerror(terrno), - GET_TASKID(pTaskInfo)); - metaReaderClear(&mr); - T_LONG_JMP(pTaskInfo->env, terrno); - } - - for (int32_t j = 0; j < pOperator->exprSupp.numOfExprs; ++j) { - SColumnInfoData* pDst = taosArrayGet(pRes->pDataBlock, pExprInfo[j].base.resSchema.slotId); - - // refactor later - if (fmIsScanPseudoColumnFunc(pExprInfo[j].pExpr->_function.functionId)) { - STR_TO_VARSTR(str, mr.me.name); - colDataSetVal(pDst, count, str, false); - } else { // it is a tag value - STagVal val = {0}; - val.cid = pExprInfo[j].base.pParam[0].pCol->colId; - const char* p = metaGetTableTagVal(mr.me.ctbEntry.pTags, pDst->info.type, &val); - - char* data = NULL; - if (pDst->info.type != TSDB_DATA_TYPE_JSON && p != NULL) { - data = tTagValToData((const STagVal*)p, false); - } else { - data = (char*)p; - } - colDataSetVal(pDst, count, data, - (data == NULL) || (pDst->info.type == TSDB_DATA_TYPE_JSON && tTagIsJsonNull(data))); - - if (pDst->info.type != TSDB_DATA_TYPE_JSON && p != NULL && IS_VAR_DATA_TYPE(((const STagVal*)p)->type) && - data != NULL) { - taosMemoryFree(data); - } - } - } - - count += 1; + doTagScanOneTable(pOperator, pRes, count, &mr); + ++count; if (++pInfo->curPos >= size) { setOperatorCompleted(pOperator); } + // each table with tbname is a group, hence its own block, but only group when slimit exists for performance reason. + if (pInfo->pSlimit != NULL) { + if (pInfo->curPos < pInfo->pSlimit->offset) { + continue; + } + pInfo->pRes->info.id.groupId = calcGroupId(mr.me.name, strlen(mr.me.name)); + if (pInfo->curPos >= (pInfo->pSlimit->offset + pInfo->pSlimit->limit) - 1) { + setOperatorCompleted(pOperator); + } + break; + } } metaReaderClear(&mr); @@ -2631,6 +2650,7 @@ SOperatorInfo* createTagScanOperatorInfo(SReadHandle* pReadHandle, STagScanPhysi pInfo->pRes = createDataBlockFromDescNode(pDescNode); pInfo->readHandle = *pReadHandle; pInfo->curPos = 0; + pInfo->pSlimit = (SLimitNode*)pPhyNode->node.pSlimit; //TODO: slimit now only indicate group setOperatorInfo(pOperator, "TagScanOperator", QUERY_NODE_PHYSICAL_PLAN_TAG_SCAN, false, OP_NOT_OPENED, pInfo, pTaskInfo); diff --git a/source/libs/executor/src/sortoperator.c b/source/libs/executor/src/sortoperator.c index 10933f285c..718bb596c5 100644 --- a/source/libs/executor/src/sortoperator.c +++ b/source/libs/executor/src/sortoperator.c @@ -547,6 +547,7 @@ typedef struct SMultiwayMergeOperatorInfo { SSDataBlock* pIntermediateBlock; // to hold the intermediate result int64_t startTs; // sort start time bool groupSort; + bool ignoreGroupId; uint64_t groupId; STupleHandle* prefetchedTuple; } SMultiwayMergeOperatorInfo; @@ -696,7 +697,11 @@ SSDataBlock* getMultiwaySortedBlockData(SSortHandle* pHandle, SSDataBlock* pData } pDataBlock->info.rows = p->info.rows; - pDataBlock->info.id.groupId = pInfo->groupId; + if (pInfo->ignoreGroupId) { + pDataBlock->info.id.groupId = 0; + } else { + pDataBlock->info.id.groupId = pInfo->groupId; + } pDataBlock->info.dataLoad = 1; } @@ -787,6 +792,7 @@ SOperatorInfo* createMultiwayMergeOperatorInfo(SOperatorInfo** downStreams, size blockDataEnsureCapacity(pInfo->binfo.pRes, pOperator->resultInfo.capacity); pInfo->groupSort = pMergePhyNode->groupSort; + pInfo->ignoreGroupId = pMergePhyNode->ignoreGroupId; pInfo->pSortInfo = createSortInfo(pMergePhyNode->pMergeKeys); pInfo->pInputBlock = pInputBlock; size_t numOfCols = taosArrayGetSize(pInfo->binfo.pRes->pDataBlock); diff --git a/source/libs/executor/src/timewindowoperator.c b/source/libs/executor/src/timewindowoperator.c index 27f9a1ac28..63b60b6d2c 100644 --- a/source/libs/executor/src/timewindowoperator.c +++ b/source/libs/executor/src/timewindowoperator.c @@ -1106,7 +1106,7 @@ static void doStateWindowAggImpl(SOperatorInfo* pOperator, SStateWindowOperatorI bool masterScan = true; int32_t numOfOutput = pOperator->exprSupp.numOfExprs; - int16_t bytes = pStateColInfoData->info.bytes; + int32_t bytes = pStateColInfoData->info.bytes; SColumnInfoData* pColInfoData = taosArrayGet(pBlock->pDataBlock, pInfo->tsSlotId); TSKEY* tsList = (TSKEY*)pColInfoData->pData; @@ -2892,8 +2892,8 @@ int32_t initStreamAggSupporter(SStreamAggSupporter* pSup, SqlFunctionCtx* pCtx, bufSize = pageSize * 4; } if (!osTempSpaceAvailable()) { - terrno = TSDB_CODE_NO_AVAIL_DISK; - qError("Init stream agg supporter failed since %s", terrstr(terrno)); + terrno = TSDB_CODE_NO_DISKSPACE; + qError("Init stream agg supporter failed since %s, tempDir:%s", terrstr(), tsTempDir); return terrno; } int32_t code = createDiskbasedBuf(&pSup->pResultBuf, pageSize, bufSize, "function", tsTempDir); diff --git a/source/libs/executor/src/tlinearhash.c b/source/libs/executor/src/tlinearhash.c index 2cba3855c7..023583fcde 100644 --- a/source/libs/executor/src/tlinearhash.c +++ b/source/libs/executor/src/tlinearhash.c @@ -248,8 +248,8 @@ SLHashObj* tHashInit(int32_t inMemPages, int32_t pageSize, _hash_fn_t fn, int32_ } if (!osTempSpaceAvailable()) { - terrno = TSDB_CODE_NO_AVAIL_DISK; - printf("tHash Init failed since %s", terrstr(terrno)); + terrno = TSDB_CODE_NO_DISKSPACE; + printf("tHash Init failed since %s, tempDir:%s", terrstr(), tsTempDir); taosMemoryFree(pHashObj); return NULL; } diff --git a/source/libs/executor/src/tsort.c b/source/libs/executor/src/tsort.c index 36e41b0c5d..211f78b981 100644 --- a/source/libs/executor/src/tsort.c +++ b/source/libs/executor/src/tsort.c @@ -195,8 +195,8 @@ static int32_t doAddToBuf(SSDataBlock* pDataBlock, SSortHandle* pHandle) { if (pHandle->pBuf == NULL) { if (!osTempSpaceAvailable()) { - terrno = TSDB_CODE_NO_AVAIL_DISK; - qError("Add to buf failed since %s", terrstr(terrno)); + terrno = TSDB_CODE_NO_DISKSPACE; + qError("Add to buf failed since %s, tempDir:%s", terrstr(), tsTempDir); return terrno; } @@ -261,9 +261,8 @@ static int32_t sortComparInit(SMsortComparParam* pParam, SArray* pSources, int32 // multi-pass internal merge sort is required if (pHandle->pBuf == NULL) { if (!osTempSpaceAvailable()) { - code = TSDB_CODE_NO_AVAIL_DISK; - terrno = code; - qError("Sort compare init failed since %s, %s", tstrerror(code), pHandle->idStr); + code = terrno = TSDB_CODE_NO_DISKSPACE; + qError("Sort compare init failed since %s, tempDir:%s, idStr:%s", terrstr(), tsTempDir, pHandle->idStr); return code; } diff --git a/source/libs/function/inc/tpercentile.h b/source/libs/function/inc/tpercentile.h index 80159460f5..65b7b38a05 100644 --- a/source/libs/function/inc/tpercentile.h +++ b/source/libs/function/inc/tpercentile.h @@ -53,7 +53,7 @@ typedef int32_t (*__perc_hash_func_t)(struct tMemBucket *pBucket, const void *va typedef struct tMemBucket { int16_t numOfSlots; int16_t type; - int16_t bytes; + int32_t bytes; int32_t total; int32_t elemPerPage; // number of elements for each object int32_t maxCapacity; // maximum allowed number of elements that can be sort directly to get the result @@ -67,7 +67,7 @@ typedef struct tMemBucket { SHashObj *groupPagesMap; // disk page map for different groups; } tMemBucket; -tMemBucket *tMemBucketCreate(int16_t nElemSize, int16_t dataType, double minval, double maxval); +tMemBucket *tMemBucketCreate(int32_t nElemSize, int16_t dataType, double minval, double maxval); void tMemBucketDestroy(tMemBucket *pBucket); diff --git a/source/libs/function/src/builtinsimpl.c b/source/libs/function/src/builtinsimpl.c index af318a6bc5..80b26bd39b 100644 --- a/source/libs/function/src/builtinsimpl.c +++ b/source/libs/function/src/builtinsimpl.c @@ -196,11 +196,11 @@ typedef struct SMavgInfo { } SMavgInfo; typedef struct SSampleInfo { - int32_t samples; - int32_t totalPoints; - int32_t numSampled; - uint8_t colType; - int16_t colBytes; + int32_t samples; + int32_t totalPoints; + int32_t numSampled; + uint8_t colType; + uint16_t colBytes; STuplePos nullTuplePos; bool nullTupleSaved; @@ -220,7 +220,7 @@ typedef struct STailInfo { int32_t numAdded; int32_t offset; uint8_t colType; - int16_t colBytes; + uint16_t colBytes; STailItem** pItems; } STailInfo; @@ -233,7 +233,7 @@ typedef struct SUniqueItem { typedef struct SUniqueInfo { int32_t numOfPoints; uint8_t colType; - int16_t colBytes; + uint16_t colBytes; bool hasNull; // null is not hashable, handle separately SHashObj* pHash; char pItems[]; @@ -247,13 +247,13 @@ typedef struct SModeItem { typedef struct SModeInfo { uint8_t colType; - int16_t colBytes; + uint16_t colBytes; SHashObj* pHash; STuplePos nullTuplePos; bool nullTupleSaved; - char* buf; // serialize data buffer + char* buf; // serialize data buffer } SModeInfo; typedef struct SDerivInfo { @@ -855,7 +855,9 @@ int32_t setSelectivityValue(SqlFunctionCtx* pCtx, SSDataBlock* pBlock, const STu int32_t numOfCols = pCtx->subsidiaries.num; const char* p = loadTupleData(pCtx, pTuplePos); if (p == NULL) { - terrno = TSDB_CODE_NO_AVAIL_DISK; + terrno = TSDB_CODE_NOT_FOUND; + qError("Load tuple data failed since %s, groupId:%" PRIu64 ", ts:%" PRId64, terrstr(), + pTuplePos->streamTupleKey.groupId, pTuplePos->streamTupleKey.ts); return terrno; } @@ -5098,7 +5100,9 @@ int32_t modeFinalize(SqlFunctionCtx* pCtx, SSDataBlock* pBlock) { if (maxCount != 0) { const char* pData = loadTupleData(pCtx, &resDataPos); if (pData == NULL) { - code = TSDB_CODE_NO_AVAIL_DISK; + code = terrno = TSDB_CODE_NOT_FOUND; + qError("Load tuple data failed since %s, groupId:%" PRIu64 ", ts:%" PRId64, terrstr(), + resDataPos.streamTupleKey.groupId, resDataPos.streamTupleKey.ts); modeFunctionCleanup(pInfo); return code; } diff --git a/source/libs/function/src/tpercentile.c b/source/libs/function/src/tpercentile.c index de381fadbd..3ec802a7ce 100644 --- a/source/libs/function/src/tpercentile.c +++ b/source/libs/function/src/tpercentile.c @@ -236,7 +236,7 @@ static void resetSlotInfo(tMemBucket *pBucket) { } } -tMemBucket *tMemBucketCreate(int16_t nElemSize, int16_t dataType, double minval, double maxval) { +tMemBucket *tMemBucketCreate(int32_t nElemSize, int16_t dataType, double minval, double maxval) { tMemBucket *pBucket = (tMemBucket *)taosMemoryCalloc(1, sizeof(tMemBucket)); if (pBucket == NULL) { return NULL; @@ -277,7 +277,7 @@ tMemBucket *tMemBucketCreate(int16_t nElemSize, int16_t dataType, double minval, resetSlotInfo(pBucket); if (!osTempSpaceAvailable()) { - terrno = TSDB_CODE_NO_AVAIL_DISK; + terrno = TSDB_CODE_NO_DISKSPACE; // qError("MemBucket create disk based Buf failed since %s", terrstr(terrno)); tMemBucketDestroy(pBucket); return NULL; diff --git a/source/libs/function/src/tudf.c b/source/libs/function/src/tudf.c index 8c8b99a6f8..6b70422ac8 100644 --- a/source/libs/function/src/tudf.c +++ b/source/libs/function/src/tudf.c @@ -1580,7 +1580,7 @@ int32_t udfcStartUvTask(SClientUvTaskNode *uvTask) { case UV_TASK_REQ_RSP: { uv_pipe_t *pipe = uvTask->pipe; if (pipe == NULL) { - code = TSDB_CODE_UDF_PIPE_NO_PIPE; + code = TSDB_CODE_UDF_PIPE_NOT_EXIST; } else { uv_write_t *write = taosMemoryMalloc(sizeof(uv_write_t)); write->data = pipe->data; @@ -1598,7 +1598,7 @@ int32_t udfcStartUvTask(SClientUvTaskNode *uvTask) { case UV_TASK_DISCONNECT: { uv_pipe_t *pipe = uvTask->pipe; if (pipe == NULL) { - code = TSDB_CODE_UDF_PIPE_NO_PIPE; + code = TSDB_CODE_UDF_PIPE_NOT_EXIST; } else { SClientUvConn *conn = pipe->data; QUEUE_INSERT_TAIL(&conn->taskQueue, &uvTask->connTaskQueue); @@ -1759,9 +1759,6 @@ int32_t udfcRunUdfUvTask(SClientUdfTask *task, int8_t uvTaskType) { } int32_t doSetupUdf(char udfName[], UdfcFuncHandle *funcHandle) { - if (gUdfcProxy.udfcState != UDFC_STATE_READY) { - return TSDB_CODE_UDF_INVALID_STATE; - } SClientUdfTask *task = taosMemoryCalloc(1, sizeof(SClientUdfTask)); task->errCode = 0; task->session = taosMemoryCalloc(1, sizeof(SUdfcUvSession)); @@ -1804,7 +1801,7 @@ int32_t callUdf(UdfcFuncHandle handle, int8_t callType, SSDataBlock *input, SUdf SUdfcUvSession *session = (SUdfcUvSession *)handle; if (session->udfUvPipe == NULL) { fnError("No pipe to udfd"); - return TSDB_CODE_UDF_PIPE_NO_PIPE; + return TSDB_CODE_UDF_PIPE_NOT_EXIST; } SClientUdfTask *task = taosMemoryCalloc(1, sizeof(SClientUdfTask)); task->errCode = 0; @@ -1928,7 +1925,7 @@ int32_t doTeardownUdf(UdfcFuncHandle handle) { if (session->udfUvPipe == NULL) { fnError("tear down udf. pipe to udfd does not exist. udf name: %s", session->udfName); taosMemoryFree(session); - return TSDB_CODE_UDF_PIPE_NO_PIPE; + return TSDB_CODE_UDF_PIPE_NOT_EXIST; } SClientUdfTask *task = taosMemoryCalloc(1, sizeof(SClientUdfTask)); diff --git a/source/libs/function/src/udfd.c b/source/libs/function/src/udfd.c index 5034af2f82..d2553ba96d 100644 --- a/source/libs/function/src/udfd.c +++ b/source/libs/function/src/udfd.c @@ -844,7 +844,7 @@ void udfdGetFuncBodyPath(const SUdf *udf, char *path) { int32_t udfdSaveFuncBodyToFile(SFuncInfo *pFuncInfo, SUdf *udf) { if (!osDataSpaceAvailable()) { - terrno = TSDB_CODE_NO_AVAIL_DISK; + terrno = TSDB_CODE_NO_DISKSPACE; fnError("udfd create shared library failed since %s", terrstr(terrno)); return terrno; } diff --git a/source/libs/nodes/src/nodesCloneFuncs.c b/source/libs/nodes/src/nodesCloneFuncs.c index d9a4c5178f..0f4e7bde63 100644 --- a/source/libs/nodes/src/nodesCloneFuncs.c +++ b/source/libs/nodes/src/nodesCloneFuncs.c @@ -455,6 +455,7 @@ static int32_t logicMergeCopy(const SMergeLogicNode* pSrc, SMergeLogicNode* pDst COPY_SCALAR_FIELD(numOfChannels); COPY_SCALAR_FIELD(srcGroupId); COPY_SCALAR_FIELD(groupSort); + COPY_SCALAR_FIELD(ignoreGroupId); return TSDB_CODE_SUCCESS; } diff --git a/source/libs/nodes/src/nodesCodeFuncs.c b/source/libs/nodes/src/nodesCodeFuncs.c index 919be78125..0ec757d6cc 100644 --- a/source/libs/nodes/src/nodesCodeFuncs.c +++ b/source/libs/nodes/src/nodesCodeFuncs.c @@ -2031,6 +2031,7 @@ static const char* jkMergePhysiPlanTargets = "Targets"; static const char* jkMergePhysiPlanNumOfChannels = "NumOfChannels"; static const char* jkMergePhysiPlanSrcGroupId = "SrcGroupId"; static const char* jkMergePhysiPlanGroupSort = "GroupSort"; +static const char* jkMergePhysiPlanIgnoreGroupID = "IgnoreGroupID"; static int32_t physiMergeNodeToJson(const void* pObj, SJson* pJson) { const SMergePhysiNode* pNode = (const SMergePhysiNode*)pObj; @@ -2051,6 +2052,9 @@ static int32_t physiMergeNodeToJson(const void* pObj, SJson* pJson) { if (TSDB_CODE_SUCCESS == code) { code = tjsonAddBoolToObject(pJson, jkMergePhysiPlanGroupSort, pNode->groupSort); } + if (TSDB_CODE_SUCCESS == code) { + code = tjsonAddBoolToObject(pJson, jkMergePhysiPlanIgnoreGroupID, pNode->ignoreGroupId); + } return code; } @@ -2074,6 +2078,9 @@ static int32_t jsonToPhysiMergeNode(const SJson* pJson, void* pObj) { if (TSDB_CODE_SUCCESS == code) { code = tjsonGetBoolValue(pJson, jkMergePhysiPlanGroupSort, &pNode->groupSort); } + if (TSDB_CODE_SUCCESS == code) { + code = tjsonGetBoolValue(pJson, jkMergePhysiPlanIgnoreGroupID, &pNode->ignoreGroupId); + } return code; } diff --git a/source/libs/nodes/src/nodesMsgFuncs.c b/source/libs/nodes/src/nodesMsgFuncs.c index 6c6b6c0e81..c06eb62771 100644 --- a/source/libs/nodes/src/nodesMsgFuncs.c +++ b/source/libs/nodes/src/nodesMsgFuncs.c @@ -2512,7 +2512,8 @@ enum { PHY_MERGE_CODE_TARGETS, PHY_MERGE_CODE_NUM_OF_CHANNELS, PHY_MERGE_CODE_SRC_GROUP_ID, - PHY_MERGE_CODE_GROUP_SORT + PHY_MERGE_CODE_GROUP_SORT, + PHY_MERGE_CODE_IGNORE_GROUP_ID, }; static int32_t physiMergeNodeToMsg(const void* pObj, STlvEncoder* pEncoder) { @@ -2534,6 +2535,9 @@ static int32_t physiMergeNodeToMsg(const void* pObj, STlvEncoder* pEncoder) { if (TSDB_CODE_SUCCESS == code) { code = tlvEncodeBool(pEncoder, PHY_MERGE_CODE_GROUP_SORT, pNode->groupSort); } + if (TSDB_CODE_SUCCESS == code) { + code = tlvEncodeBool(pEncoder, PHY_MERGE_CODE_IGNORE_GROUP_ID, pNode->ignoreGroupId); + } return code; } @@ -2563,6 +2567,9 @@ static int32_t msgToPhysiMergeNode(STlvDecoder* pDecoder, void* pObj) { case PHY_MERGE_CODE_GROUP_SORT: code = tlvDecodeBool(pTlv, &pNode->groupSort); break; + case PHY_MERGE_CODE_IGNORE_GROUP_ID: + code = tlvDecodeBool(pTlv, &pNode->ignoreGroupId); + break; default: break; } diff --git a/source/libs/parser/src/parAstCreater.c b/source/libs/parser/src/parAstCreater.c index 41648175b9..02af95a10c 100644 --- a/source/libs/parser/src/parAstCreater.c +++ b/source/libs/parser/src/parAstCreater.c @@ -1224,7 +1224,7 @@ SDataType createDataType(uint8_t type) { } SDataType createVarLenDataType(uint8_t type, const SToken* pLen) { - SDataType dt = {.type = type, .precision = 0, .scale = 0, .bytes = taosStr2Int16(pLen->z, NULL, 10)}; + SDataType dt = {.type = type, .precision = 0, .scale = 0, .bytes = taosStr2Int32(pLen->z, NULL, 10)}; return dt; } diff --git a/source/libs/parser/src/parInsertSml.c b/source/libs/parser/src/parInsertSml.c index 1c921b2d7c..dbf5931e8a 100644 --- a/source/libs/parser/src/parInsertSml.c +++ b/source/libs/parser/src/parInsertSml.c @@ -250,7 +250,7 @@ end: int32_t smlBindData(SQuery* query, bool dataFormat, SArray* tags, SArray* colsSchema, SArray* cols, STableMeta* pTableMeta, char* tableName, const char* sTableName, int32_t sTableNameLen, int32_t ttl, - char* msgBuf, int16_t msgBufLen) { + char* msgBuf, int32_t msgBufLen) { SMsgBuf pBuf = {.buf = msgBuf, .len = msgBufLen}; SSchema* pTagsSchema = getTableTagSchema(pTableMeta); diff --git a/source/libs/parser/src/parTranslater.c b/source/libs/parser/src/parTranslater.c index af7dced201..ec4e203133 100644 --- a/source/libs/parser/src/parTranslater.c +++ b/source/libs/parser/src/parTranslater.c @@ -2294,6 +2294,25 @@ static int32_t checkAggColCoexist(STranslateContext* pCxt, SSelectStmt* pSelect) return TSDB_CODE_SUCCESS; } +static int32_t checkHavingGroupBy(STranslateContext* pCxt, SSelectStmt* pSelect) { + int32_t code = TSDB_CODE_SUCCESS; + if (NULL == getGroupByList(pCxt) && NULL == pSelect->pPartitionByList && NULL == pSelect->pWindow) { + return code; + } + if (NULL != pSelect->pHaving) { + code = checkExprForGroupBy(pCxt, &pSelect->pHaving); + } +/* + if (TSDB_CODE_SUCCESS == code && NULL != pSelect->pProjectionList) { + code = checkExprListForGroupBy(pCxt, pSelect, pSelect->pProjectionList); + } + if (TSDB_CODE_SUCCESS == code && NULL != pSelect->pOrderByList) { + code = checkExprListForGroupBy(pCxt, pSelect, pSelect->pOrderByList); + } +*/ + return code; +} + static int32_t checkWindowFuncCoexist(STranslateContext* pCxt, SSelectStmt* pSelect) { if (NULL == pSelect->pWindow) { return TSDB_CODE_SUCCESS; @@ -3092,7 +3111,7 @@ static int32_t translateSelectList(STranslateContext* pCxt, SSelectStmt* pSelect } if (TSDB_CODE_SUCCESS == code) { code = checkExprListForGroupBy(pCxt, pSelect, pSelect->pProjectionList); - } + } if (TSDB_CODE_SUCCESS == code) { code = translateFillValues(pCxt, pSelect); } @@ -3109,9 +3128,6 @@ static int32_t translateHaving(STranslateContext* pCxt, SSelectStmt* pSelect) { } pCxt->currClause = SQL_CLAUSE_HAVING; int32_t code = translateExpr(pCxt, &pSelect->pHaving); - if (TSDB_CODE_SUCCESS == code && (NULL != pSelect->pGroupByList || NULL != pSelect->pWindow)) { - code = checkExprForGroupBy(pCxt, &pSelect->pHaving); - } return code; } @@ -3694,6 +3710,9 @@ static int32_t translateSelectFrom(STranslateContext* pCxt, SSelectStmt* pSelect if (TSDB_CODE_SUCCESS == code) { code = translateSelectList(pCxt, pSelect); } + if (TSDB_CODE_SUCCESS == code) { + code = checkHavingGroupBy(pCxt, pSelect); + } if (TSDB_CODE_SUCCESS == code) { code = translateOrderBy(pCxt, pSelect); } @@ -4593,8 +4612,8 @@ static int32_t checkTableTagsSchema(STranslateContext* pCxt, SHashObj* pHash, SN code = generateSyntaxErrMsg(&pCxt->msgBuf, TSDB_CODE_PAR_ONLY_ONE_JSON_TAG); } if (TSDB_CODE_SUCCESS == code) { - if ((TSDB_DATA_TYPE_VARCHAR == pTag->dataType.type && calcTypeBytes(pTag->dataType) > TSDB_MAX_BINARY_LEN) || - (TSDB_DATA_TYPE_NCHAR == pTag->dataType.type && calcTypeBytes(pTag->dataType) > TSDB_MAX_NCHAR_LEN)) { + if ((TSDB_DATA_TYPE_VARCHAR == pTag->dataType.type && calcTypeBytes(pTag->dataType) > TSDB_MAX_TAGS_LEN) || + (TSDB_DATA_TYPE_NCHAR == pTag->dataType.type && calcTypeBytes(pTag->dataType) > TSDB_MAX_TAGS_LEN)) { code = generateSyntaxErrMsg(&pCxt->msgBuf, TSDB_CODE_PAR_INVALID_VAR_COLUMN_LEN); } } @@ -5340,7 +5359,7 @@ static int32_t checkAlterSuperTableBySchema(STranslateContext* pCxt, SAlterTable } if (TSDB_ALTER_TABLE_UPDATE_TAG_BYTES == pStmt->alterType) { - if (calcTypeBytes(pStmt->dataType) > TSDB_MAX_FIELD_LEN) { + if (calcTypeBytes(pStmt->dataType) > TSDB_MAX_TAGS_LEN) { return generateSyntaxErrMsg(&pCxt->msgBuf, TSDB_CODE_PAR_INVALID_VAR_COLUMN_LEN); } diff --git a/source/libs/parser/test/parSelectTest.cpp b/source/libs/parser/test/parSelectTest.cpp index ec6c69ea8d..2d8ce55b72 100644 --- a/source/libs/parser/test/parSelectTest.cpp +++ b/source/libs/parser/test/parSelectTest.cpp @@ -239,6 +239,19 @@ TEST_F(ParserSelectTest, groupBySemanticCheck) { run("SELECT COUNT(*) cnt, c2 FROM t1 WHERE c1 > 0 GROUP BY c1", TSDB_CODE_PAR_GROUPBY_LACK_EXPRESSION); } +TEST_F(ParserSelectTest, havingCheck) { + useDb("root", "test"); + + run("select tbname,count(*) from st1 partition by tbname having c1>0", TSDB_CODE_PAR_INVALID_OPTR_USAGE); + + run("select tbname,count(*) from st1 group by tbname having c1>0", TSDB_CODE_PAR_GROUPBY_LACK_EXPRESSION); + + run("select max(c1) from st1 group by tbname having c1>0"); + + run("select max(c1) from st1 partition by tbname having c1>0"); +} + + TEST_F(ParserSelectTest, orderBy) { useDb("root", "test"); diff --git a/source/libs/planner/src/planOptimizer.c b/source/libs/planner/src/planOptimizer.c index 0216a214c2..ea6ffed1f6 100644 --- a/source/libs/planner/src/planOptimizer.c +++ b/source/libs/planner/src/planOptimizer.c @@ -2418,6 +2418,36 @@ static bool tagScanOptShouldBeOptimized(SLogicNode* pNode) { return true; } +static SLogicNode* tagScanOptFindAncestorWithSlimit(SLogicNode* pTableScanNode) { + SLogicNode* pNode = pTableScanNode->pParent; + while (NULL != pNode) { + if (QUERY_NODE_LOGIC_PLAN_PARTITION == nodeType(pNode) || QUERY_NODE_LOGIC_PLAN_AGG == nodeType(pNode) || + QUERY_NODE_LOGIC_PLAN_WINDOW == nodeType(pNode) || QUERY_NODE_LOGIC_PLAN_SORT == nodeType(pNode)) { + return NULL; + } + if (NULL != pNode->pSlimit) { + return pNode; + } + pNode = pNode->pParent; + } + return NULL; +} + +static void tagScanOptCloneAncestorSlimit(SLogicNode* pTableScanNode) { + if (NULL != pTableScanNode->pSlimit) { + return; + } + + SLogicNode* pNode = tagScanOptFindAncestorWithSlimit(pTableScanNode); + if (NULL != pNode) { + //TODO: only set the slimit now. push down slimit later + pTableScanNode->pSlimit = nodesCloneNode(pNode->pSlimit); + ((SLimitNode*)pTableScanNode->pSlimit)->limit += ((SLimitNode*)pTableScanNode->pSlimit)->offset; + ((SLimitNode*)pTableScanNode->pSlimit)->offset = 0; + } + return; +} + static int32_t tagScanOptimize(SOptimizeContext* pCxt, SLogicSubplan* pLogicSubplan) { SScanLogicNode* pScanNode = (SScanLogicNode*)optFindPossibleNode(pLogicSubplan->pNode, tagScanOptShouldBeOptimized); if (NULL == pScanNode) { @@ -2458,6 +2488,7 @@ static int32_t tagScanOptimize(SOptimizeContext* pCxt, SLogicSubplan* pLogicSubp NODES_CLEAR_LIST(pAgg->pChildren); } nodesDestroyNode((SNode*)pAgg); + tagScanOptCloneAncestorSlimit((SLogicNode*)pScanNode); pCxt->optimized = true; return TSDB_CODE_SUCCESS; } diff --git a/source/libs/planner/src/planPhysiCreater.c b/source/libs/planner/src/planPhysiCreater.c index e2c2e4c655..be43bb008c 100644 --- a/source/libs/planner/src/planPhysiCreater.c +++ b/source/libs/planner/src/planPhysiCreater.c @@ -1559,6 +1559,7 @@ static int32_t createMergePhysiNode(SPhysiPlanContext* pCxt, SMergeLogicNode* pM pMerge->numOfChannels = pMergeLogicNode->numOfChannels; pMerge->srcGroupId = pMergeLogicNode->srcGroupId; pMerge->groupSort = pMergeLogicNode->groupSort; + pMerge->ignoreGroupId = pMergeLogicNode->ignoreGroupId; int32_t code = addDataBlockSlots(pCxt, pMergeLogicNode->pInputs, pMerge->node.pOutputDataBlockDesc); diff --git a/source/libs/planner/src/planSpliter.c b/source/libs/planner/src/planSpliter.c index fd77261818..504db0d07b 100644 --- a/source/libs/planner/src/planSpliter.c +++ b/source/libs/planner/src/planSpliter.c @@ -532,6 +532,25 @@ static int32_t stbSplGetNumOfVgroups(SLogicNode* pNode) { return 0; } +static int32_t stbSplRewriteFromMergeNode(SMergeLogicNode* pMerge, SLogicNode* pNode) { + int32_t code = TSDB_CODE_SUCCESS; + + switch (nodeType(pNode)) { + case QUERY_NODE_LOGIC_PLAN_PROJECT: { + SProjectLogicNode *pLogicNode = (SProjectLogicNode*)pNode; + if (pLogicNode->ignoreGroupId && (pMerge->node.pLimit || pMerge->node.pSlimit)) { + pMerge->ignoreGroupId = true; + pLogicNode->ignoreGroupId = false; + } + break; + } + default: + break; + } + + return code; +} + static int32_t stbSplCreateMergeNode(SSplitContext* pCxt, SLogicSubplan* pSubplan, SLogicNode* pSplitNode, SNodeList* pMergeKeys, SLogicNode* pPartChild, bool groupSort) { SMergeLogicNode* pMerge = (SMergeLogicNode*)nodesMakeNode(QUERY_NODE_LOGIC_PLAN_MERGE); @@ -563,6 +582,9 @@ static int32_t stbSplCreateMergeNode(SSplitContext* pCxt, SLogicSubplan* pSubpla ((SLimitNode*)pSplitNode->pLimit)->limit += ((SLimitNode*)pSplitNode->pLimit)->offset; ((SLimitNode*)pSplitNode->pLimit)->offset = 0; } + if (TSDB_CODE_SUCCESS == code) { + code = stbSplRewriteFromMergeNode(pMerge, pSplitNode); + } if (TSDB_CODE_SUCCESS == code) { if (NULL == pSubplan) { code = nodesListMakeAppend(&pSplitNode->pChildren, (SNode*)pMerge); diff --git a/source/libs/scalar/src/sclfunc.c b/source/libs/scalar/src/sclfunc.c index 42c2c7f360..1afcf44958 100644 --- a/source/libs/scalar/src/sclfunc.c +++ b/source/libs/scalar/src/sclfunc.c @@ -12,7 +12,7 @@ typedef double (*_double_fn)(double); typedef double (*_double_fn_2)(double, double); typedef int (*_conv_fn)(int); typedef void (*_trim_fn)(char *, char *, int32_t, int32_t); -typedef int16_t (*_len_fn)(char *, int32_t); +typedef uint16_t (*_len_fn)(char *, int32_t); /** Math functions **/ static double tlog(double v) { return log(v); } @@ -286,9 +286,9 @@ static int32_t doScalarFunction(SScalarParam *pInput, int32_t inputNum, SScalarP } /** String functions **/ -static int16_t tlength(char *input, int32_t type) { return varDataLen(input); } +static VarDataLenT tlength(char *input, int32_t type) { return varDataLen(input); } -static int16_t tcharlength(char *input, int32_t type) { +static VarDataLenT tcharlength(char *input, int32_t type) { if (type == TSDB_DATA_TYPE_VARCHAR) { return varDataLen(input); } else { // NCHAR @@ -377,7 +377,7 @@ static int32_t doLengthFunction(SScalarParam *pInput, int32_t inputNum, SScalarP return TSDB_CODE_SUCCESS; } -static int32_t concatCopyHelper(const char *input, char *output, bool hasNchar, int32_t type, int16_t *dataLen) { +static int32_t concatCopyHelper(const char *input, char *output, bool hasNchar, int32_t type, VarDataLenT *dataLen) { if (hasNchar && type == TSDB_DATA_TYPE_VARCHAR) { TdUcs4 *newBuf = taosMemoryCalloc((varDataLen(input) + 1) * TSDB_NCHAR_SIZE, 1); int32_t len = varDataLen(input); @@ -457,7 +457,7 @@ int32_t concatFunction(SScalarParam *pInput, int32_t inputNum, SScalarParam *pOu continue; } - int16_t dataLen = 0; + VarDataLenT dataLen = 0; for (int32_t i = 0; i < inputNum; ++i) { int32_t rowIdx = (pInput[i].numOfRows == 1) ? 0 : k; input[i] = colDataGetData(pInputData[i], rowIdx); @@ -526,8 +526,8 @@ int32_t concatWsFunction(SScalarParam *pInput, int32_t inputNum, SScalarParam *p continue; } - int16_t dataLen = 0; - bool hasNull = false; + VarDataLenT dataLen = 0; + bool hasNull = false; for (int32_t i = 1; i < inputNum; ++i) { if (colDataIsNull_s(pInputData[i], k) || IS_NULL_TYPE(GET_PARAM_TYPE(&pInput[i]))) { hasNull = true; @@ -695,7 +695,7 @@ int32_t substrFunction(SScalarParam *pInput, int32_t inputNum, SScalarParam *pOu /** Conversion functions **/ int32_t castFunction(SScalarParam *pInput, int32_t inputNum, SScalarParam *pOutput) { int16_t inputType = GET_PARAM_TYPE(&pInput[0]); - int16_t inputLen = GET_PARAM_BYTES(&pInput[0]); + int32_t inputLen = GET_PARAM_BYTES(&pInput[0]); int16_t outputType = GET_PARAM_TYPE(&pOutput[0]); int64_t outputLen = GET_PARAM_BYTES(&pOutput[0]); diff --git a/source/libs/sync/inc/syncIndexMgr.h b/source/libs/sync/inc/syncIndexMgr.h index 4e6ab284f8..a4e83dc495 100644 --- a/source/libs/sync/inc/syncIndexMgr.h +++ b/source/libs/sync/inc/syncIndexMgr.h @@ -24,12 +24,13 @@ extern "C" { // SIndexMgr ----------------------------- typedef struct SSyncIndexMgr { - SRaftId (*replicas)[TSDB_MAX_REPLICA]; - SyncIndex index[TSDB_MAX_REPLICA]; - SyncTerm privateTerm[TSDB_MAX_REPLICA]; // for advanced function - int64_t startTimeArr[TSDB_MAX_REPLICA]; - int64_t recvTimeArr[TSDB_MAX_REPLICA]; + SRaftId (*replicas)[TSDB_MAX_REPLICA + TSDB_MAX_LEARNER_REPLICA]; + SyncIndex index[TSDB_MAX_REPLICA + TSDB_MAX_LEARNER_REPLICA]; + SyncTerm privateTerm[TSDB_MAX_REPLICA + TSDB_MAX_LEARNER_REPLICA]; // for advanced function + int64_t startTimeArr[TSDB_MAX_REPLICA + TSDB_MAX_LEARNER_REPLICA]; + int64_t recvTimeArr[TSDB_MAX_REPLICA + TSDB_MAX_LEARNER_REPLICA]; int32_t replicaNum; + int32_t totalReplicaNum; SSyncNode *pNode; } SSyncIndexMgr; diff --git a/source/libs/sync/inc/syncInt.h b/source/libs/sync/inc/syncInt.h index 6f2c1a1ad0..7d336c8313 100644 --- a/source/libs/sync/inc/syncInt.h +++ b/source/libs/sync/inc/syncInt.h @@ -54,13 +54,13 @@ typedef struct SSyncLogReplMgr SSyncLogReplMgr; #define MAX_CONFIG_INDEX_COUNT 256 typedef struct SRaftCfg { - SSyncCfg cfg; - int32_t batchSize; - int8_t isStandBy; - int8_t snapshotStrategy; - SyncIndex lastConfigIndex; - int32_t configIndexCount; - SyncIndex configIndexArr[MAX_CONFIG_INDEX_COUNT]; + SSyncCfg cfg; + int32_t batchSize; + int8_t isStandBy; + int8_t snapshotStrategy; + SyncIndex lastConfigIndex; + int32_t configIndexCount; + SyncIndex configIndexArr[MAX_CONFIG_INDEX_COUNT]; } SRaftCfg; typedef struct SRaftId { @@ -127,12 +127,13 @@ typedef struct SSyncNode { SRaftId myRaftId; int32_t peersNum; - SNodeInfo peersNodeInfo[TSDB_MAX_REPLICA]; - SEpSet peersEpset[TSDB_MAX_REPLICA]; - SRaftId peersId[TSDB_MAX_REPLICA]; + SNodeInfo peersNodeInfo[TSDB_MAX_REPLICA + TSDB_MAX_LEARNER_REPLICA]; + SEpSet peersEpset[TSDB_MAX_REPLICA + TSDB_MAX_LEARNER_REPLICA]; + SRaftId peersId[TSDB_MAX_REPLICA + TSDB_MAX_LEARNER_REPLICA]; int32_t replicaNum; - SRaftId replicasId[TSDB_MAX_REPLICA]; + int32_t totalReplicaNum; + SRaftId replicasId[TSDB_MAX_REPLICA + TSDB_MAX_LEARNER_REPLICA]; // raft algorithm SSyncFSM* pFsm; @@ -188,7 +189,7 @@ typedef struct SSyncNode { uint64_t heartbeatTimerCounter; // peer heartbeat timer - SSyncTimer peerHeartbeatTimerArr[TSDB_MAX_REPLICA]; + SSyncTimer peerHeartbeatTimerArr[TSDB_MAX_REPLICA + TSDB_MAX_LEARNER_REPLICA]; // tools SSyncRespMgr* pSyncRespMgr; @@ -196,13 +197,13 @@ typedef struct SSyncNode { // restore state bool restoreFinish; // SSnapshot* pSnapshot; - SSyncSnapshotSender* senders[TSDB_MAX_REPLICA]; + SSyncSnapshotSender* senders[TSDB_MAX_REPLICA + TSDB_MAX_LEARNER_REPLICA]; SSyncSnapshotReceiver* pNewNodeReceiver; // log replication mgr - SSyncLogReplMgr* logReplMgrs[TSDB_MAX_REPLICA]; + SSyncLogReplMgr* logReplMgrs[TSDB_MAX_REPLICA + TSDB_MAX_LEARNER_REPLICA]; - SPeerState peerStates[TSDB_MAX_REPLICA]; + SPeerState peerStates[TSDB_MAX_REPLICA + TSDB_MAX_LEARNER_REPLICA]; // is config changing bool changing; @@ -275,6 +276,7 @@ void syncNodeUpdateTerm(SSyncNode* pSyncNode, SyncTerm term); void syncNodeUpdateTermWithoutStepDown(SSyncNode* pSyncNode, SyncTerm term); void syncNodeStepDown(SSyncNode* pSyncNode, SyncTerm newTerm); void syncNodeBecomeFollower(SSyncNode* pSyncNode, const char* debugStr); +void syncNodeBecomeLearner(SSyncNode* pSyncNode, const char* debugStr); void syncNodeBecomeLeader(SSyncNode* pSyncNode, const char* debugStr); void syncNodeCandidate2Leader(SSyncNode* pSyncNode); void syncNodeFollower2Candidate(SSyncNode* pSyncNode); diff --git a/source/libs/sync/inc/syncMessage.h b/source/libs/sync/inc/syncMessage.h index c3566c7c82..f8c96d8be2 100644 --- a/source/libs/sync/inc/syncMessage.h +++ b/source/libs/sync/inc/syncMessage.h @@ -237,6 +237,7 @@ typedef struct SyncLeaderTransfer { typedef enum { SYNC_LOCAL_CMD_STEP_DOWN = 100, SYNC_LOCAL_CMD_FOLLOWER_CMT, + SYNC_LOCAL_CMD_LEARNER_CMT, } ESyncLocalCmd; typedef struct SyncLocalCmd { diff --git a/source/libs/sync/inc/syncPipeline.h b/source/libs/sync/inc/syncPipeline.h index d709e33cd4..02790732a2 100644 --- a/source/libs/sync/inc/syncPipeline.h +++ b/source/libs/sync/inc/syncPipeline.h @@ -56,6 +56,8 @@ typedef struct SSyncLogBuffer { int64_t size; TdThreadMutex mutex; TdThreadMutexAttr attr; + int64_t totalIndex; + bool isCatchup; } SSyncLogBuffer; // SSyncLogRepMgr diff --git a/source/libs/sync/src/syncAppendEntries.c b/source/libs/sync/src/syncAppendEntries.c index deae4b0b3f..6f065f56e8 100644 --- a/source/libs/sync/src/syncAppendEntries.c +++ b/source/libs/sync/src/syncAppendEntries.c @@ -126,8 +126,10 @@ int32_t syncNodeOnAppendEntries(SSyncNode* ths, const SRpcMsg* pRpcMsg) { pReply->term = pMsg->term; } - syncNodeStepDown(ths, pMsg->term); - resetElect = true; + if(ths->raftCfg.cfg.nodeInfo[ths->raftCfg.cfg.myIndex].nodeRole != TAOS_SYNC_ROLE_LEARNER){ + syncNodeStepDown(ths, pMsg->term); + resetElect = true; + } if (pMsg->dataLen < sizeof(SSyncRaftEntry)) { sError("vgId:%d, incomplete append entries received. prev index:%" PRId64 ", term:%" PRId64 ", datalen:%d", diff --git a/source/libs/sync/src/syncCommit.c b/source/libs/sync/src/syncCommit.c index 01f1f00c8b..47e4049c73 100644 --- a/source/libs/sync/src/syncCommit.c +++ b/source/libs/sync/src/syncCommit.c @@ -59,10 +59,12 @@ bool syncNodeAgreedUpon(SSyncNode* pNode, SyncIndex index) { SSyncIndexMgr* pMatches = pNode->pMatchIndex; ASSERT(pNode->replicaNum == pMatches->replicaNum); - for (int i = 0; i < pNode->replicaNum; i++) { - SyncIndex matchIndex = pMatches->index[i]; - if (matchIndex >= index) { - count++; + for (int i = 0; i < pNode->totalReplicaNum; i++) { + if(pNode->raftCfg.cfg.nodeInfo[i].nodeRole == TAOS_SYNC_ROLE_VOTER){ + SyncIndex matchIndex = pMatches->index[i]; + if (matchIndex >= index) { + count++; + } } } diff --git a/source/libs/sync/src/syncElection.c b/source/libs/sync/src/syncElection.c index 3699efbc59..86e28db90c 100644 --- a/source/libs/sync/src/syncElection.c +++ b/source/libs/sync/src/syncElection.c @@ -41,6 +41,8 @@ static int32_t syncNodeRequestVotePeers(SSyncNode* pNode) { int32_t ret = 0; for (int i = 0; i < pNode->peersNum; ++i) { + if(pNode->peersNodeInfo[i].nodeRole == TAOS_SYNC_ROLE_LEARNER) continue; + SRpcMsg rpcMsg = {0}; ret = syncBuildRequestVote(&rpcMsg, pNode->vgId); if (ret < 0) { diff --git a/source/libs/sync/src/syncIndexMgr.c b/source/libs/sync/src/syncIndexMgr.c index 7ecb9d7782..e3c3f63a4f 100644 --- a/source/libs/sync/src/syncIndexMgr.c +++ b/source/libs/sync/src/syncIndexMgr.c @@ -26,6 +26,7 @@ SSyncIndexMgr *syncIndexMgrCreate(SSyncNode *pNode) { pIndexMgr->replicas = &pNode->replicasId; pIndexMgr->replicaNum = pNode->replicaNum; + pIndexMgr->totalReplicaNum = pNode->totalReplicaNum; pIndexMgr->pNode = pNode; syncIndexMgrClear(pIndexMgr); @@ -35,6 +36,7 @@ SSyncIndexMgr *syncIndexMgrCreate(SSyncNode *pNode) { void syncIndexMgrUpdate(SSyncIndexMgr *pIndexMgr, SSyncNode *pNode) { pIndexMgr->replicas = &pNode->replicasId; pIndexMgr->replicaNum = pNode->replicaNum; + pIndexMgr->totalReplicaNum = pNode->totalReplicaNum; pIndexMgr->pNode = pNode; syncIndexMgrClear(pIndexMgr); } @@ -50,14 +52,14 @@ void syncIndexMgrClear(SSyncIndexMgr *pIndexMgr) { memset(pIndexMgr->privateTerm, 0, sizeof(pIndexMgr->privateTerm)); int64_t timeNow = taosGetTimestampMs(); - for (int i = 0; i < pIndexMgr->replicaNum; ++i) { + for (int i = 0; i < pIndexMgr->totalReplicaNum; ++i) { pIndexMgr->startTimeArr[i] = 0; pIndexMgr->recvTimeArr[i] = timeNow; } } void syncIndexMgrSetIndex(SSyncIndexMgr *pIndexMgr, const SRaftId *pRaftId, SyncIndex index) { - for (int i = 0; i < pIndexMgr->replicaNum; ++i) { + for (int i = 0; i < pIndexMgr->totalReplicaNum; ++i) { if (syncUtilSameId(&((*(pIndexMgr->replicas))[i]), pRaftId)) { (pIndexMgr->index)[i] = index; return; @@ -69,7 +71,7 @@ void syncIndexMgrSetIndex(SSyncIndexMgr *pIndexMgr, const SRaftId *pRaftId, Sync } SSyncLogReplMgr *syncNodeGetLogReplMgr(SSyncNode *pNode, SRaftId *pRaftId) { - for (int i = 0; i < pNode->replicaNum; i++) { + for (int i = 0; i < pNode->totalReplicaNum; i++) { if (syncUtilSameId(&pNode->replicasId[i], pRaftId)) { return pNode->logReplMgrs[i]; } @@ -80,7 +82,7 @@ SSyncLogReplMgr *syncNodeGetLogReplMgr(SSyncNode *pNode, SRaftId *pRaftId) { } SyncIndex syncIndexMgrGetIndex(SSyncIndexMgr *pIndexMgr, const SRaftId *pRaftId) { - for (int i = 0; i < pIndexMgr->replicaNum; ++i) { + for (int i = 0; i < pIndexMgr->totalReplicaNum; ++i) { if (syncUtilSameId(&((*(pIndexMgr->replicas))[i]), pRaftId)) { SyncIndex idx = (pIndexMgr->index)[i]; return idx; @@ -93,7 +95,7 @@ SyncIndex syncIndexMgrGetIndex(SSyncIndexMgr *pIndexMgr, const SRaftId *pRaftId) } void syncIndexMgrSetStartTime(SSyncIndexMgr *pIndexMgr, const SRaftId *pRaftId, int64_t startTime) { - for (int i = 0; i < pIndexMgr->replicaNum; ++i) { + for (int i = 0; i < pIndexMgr->totalReplicaNum; ++i) { if (syncUtilSameId(&((*(pIndexMgr->replicas))[i]), pRaftId)) { (pIndexMgr->startTimeArr)[i] = startTime; return; @@ -105,7 +107,7 @@ void syncIndexMgrSetStartTime(SSyncIndexMgr *pIndexMgr, const SRaftId *pRaftId, } int64_t syncIndexMgrGetStartTime(SSyncIndexMgr *pIndexMgr, const SRaftId *pRaftId) { - for (int i = 0; i < pIndexMgr->replicaNum; ++i) { + for (int i = 0; i < pIndexMgr->totalReplicaNum; ++i) { if (syncUtilSameId(&((*(pIndexMgr->replicas))[i]), pRaftId)) { int64_t startTime = (pIndexMgr->startTimeArr)[i]; return startTime; @@ -118,7 +120,7 @@ int64_t syncIndexMgrGetStartTime(SSyncIndexMgr *pIndexMgr, const SRaftId *pRaftI } void syncIndexMgrSetRecvTime(SSyncIndexMgr *pIndexMgr, const SRaftId *pRaftId, int64_t recvTime) { - for (int i = 0; i < pIndexMgr->replicaNum; ++i) { + for (int i = 0; i < pIndexMgr->totalReplicaNum; ++i) { if (syncUtilSameId(&((*(pIndexMgr->replicas))[i]), pRaftId)) { (pIndexMgr->recvTimeArr)[i] = recvTime; return; @@ -130,7 +132,7 @@ void syncIndexMgrSetRecvTime(SSyncIndexMgr *pIndexMgr, const SRaftId *pRaftId, i } int64_t syncIndexMgrGetRecvTime(SSyncIndexMgr *pIndexMgr, const SRaftId *pRaftId) { - for (int i = 0; i < pIndexMgr->replicaNum; ++i) { + for (int i = 0; i < pIndexMgr->totalReplicaNum; ++i) { if (syncUtilSameId(&((*(pIndexMgr->replicas))[i]), pRaftId)) { int64_t recvTime = (pIndexMgr->recvTimeArr)[i]; return recvTime; @@ -143,7 +145,7 @@ int64_t syncIndexMgrGetRecvTime(SSyncIndexMgr *pIndexMgr, const SRaftId *pRaftId } void syncIndexMgrSetTerm(SSyncIndexMgr *pIndexMgr, const SRaftId *pRaftId, SyncTerm term) { - for (int i = 0; i < pIndexMgr->replicaNum; ++i) { + for (int i = 0; i < pIndexMgr->totalReplicaNum; ++i) { if (syncUtilSameId(&((*(pIndexMgr->replicas))[i]), pRaftId)) { (pIndexMgr->privateTerm)[i] = term; return; @@ -155,7 +157,7 @@ void syncIndexMgrSetTerm(SSyncIndexMgr *pIndexMgr, const SRaftId *pRaftId, SyncT } SyncTerm syncIndexMgrGetTerm(SSyncIndexMgr *pIndexMgr, const SRaftId *pRaftId) { - for (int i = 0; i < pIndexMgr->replicaNum; ++i) { + for (int i = 0; i < pIndexMgr->totalReplicaNum; ++i) { if (syncUtilSameId(&((*(pIndexMgr->replicas))[i]), pRaftId)) { SyncTerm term = (pIndexMgr->privateTerm)[i]; return term; diff --git a/source/libs/sync/src/syncMain.c b/source/libs/sync/src/syncMain.c index 966b3ed093..499df4a98b 100644 --- a/source/libs/sync/src/syncMain.c +++ b/source/libs/sync/src/syncMain.c @@ -141,6 +141,13 @@ int32_t syncReconfig(int64_t rid, SSyncCfg* pNewCfg) { SSyncNode* pSyncNode = syncNodeAcquire(rid); if (pSyncNode == NULL) return -1; + if(pSyncNode->raftCfg.lastConfigIndex >= pNewCfg->lastIndex){ + syncNodeRelease(pSyncNode); + sInfo("vgId:%d, no need Reconfig, current index:%" PRId64 ", new index:%" PRId64, pSyncNode->vgId, + pSyncNode->raftCfg.lastConfigIndex, pNewCfg->lastIndex); + return 0; + } + if (!syncNodeCheckNewConfig(pSyncNode, pNewCfg)) { syncNodeRelease(pSyncNode); terrno = TSDB_CODE_SYN_NEW_CONFIG_ERROR; @@ -149,12 +156,12 @@ int32_t syncReconfig(int64_t rid, SSyncCfg* pNewCfg) { } syncNodeUpdateNewConfigIndex(pSyncNode, pNewCfg); - syncNodeDoConfigChange(pSyncNode, pNewCfg, SYNC_INDEX_INVALID); + syncNodeDoConfigChange(pSyncNode, pNewCfg, pNewCfg->lastIndex); if (pSyncNode->state == TAOS_SYNC_STATE_LEADER) { syncNodeStopHeartbeatTimer(pSyncNode); - for (int32_t i = 0; i < TSDB_MAX_REPLICA; ++i) { + for (int32_t i = 0; i < TSDB_MAX_REPLICA + TSDB_MAX_LEARNER_REPLICA; ++i) { syncHbTimerInit(pSyncNode, &pSyncNode->peerHeartbeatTimerArr[i], pSyncNode->replicasId[i]); } @@ -315,8 +322,9 @@ int32_t syncBeginSnapshot(int64_t rid, int64_t lastApplyIndex) { } } - if (pSyncNode->replicaNum > 1) { - if (pSyncNode->state != TAOS_SYNC_STATE_LEADER && pSyncNode->state != TAOS_SYNC_STATE_FOLLOWER) { + if (pSyncNode->totalReplicaNum > 1) { + if (pSyncNode->state != TAOS_SYNC_STATE_LEADER && pSyncNode->state != TAOS_SYNC_STATE_FOLLOWER + && pSyncNode->state != TAOS_SYNC_STATE_LEARNER) { sNTrace(pSyncNode, "new-snapshot-index:%" PRId64 " candidate or unknown state, do not delete wal", lastApplyIndex); syncNodeRelease(pSyncNode); @@ -455,8 +463,7 @@ bool syncSnapshotRecving(int64_t rid) { int32_t syncNodeLeaderTransfer(SSyncNode* pSyncNode) { if (pSyncNode->peersNum == 0) { sDebug("vgId:%d, only one replica, cannot leader transfer", pSyncNode->vgId); - terrno = TSDB_CODE_SYN_ONE_REPLICA; - return -1; + return 0; } int32_t ret = 0; @@ -478,7 +485,6 @@ int32_t syncNodeLeaderTransfer(SSyncNode* pSyncNode) { int32_t syncNodeLeaderTransferTo(SSyncNode* pSyncNode, SNodeInfo newLeader) { if (pSyncNode->replicaNum == 1) { sDebug("vgId:%d, only one replica, cannot leader transfer", pSyncNode->vgId); - terrno = TSDB_CODE_SYN_ONE_REPLICA; return -1; } @@ -537,7 +543,8 @@ void syncGetRetryEpSet(int64_t rid, SEpSet* pEpSet) { SSyncNode* pSyncNode = syncNodeAcquire(rid); if (pSyncNode == NULL) return; - for (int32_t i = 0; i < pSyncNode->raftCfg.cfg.replicaNum; ++i) { + for (int32_t i = 0; i < pSyncNode->raftCfg.cfg.totalReplicaNum; ++i) { + if(pSyncNode->raftCfg.cfg.nodeInfo[i].nodeRole == TAOS_SYNC_ROLE_LEARNER) continue; SEp* pEp = &pEpSet->eps[i]; tstrncpy(pEp->fqdn, pSyncNode->raftCfg.cfg.nodeInfo[i].nodeFqdn, TSDB_FQDN_LEN); pEp->port = (pSyncNode->raftCfg.cfg.nodeInfo)[i].nodePort; @@ -564,6 +571,46 @@ int32_t syncPropose(int64_t rid, SRpcMsg* pMsg, bool isWeak, int64_t* seq) { return ret; } +int32_t syncIsCatchUp(int64_t rid) { + SSyncNode* pSyncNode = syncNodeAcquire(rid); + if (pSyncNode == NULL) { + sError("sync Node Acquire error since %d", errno); + return -1; + } + + int32_t isCatchUp = 0; + if(pSyncNode->pLogBuf->totalIndex < 0 || pSyncNode->pLogBuf->commitIndex < 0 || + pSyncNode->pLogBuf->totalIndex < pSyncNode->pLogBuf->commitIndex || + pSyncNode->pLogBuf->totalIndex - pSyncNode->pLogBuf->commitIndex > SYNC_LEARNER_CATCHUP){ + sInfo("vgId:%d, Not catch up, wait one second, totalIndex:%" PRId64 " commitIndex:%" PRId64 " matchIndex:%" PRId64, + pSyncNode->vgId, pSyncNode->pLogBuf->totalIndex, pSyncNode->pLogBuf->commitIndex, + pSyncNode->pLogBuf->matchIndex); + isCatchUp = 0; + } + else{ + sInfo("vgId:%d, Catch up, totalIndex:%" PRId64 " commitIndex:%" PRId64 " matchIndex:%" PRId64, + pSyncNode->vgId, pSyncNode->pLogBuf->totalIndex, pSyncNode->pLogBuf->commitIndex, + pSyncNode->pLogBuf->matchIndex); + isCatchUp = 1; + } + + syncNodeRelease(pSyncNode); + return isCatchUp; +} + +ESyncRole syncGetRole(int64_t rid) { + SSyncNode* pSyncNode = syncNodeAcquire(rid); + if (pSyncNode == NULL) { + sError("sync Node Acquire error since %d", errno); + return -1; + } + + ESyncRole role = pSyncNode->raftCfg.cfg.nodeInfo[pSyncNode->raftCfg.cfg.myIndex].nodeRole; + + syncNodeRelease(pSyncNode); + return role; +} + int32_t syncNodePropose(SSyncNode* pSyncNode, SRpcMsg* pMsg, bool isWeak, int64_t* seq) { if (pSyncNode->state != TAOS_SYNC_STATE_LEADER) { terrno = TSDB_CODE_SYN_NOT_LEADER; @@ -655,6 +702,8 @@ static int32_t syncHbTimerStart(SSyncNode* pSyncNode, SSyncTimer* pSyncTimer) { pData->logicClock = pSyncTimer->logicClock; pData->execTime = tsNow + pSyncTimer->timerMS; + sTrace("vgId:%d, start hb timer, rid:%" PRId64 " addr:%" PRId64, pSyncNode->vgId, pData->rid, pData->destId.addr); + taosTmrReset(pSyncTimer->timerCb, pSyncTimer->timerMS / HEARTBEAT_TICK_NUM, (void*)(pData->rid), syncEnv()->pTimerManager, &pSyncTimer->pTimer); } else { @@ -719,7 +768,7 @@ SSyncNode* syncNodeOpen(SSyncInfo* pSyncInfo) { sInfo("vgId:%d, create a new raft config file", pSyncNode->vgId); pSyncNode->raftCfg.isStandBy = pSyncInfo->isStandBy; pSyncNode->raftCfg.snapshotStrategy = pSyncInfo->snapshotStrategy; - pSyncNode->raftCfg.lastConfigIndex = SYNC_INDEX_INVALID; + pSyncNode->raftCfg.lastConfigIndex = pSyncInfo->syncCfg.lastIndex; pSyncNode->raftCfg.batchSize = pSyncInfo->batchSize; pSyncNode->raftCfg.cfg = pSyncInfo->syncCfg; pSyncNode->raftCfg.configIndexCount = 1; @@ -736,7 +785,7 @@ SSyncNode* syncNodeOpen(SSyncInfo* pSyncInfo) { goto _error; } - if (pSyncInfo->syncCfg.replicaNum > 0 && syncIsConfigChanged(&pSyncNode->raftCfg.cfg, &pSyncInfo->syncCfg)) { + if (pSyncInfo->syncCfg.totalReplicaNum > 0 && syncIsConfigChanged(&pSyncNode->raftCfg.cfg, &pSyncInfo->syncCfg)) { sInfo("vgId:%d, use sync config from input options and write to cfg file", pSyncNode->vgId); pSyncNode->raftCfg.cfg = pSyncInfo->syncCfg; if (syncWriteCfgFile(pSyncNode) != 0) { @@ -753,8 +802,9 @@ SSyncNode* syncNodeOpen(SSyncInfo* pSyncInfo) { pSyncNode->vgId = pSyncInfo->vgId; SSyncCfg* pCfg = &pSyncNode->raftCfg.cfg; bool updated = false; - sInfo("vgId:%d, start to open sync node, replica:%d selfIndex:%d", pSyncNode->vgId, pCfg->replicaNum, pCfg->myIndex); - for (int32_t i = 0; i < pCfg->replicaNum; ++i) { + sInfo("vgId:%d, start to open sync node, totalReplicaNum:%d replicaNum:%d selfIndex:%d", + pSyncNode->vgId, pCfg->totalReplicaNum, pCfg->replicaNum, pCfg->myIndex); + for (int32_t i = 0; i < pCfg->totalReplicaNum; ++i) { SNodeInfo* pNode = &pCfg->nodeInfo[i]; if (tmsgUpdateDnodeInfo(&pNode->nodeId, &pNode->clusterId, pNode->nodeFqdn, &pNode->nodePort)) { updated = true; @@ -792,9 +842,9 @@ SSyncNode* syncNodeOpen(SSyncInfo* pSyncInfo) { } // init peersNum, peers, peersId - pSyncNode->peersNum = pSyncNode->raftCfg.cfg.replicaNum - 1; + pSyncNode->peersNum = pSyncNode->raftCfg.cfg.totalReplicaNum - 1; int32_t j = 0; - for (int32_t i = 0; i < pSyncNode->raftCfg.cfg.replicaNum; ++i) { + for (int32_t i = 0; i < pSyncNode->raftCfg.cfg.totalReplicaNum; ++i) { if (i != pSyncNode->raftCfg.cfg.myIndex) { pSyncNode->peersNodeInfo[j] = pSyncNode->raftCfg.cfg.nodeInfo[i]; syncUtilNodeInfo2EpSet(&pSyncNode->peersNodeInfo[j], &pSyncNode->peersEpset[j]); @@ -810,7 +860,8 @@ SSyncNode* syncNodeOpen(SSyncInfo* pSyncInfo) { // init replicaNum, replicasId pSyncNode->replicaNum = pSyncNode->raftCfg.cfg.replicaNum; - for (int32_t i = 0; i < pSyncNode->raftCfg.cfg.replicaNum; ++i) { + pSyncNode->totalReplicaNum = pSyncNode->raftCfg.cfg.totalReplicaNum; + for (int32_t i = 0; i < pSyncNode->raftCfg.cfg.totalReplicaNum; ++i) { if (!syncUtilNodeInfo2RaftId(&pSyncNode->raftCfg.cfg.nodeInfo[i], pSyncNode->vgId, &pSyncNode->replicasId[i])) { sError("vgId:%d, failed to determine raft member id, replica:%d", pSyncNode->vgId, i); goto _error; @@ -934,7 +985,7 @@ SSyncNode* syncNodeOpen(SSyncInfo* pSyncInfo) { pSyncNode->heartbeatTimerCounter = 0; // init peer heartbeat timer - for (int32_t i = 0; i < TSDB_MAX_REPLICA; ++i) { + for (int32_t i = 0; i < TSDB_MAX_REPLICA + TSDB_MAX_LEARNER_REPLICA; ++i) { syncHbTimerInit(pSyncNode, &(pSyncNode->peerHeartbeatTimerArr[i]), (pSyncNode->replicasId)[i]); } @@ -949,7 +1000,7 @@ SSyncNode* syncNodeOpen(SSyncInfo* pSyncInfo) { pSyncNode->restoreFinish = false; // snapshot senders - for (int32_t i = 0; i < TSDB_MAX_REPLICA; ++i) { + for (int32_t i = 0; i < TSDB_MAX_REPLICA + TSDB_MAX_LEARNER_REPLICA; ++i) { SSyncSnapshotSender* pSender = snapshotSenderCreate(pSyncNode, i); if (pSender == NULL) return NULL; @@ -1059,14 +1110,19 @@ int32_t syncNodeRestore(SSyncNode* pSyncNode) { int32_t syncNodeStart(SSyncNode* pSyncNode) { // start raft - if (pSyncNode->replicaNum == 1) { - raftStoreNextTerm(pSyncNode); - syncNodeBecomeLeader(pSyncNode, "one replica start"); + if(pSyncNode->raftCfg.cfg.nodeInfo[pSyncNode->raftCfg.cfg.myIndex].nodeRole == TAOS_SYNC_ROLE_LEARNER){ + syncNodeBecomeLearner(pSyncNode, "first start"); + } + else{ + if (pSyncNode->replicaNum == 1) { + raftStoreNextTerm(pSyncNode); + syncNodeBecomeLeader(pSyncNode, "one replica start"); - // Raft 3.6.2 Committing entries from previous terms - syncNodeAppendNoop(pSyncNode); - } else { - syncNodeBecomeFollower(pSyncNode, "first start"); + // Raft 3.6.2 Committing entries from previous terms + syncNodeAppendNoop(pSyncNode); + } else { + syncNodeBecomeFollower(pSyncNode, "first start"); + } } int32_t ret = 0; @@ -1157,7 +1213,7 @@ void syncNodeClose(SSyncNode* pSyncNode) { syncLogBufferDestroy(pSyncNode->pLogBuf); pSyncNode->pLogBuf = NULL; - for (int32_t i = 0; i < TSDB_MAX_REPLICA; ++i) { + for (int32_t i = 0; i < TSDB_MAX_REPLICA + TSDB_MAX_LEARNER_REPLICA; ++i) { if (pSyncNode->senders[i] != NULL) { sDebug("vgId:%d, snapshot sender destroy while close, data:%p", pSyncNode->vgId, pSyncNode->senders[i]); @@ -1350,7 +1406,7 @@ inline bool syncNodeInConfig(SSyncNode* pNode, const SSyncCfg* pCfg) { bool b1 = false; bool b2 = false; - for (int32_t i = 0; i < pCfg->replicaNum; ++i) { + for (int32_t i = 0; i < pCfg->totalReplicaNum; ++i) { if (strcmp(pCfg->nodeInfo[i].nodeFqdn, pNode->myNodeInfo.nodeFqdn) == 0 && pCfg->nodeInfo[i].nodePort == pNode->myNodeInfo.nodePort) { b1 = true; @@ -1358,7 +1414,7 @@ inline bool syncNodeInConfig(SSyncNode* pNode, const SSyncCfg* pCfg) { } } - for (int32_t i = 0; i < pCfg->replicaNum; ++i) { + for (int32_t i = 0; i < pCfg->totalReplicaNum; ++i) { SRaftId raftId = { .addr = SYNC_ADDR(&pCfg->nodeInfo[i]), .vgId = pNode->vgId, @@ -1375,13 +1431,14 @@ inline bool syncNodeInConfig(SSyncNode* pNode, const SSyncCfg* pCfg) { } static bool syncIsConfigChanged(const SSyncCfg* pOldCfg, const SSyncCfg* pNewCfg) { - if (pOldCfg->replicaNum != pNewCfg->replicaNum) return true; + if (pOldCfg->totalReplicaNum != pNewCfg->totalReplicaNum) return true; if (pOldCfg->myIndex != pNewCfg->myIndex) return true; - for (int32_t i = 0; i < pOldCfg->replicaNum; ++i) { + for (int32_t i = 0; i < pOldCfg->totalReplicaNum; ++i) { const SNodeInfo* pOldInfo = &pOldCfg->nodeInfo[i]; const SNodeInfo* pNewInfo = &pNewCfg->nodeInfo[i]; if (strcmp(pOldInfo->nodeFqdn, pNewInfo->nodeFqdn) != 0) return true; if (pOldInfo->nodePort != pNewInfo->nodePort) return true; + if(pOldInfo->nodeRole != pNewInfo->nodeRole) return true; } return false; @@ -1418,8 +1475,10 @@ void syncNodeDoConfigChange(SSyncNode* pSyncNode, SSyncCfg* pNewConfig, SyncInde } // log begin config change - sNInfo(pSyncNode, "begin do config change, from %d to %d, replicas:%d", pSyncNode->vgId, oldConfig.replicaNum, - pNewConfig->replicaNum); + sNInfo(pSyncNode, "begin do config change, from %d to %d, from %" PRId64 " to %" PRId64 ", replicas:%d", + pSyncNode->vgId, + oldConfig.totalReplicaNum, pNewConfig->totalReplicaNum, + oldConfig.lastIndex, pNewConfig->lastIndex); if (IamInNew) { pSyncNode->raftCfg.isStandBy = 0; // change isStandBy to normal @@ -1436,11 +1495,10 @@ void syncNodeDoConfigChange(SSyncNode* pSyncNode, SSyncCfg* pNewConfig, SyncInde int32_t ret = 0; // save snapshot senders - int32_t oldReplicaNum = pSyncNode->replicaNum; - SRaftId oldReplicasId[TSDB_MAX_REPLICA]; + SRaftId oldReplicasId[TSDB_MAX_REPLICA + TSDB_MAX_LEARNER_REPLICA]; memcpy(oldReplicasId, pSyncNode->replicasId, sizeof(oldReplicasId)); - SSyncSnapshotSender* oldSenders[TSDB_MAX_REPLICA]; - for (int32_t i = 0; i < TSDB_MAX_REPLICA; ++i) { + SSyncSnapshotSender* oldSenders[TSDB_MAX_REPLICA + TSDB_MAX_LEARNER_REPLICA]; + for (int32_t i = 0; i < TSDB_MAX_REPLICA + TSDB_MAX_LEARNER_REPLICA; ++i) { oldSenders[i] = pSyncNode->senders[i]; sSTrace(oldSenders[i], "snapshot sender save old"); } @@ -1450,9 +1508,9 @@ void syncNodeDoConfigChange(SSyncNode* pSyncNode, SSyncCfg* pNewConfig, SyncInde syncUtilNodeInfo2RaftId(&pSyncNode->myNodeInfo, pSyncNode->vgId, &pSyncNode->myRaftId); // init peersNum, peers, peersId - pSyncNode->peersNum = pSyncNode->raftCfg.cfg.replicaNum - 1; + pSyncNode->peersNum = pSyncNode->raftCfg.cfg.totalReplicaNum - 1; int32_t j = 0; - for (int32_t i = 0; i < pSyncNode->raftCfg.cfg.replicaNum; ++i) { + for (int32_t i = 0; i < pSyncNode->raftCfg.cfg.totalReplicaNum; ++i) { if (i != pSyncNode->raftCfg.cfg.myIndex) { pSyncNode->peersNodeInfo[j] = pSyncNode->raftCfg.cfg.nodeInfo[i]; syncUtilNodeInfo2EpSet(&pSyncNode->peersNodeInfo[j], &pSyncNode->peersEpset[j]); @@ -1465,7 +1523,8 @@ void syncNodeDoConfigChange(SSyncNode* pSyncNode, SSyncCfg* pNewConfig, SyncInde // init replicaNum, replicasId pSyncNode->replicaNum = pSyncNode->raftCfg.cfg.replicaNum; - for (int32_t i = 0; i < pSyncNode->raftCfg.cfg.replicaNum; ++i) { + pSyncNode->totalReplicaNum = pSyncNode->raftCfg.cfg.totalReplicaNum; + for (int32_t i = 0; i < pSyncNode->raftCfg.cfg.totalReplicaNum; ++i) { syncUtilNodeInfo2RaftId(&pSyncNode->raftCfg.cfg.nodeInfo[i], pSyncNode->vgId, &pSyncNode->replicasId[i]); } @@ -1480,15 +1539,15 @@ void syncNodeDoConfigChange(SSyncNode* pSyncNode, SSyncCfg* pNewConfig, SyncInde // reset snapshot senders // clear new - for (int32_t i = 0; i < TSDB_MAX_REPLICA; ++i) { + for (int32_t i = 0; i < TSDB_MAX_REPLICA + TSDB_MAX_LEARNER_REPLICA; ++i) { pSyncNode->senders[i] = NULL; } // reset new - for (int32_t i = 0; i < pSyncNode->replicaNum; ++i) { + for (int32_t i = 0; i < pSyncNode->totalReplicaNum; ++i) { // reset sender bool reset = false; - for (int32_t j = 0; j < TSDB_MAX_REPLICA; ++j) { + for (int32_t j = 0; j < TSDB_MAX_REPLICA + TSDB_MAX_LEARNER_REPLICA; ++j) { if (syncUtilSameId(&(pSyncNode->replicasId)[i], &oldReplicasId[j]) && oldSenders[j] != NULL) { sNTrace(pSyncNode, "snapshot sender reset for:%" PRId64 ", newIndex:%d, dnode:%d, %p", (pSyncNode->replicasId)[i].addr, i, DID(&pSyncNode->replicasId[i]), oldSenders[j]); @@ -1510,7 +1569,7 @@ void syncNodeDoConfigChange(SSyncNode* pSyncNode, SSyncCfg* pNewConfig, SyncInde } // create new - for (int32_t i = 0; i < TSDB_MAX_REPLICA; ++i) { + for (int32_t i = 0; i < TSDB_MAX_REPLICA + TSDB_MAX_LEARNER_REPLICA; ++i) { if (pSyncNode->senders[i] == NULL) { pSyncNode->senders[i] = snapshotSenderCreate(pSyncNode, i); if (pSyncNode->senders[i] == NULL) { @@ -1525,7 +1584,7 @@ void syncNodeDoConfigChange(SSyncNode* pSyncNode, SSyncCfg* pNewConfig, SyncInde } // free old - for (int32_t i = 0; i < TSDB_MAX_REPLICA; ++i) { + for (int32_t i = 0; i < TSDB_MAX_REPLICA + TSDB_MAX_LEARNER_REPLICA; ++i) { if (oldSenders[i] != NULL) { sSDebug(oldSenders[i], "snapshot sender destroy old, data:%p replica-index:%d", oldSenders[i], i); snapshotSenderDestroy(oldSenders[i]); @@ -1550,12 +1609,12 @@ void syncNodeDoConfigChange(SSyncNode* pSyncNode, SSyncCfg* pNewConfig, SyncInde } else { // persist cfg syncWriteCfgFile(pSyncNode); - sNInfo(pSyncNode, "do not config change from %d to %d", oldConfig.replicaNum, pNewConfig->replicaNum); + sNInfo(pSyncNode, "do not config change from %d to %d", oldConfig.totalReplicaNum, pNewConfig->totalReplicaNum); } _END: // log end config change - sNInfo(pSyncNode, "end do config change, from %d to %d", oldConfig.replicaNum, pNewConfig->replicaNum); + sNInfo(pSyncNode, "end do config change, from %d to %d", oldConfig.totalReplicaNum, pNewConfig->totalReplicaNum); } // raft state change -------------- @@ -1635,6 +1694,27 @@ void syncNodeBecomeFollower(SSyncNode* pSyncNode, const char* debugStr) { syncNodeResetElectTimer(pSyncNode); } +void syncNodeBecomeLearner(SSyncNode* pSyncNode, const char* debugStr) { + pSyncNode->hbSlowNum = 0; + + // state change + pSyncNode->state = TAOS_SYNC_STATE_LEARNER; + + // trace log + sNTrace(pSyncNode, "become learner %s", debugStr); + + // call back + if (pSyncNode->pFsm != NULL && pSyncNode->pFsm->FpBecomeLearnerCb != NULL) { + pSyncNode->pFsm->FpBecomeLearnerCb(pSyncNode->pFsm); + } + + // min match index + pSyncNode->minMatchIndex = SYNC_INDEX_INVALID; + + // reset log buffer + syncLogBufferReset(pSyncNode->pLogBuf, pSyncNode); +} + // TLA+ Spec // \* Candidate i transitions to leader. // BecomeLeader(i) == @@ -1752,7 +1832,7 @@ void syncNodeCandidate2Leader(SSyncNode* pSyncNode) { bool syncNodeIsMnode(SSyncNode* pSyncNode) { return (pSyncNode->vgId == 1); } int32_t syncNodePeerStateInit(SSyncNode* pSyncNode) { - for (int32_t i = 0; i < TSDB_MAX_REPLICA; ++i) { + for (int32_t i = 0; i < TSDB_MAX_REPLICA + TSDB_MAX_LEARNER_REPLICA; ++i) { pSyncNode->peerStates[i].lastSendIndex = SYNC_INDEX_INVALID; pSyncNode->peerStates[i].lastSendTime = 0; } @@ -2039,7 +2119,7 @@ static void syncNodeEqHeartbeatTimer(void* param, void* tmrId) { if (!syncIsInit()) return; SSyncNode* pNode = param; - if (pNode->replicaNum > 1) { + if (pNode->totalReplicaNum > 1) { if (atomic_load_64(&pNode->heartbeatTimerLogicClockUser) <= atomic_load_64(&pNode->heartbeatTimerLogicClock)) { SRpcMsg rpcMsg = {0}; int32_t code = syncBuildTimeout(&rpcMsg, SYNC_TIMEOUT_HEARTBEAT, atomic_load_64(&pNode->heartbeatTimerLogicClock), @@ -2101,9 +2181,9 @@ static void syncNodeEqPeerHeartbeatTimer(void* param, void* tmrId) { return; } - // sTrace("vgId:%d, eq peer hb timer", pSyncNode->vgId); + sTrace("vgId:%d, eq peer hb timer, rid:%" PRId64 " addr:%" PRId64, pSyncNode->vgId, hbDataRid, pData->destId.addr); - if (pSyncNode->replicaNum > 1) { + if (pSyncNode->totalReplicaNum > 1) { int64_t timerLogicClock = atomic_load_64(&pSyncTimer->logicClock); int64_t msgLogicClock = atomic_load_64(&pData->logicClock); @@ -2130,6 +2210,7 @@ static void syncNodeEqPeerHeartbeatTimer(void* param, void* tmrId) { pSyncTimer->timeStamp = tsNow; // send msg + sTrace("vgId:%d, send heartbeat to dnode:%d", pSyncNode->vgId, DID(&(pSyncMsg->destId))); syncLogSendHeartbeat(pSyncNode, pSyncMsg, false, timerElapsed, pData->execTime); syncNodeSendHeartbeat(pSyncNode, &pSyncMsg->destId, &rpcMsg); } else { @@ -2212,7 +2293,7 @@ int32_t syncNodeAppend(SSyncNode* ths, SSyncRaftEntry* pEntry) { } bool syncNodeHeartbeatReplyTimeout(SSyncNode* pSyncNode) { - if (pSyncNode->replicaNum == 1) { + if (pSyncNode->totalReplicaNum == 1) { return false; } @@ -2237,7 +2318,7 @@ bool syncNodeHeartbeatReplyTimeout(SSyncNode* pSyncNode) { bool syncNodeSnapshotSending(SSyncNode* pSyncNode) { if (pSyncNode == NULL) return false; bool b = false; - for (int32_t i = 0; i < pSyncNode->replicaNum; ++i) { + for (int32_t i = 0; i < pSyncNode->totalReplicaNum; ++i) { if (pSyncNode->senders[i] != NULL && pSyncNode->senders[i]->start) { b = true; break; @@ -2328,35 +2409,40 @@ int32_t syncNodeOnHeartbeat(SSyncNode* ths, const SRpcMsg* pRpcMsg) { pMsgReply->startTime = ths->startTime; pMsgReply->timeStamp = tsMs; + sTrace( + "vgId:%d, heartbeat msg from dnode:%d, cluster:%d, Msgterm:%" PRId64 " currentTerm:%" PRId64, + ths->vgId, DID(&(pMsg->srcId)), CID(&(pMsg->srcId)), pMsg->term, currentTerm); + if (pMsg->term == currentTerm && ths->state != TAOS_SYNC_STATE_LEADER) { syncIndexMgrSetRecvTime(ths->pNextIndex, &(pMsg->srcId), tsMs); resetElect = true; ths->minMatchIndex = pMsg->minMatchIndex; - if (ths->state == TAOS_SYNC_STATE_FOLLOWER) { + if (ths->state == TAOS_SYNC_STATE_FOLLOWER || ths->state == TAOS_SYNC_STATE_LEARNER) { SRpcMsg rpcMsgLocalCmd = {0}; (void)syncBuildLocalCmd(&rpcMsgLocalCmd, ths->vgId); SyncLocalCmd* pSyncMsg = rpcMsgLocalCmd.pCont; - pSyncMsg->cmd = SYNC_LOCAL_CMD_FOLLOWER_CMT; + pSyncMsg->cmd = + (ths->state == TAOS_SYNC_STATE_LEARNER) ? SYNC_LOCAL_CMD_LEARNER_CMT : SYNC_LOCAL_CMD_FOLLOWER_CMT; pSyncMsg->commitIndex = pMsg->commitIndex; pSyncMsg->currentTerm = pMsg->term; - SyncIndex fcIndex = pSyncMsg->commitIndex; if (ths->syncEqMsg != NULL && ths->msgcb != NULL) { int32_t code = ths->syncEqMsg(ths->msgcb, &rpcMsgLocalCmd); if (code != 0) { - sError("vgId:%d, sync enqueue fc-commit msg error, code:%d", ths->vgId, code); + sError("vgId:%d, failed to enqueue commit msg from heartbeat since %s, code:%d", ths->vgId, terrstr(), code); rpcFreeCont(rpcMsgLocalCmd.pCont); } else { - sTrace("vgId:%d, sync enqueue fc-commit msg, fc-index:%" PRId64, ths->vgId, fcIndex); + sTrace("vgId:%d, enqueue commit msg from heartbeat, commit-index:%" PRId64 ", term:%" PRId64, ths->vgId, + pMsg->commitIndex, pMsg->term); } } } } - if (pMsg->term >= currentTerm && ths->state != TAOS_SYNC_STATE_FOLLOWER) { + if (pMsg->term >= currentTerm && ths->state == TAOS_SYNC_STATE_LEADER) { SRpcMsg rpcMsgLocalCmd = {0}; (void)syncBuildLocalCmd(&rpcMsgLocalCmd, ths->vgId); @@ -2371,7 +2457,7 @@ int32_t syncNodeOnHeartbeat(SSyncNode* ths, const SRpcMsg* pRpcMsg) { sError("vgId:%d, sync enqueue step-down msg error, code:%d", ths->vgId, code); rpcFreeCont(rpcMsgLocalCmd.pCont); } else { - sTrace("vgId:%d, sync enqueue step-down msg, new-term:%" PRId64, ths->vgId, pSyncMsg->currentTerm); + sTrace("vgId:%d, sync enqueue step-down msg, new-term:%" PRId64, ths->vgId, pMsg->term); } } } @@ -2426,7 +2512,7 @@ int32_t syncNodeOnLocalCmd(SSyncNode* ths, const SRpcMsg* pRpcMsg) { if (pMsg->cmd == SYNC_LOCAL_CMD_STEP_DOWN) { syncNodeStepDown(ths, pMsg->currentTerm); - } else if (pMsg->cmd == SYNC_LOCAL_CMD_FOLLOWER_CMT) { + } else if (pMsg->cmd == SYNC_LOCAL_CMD_FOLLOWER_CMT || pMsg->cmd == SYNC_LOCAL_CMD_LEARNER_CMT) { if (syncLogBufferIsEmpty(ths->pLogBuf)) { sError("vgId:%d, sync log buffer is empty.", ths->vgId); return 0; @@ -2502,13 +2588,15 @@ const char* syncStr(ESyncState state) { return "error"; case TAOS_SYNC_STATE_OFFLINE: return "offline"; + case TAOS_SYNC_STATE_LEARNER: + return "learner"; default: return "unknown"; } } int32_t syncNodeUpdateNewConfigIndex(SSyncNode* ths, SSyncCfg* pNewCfg) { - for (int32_t i = 0; i < pNewCfg->replicaNum; ++i) { + for (int32_t i = 0; i < pNewCfg->totalReplicaNum; ++i) { SRaftId raftId = { .addr = SYNC_ADDR(&pNewCfg->nodeInfo[i]), .vgId = ths->vgId, @@ -2528,7 +2616,7 @@ bool syncNodeIsOptimizedOneReplica(SSyncNode* ths, SRpcMsg* pMsg) { } bool syncNodeInRaftGroup(SSyncNode* ths, SRaftId* pRaftId) { - for (int32_t i = 0; i < ths->replicaNum; ++i) { + for (int32_t i = 0; i < ths->totalReplicaNum; ++i) { if (syncUtilSameId(&((ths->replicasId)[i]), pRaftId)) { return true; } @@ -2538,7 +2626,7 @@ bool syncNodeInRaftGroup(SSyncNode* ths, SRaftId* pRaftId) { SSyncSnapshotSender* syncNodeGetSnapshotSender(SSyncNode* ths, SRaftId* pDestId) { SSyncSnapshotSender* pSender = NULL; - for (int32_t i = 0; i < ths->replicaNum; ++i) { + for (int32_t i = 0; i < ths->totalReplicaNum; ++i) { if (syncUtilSameId(pDestId, &((ths->replicasId)[i]))) { pSender = (ths->senders)[i]; } @@ -2548,7 +2636,7 @@ SSyncSnapshotSender* syncNodeGetSnapshotSender(SSyncNode* ths, SRaftId* pDestId) SSyncTimer* syncNodeGetHbTimer(SSyncNode* ths, SRaftId* pDestId) { SSyncTimer* pTimer = NULL; - for (int32_t i = 0; i < ths->replicaNum; ++i) { + for (int32_t i = 0; i < ths->totalReplicaNum; ++i) { if (syncUtilSameId(pDestId, &((ths->replicasId)[i]))) { pTimer = &((ths->peerHeartbeatTimerArr)[i]); } @@ -2558,7 +2646,7 @@ SSyncTimer* syncNodeGetHbTimer(SSyncNode* ths, SRaftId* pDestId) { SPeerState* syncNodeGetPeerState(SSyncNode* ths, const SRaftId* pDestId) { SPeerState* pState = NULL; - for (int32_t i = 0; i < ths->replicaNum; ++i) { + for (int32_t i = 0; i < ths->totalReplicaNum; ++i) { if (syncUtilSameId(pDestId, &((ths->replicasId)[i]))) { pState = &((ths->peerStates)[i]); } diff --git a/source/libs/sync/src/syncPipeline.c b/source/libs/sync/src/syncPipeline.c index 5e6058ef03..8bb72de518 100644 --- a/source/libs/sync/src/syncPipeline.c +++ b/source/libs/sync/src/syncPipeline.c @@ -54,7 +54,7 @@ int32_t syncLogBufferAppend(SSyncLogBuffer* pBuf, SSyncNode* pNode, SSyncRaftEnt } if (pNode->restoreFinish && index - pBuf->commitIndex >= TSDB_SYNC_NEGOTIATION_WIN) { - terrno = TSDB_CODE_SYN_NEGO_WIN_EXCEEDED; + terrno = TSDB_CODE_SYN_NEGOTIATION_WIN_FULL; sError("vgId:%d, failed to append since %s, index:%" PRId64 ", commit-index:%" PRId64, pNode->vgId, terrstr(), index, pBuf->commitIndex); goto _err; @@ -250,6 +250,8 @@ int32_t syncLogBufferInitWithoutLock(SSyncLogBuffer* pBuf, SSyncNode* pNode) { // update startIndex pBuf->startIndex = takeDummy ? index : index + 1; + pBuf->isCatchup = false; + sInfo("vgId:%d, init sync log buffer. buffer: [%" PRId64 " %" PRId64 " %" PRId64 ", %" PRId64 ")", pNode->vgId, pBuf->startIndex, pBuf->commitIndex, pBuf->matchIndex, pBuf->endIndex); @@ -332,6 +334,15 @@ int32_t syncLogBufferAccept(SSyncLogBuffer* pBuf, SSyncNode* pNode, SSyncRaftEnt goto _out; } + if(pNode->raftCfg.cfg.nodeInfo[pNode->raftCfg.cfg.myIndex].nodeRole == TAOS_SYNC_ROLE_LEARNER && + index > 0 && index > pBuf->totalIndex){ + pBuf->totalIndex = index; + sTrace("vgId:%d, update learner progress. index:%" PRId64 ", term:%" PRId64 ": prevterm:%" PRId64 + " != lastmatch:%" PRId64 ". log buffer: [%" PRId64 " %" PRId64 " %" PRId64 ", %" PRId64 ")", + pNode->vgId, pEntry->index, pEntry->term, prevTerm, lastMatchTerm, pBuf->startIndex, pBuf->commitIndex, + pBuf->matchIndex, pBuf->endIndex); + } + if (index - pBuf->startIndex >= pBuf->size) { sWarn("vgId:%d, out of buffer range. index:%" PRId64 ", term:%" PRId64 ". log buffer: [%" PRId64 " %" PRId64 " %" PRId64 ", %" PRId64 ")", @@ -491,7 +502,7 @@ _out: } int32_t syncFsmExecute(SSyncNode* pNode, SSyncFSM* pFsm, ESyncState role, SyncTerm term, SSyncRaftEntry* pEntry, - int32_t applyCode) { + int32_t applyCode) { if (pNode->replicaNum == 1 && pNode->restoreFinish && pNode->vgId != 1) { return 0; } @@ -965,7 +976,7 @@ void syncLogReplDestroy(SSyncLogReplMgr* pMgr) { } int32_t syncNodeLogReplInit(SSyncNode* pNode) { - for (int i = 0; i < TSDB_MAX_REPLICA; i++) { + for (int i = 0; i < TSDB_MAX_REPLICA + TSDB_MAX_LEARNER_REPLICA; i++) { ASSERT(pNode->logReplMgrs[i] == NULL); pNode->logReplMgrs[i] = syncLogReplCreate(); if (pNode->logReplMgrs[i] == NULL) { @@ -978,7 +989,7 @@ int32_t syncNodeLogReplInit(SSyncNode* pNode) { } void syncNodeLogReplDestroy(SSyncNode* pNode) { - for (int i = 0; i < TSDB_MAX_REPLICA; i++) { + for (int i = 0; i < TSDB_MAX_REPLICA + TSDB_MAX_LEARNER_REPLICA; i++) { syncLogReplDestroy(pNode->logReplMgrs[i]); pNode->logReplMgrs[i] = NULL; } @@ -1108,7 +1119,7 @@ int32_t syncLogBufferReset(SSyncLogBuffer* pBuf, SSyncNode* pNode) { pBuf->endIndex = pBuf->matchIndex + 1; // reset repl mgr - for (int i = 0; i < pNode->replicaNum; i++) { + for (int i = 0; i < pNode->totalReplicaNum; i++) { SSyncLogReplMgr* pMgr = pNode->logReplMgrs[i]; syncLogReplReset(pMgr); } diff --git a/source/libs/sync/src/syncRaftCfg.c b/source/libs/sync/src/syncRaftCfg.c index f780e255ce..480ed4b2af 100644 --- a/source/libs/sync/src/syncRaftCfg.c +++ b/source/libs/sync/src/syncRaftCfg.c @@ -18,6 +18,28 @@ #include "syncUtil.h" #include "tjson.h" +const char* syncRoleToStr(ESyncRole role) { + switch (role) { + case TAOS_SYNC_ROLE_VOTER: + return "true"; + case TAOS_SYNC_ROLE_LEARNER: + return "false"; + default: + return "unknown"; + } +} + +const ESyncRole syncStrToRole(char* str) { + if(strcmp(str, "true") == 0){ + return TAOS_SYNC_ROLE_VOTER; + } + if(strcmp(str, "false") == 0){ + return TAOS_SYNC_ROLE_LEARNER; + } + + return TAOS_SYNC_ROLE_ERROR; +} + static int32_t syncEncodeSyncCfg(const void *pObj, SJson *pJson) { SSyncCfg *pCfg = (SSyncCfg *)pObj; if (tjsonAddDoubleToObject(pJson, "replicaNum", pCfg->replicaNum) < 0) return -1; @@ -26,13 +48,14 @@ static int32_t syncEncodeSyncCfg(const void *pObj, SJson *pJson) { SJson *nodeInfo = tjsonCreateArray(); if (nodeInfo == NULL) return -1; if (tjsonAddItemToObject(pJson, "nodeInfo", nodeInfo) < 0) return -1; - for (int32_t i = 0; i < pCfg->replicaNum; ++i) { + for (int32_t i = 0; i < pCfg->totalReplicaNum; ++i) { SJson *info = tjsonCreateObject(); if (info == NULL) return -1; if (tjsonAddDoubleToObject(info, "nodePort", pCfg->nodeInfo[i].nodePort) < 0) return -1; if (tjsonAddStringToObject(info, "nodeFqdn", pCfg->nodeInfo[i].nodeFqdn) < 0) return -1; if (tjsonAddIntegerToObject(info, "nodeId", pCfg->nodeInfo[i].nodeId) < 0) return -1; if (tjsonAddIntegerToObject(info, "clusterId", pCfg->nodeInfo[i].clusterId) < 0) return -1; + if (tjsonAddStringToObject(info, "isReplica", syncRoleToStr(pCfg->nodeInfo[i].nodeRole)) < 0) return -1; if (tjsonAddItemToArray(nodeInfo, info) < 0) return -1; } @@ -90,7 +113,8 @@ int32_t syncWriteCfgFile(SSyncNode *pNode) { if (taosRenameFile(file, realfile) != 0) goto _OVER; code = 0; - sInfo("vgId:%d, succeed to write sync cfg file:%s, len:%d", pNode->vgId, realfile, len); + sInfo("vgId:%d, succeed to write sync cfg file:%s, len:%d, lastConfigIndex:%" PRId64, pNode->vgId, + realfile, len, pNode->raftCfg.lastConfigIndex); _OVER: if (pJson != NULL) tjsonDelete(pJson); @@ -115,9 +139,9 @@ static int32_t syncDecodeSyncCfg(const SJson *pJson, void *pObj) { SJson *nodeInfo = tjsonGetObjectItem(pJson, "nodeInfo"); if (nodeInfo == NULL) return -1; - pCfg->replicaNum = tjsonGetArraySize(nodeInfo); + pCfg->totalReplicaNum = tjsonGetArraySize(nodeInfo); - for (int32_t i = 0; i < pCfg->replicaNum; ++i) { + for (int32_t i = 0; i < pCfg->totalReplicaNum; ++i) { SJson *info = tjsonGetArrayItem(nodeInfo, i); if (info == NULL) return -1; tjsonGetUInt16ValueFromDouble(info, "nodePort", pCfg->nodeInfo[i].nodePort, code); @@ -126,6 +150,15 @@ static int32_t syncDecodeSyncCfg(const SJson *pJson, void *pObj) { if (code < 0) return -1; tjsonGetNumberValue(info, "nodeId", pCfg->nodeInfo[i].nodeId, code); tjsonGetNumberValue(info, "clusterId", pCfg->nodeInfo[i].clusterId, code); + char role[10] = {0}; + code = tjsonGetStringValue(info, "isReplica", role); + if(code < 0) return -1; + if(strlen(role) != 0){ + pCfg->nodeInfo[i].nodeRole = syncStrToRole(role); + } + else{ + pCfg->nodeInfo[i].nodeRole = TAOS_SYNC_ROLE_VOTER; + } } return 0; diff --git a/source/libs/sync/src/syncReplication.c b/source/libs/sync/src/syncReplication.c index 2776225a39..5a3a3bbb70 100644 --- a/source/libs/sync/src/syncReplication.c +++ b/source/libs/sync/src/syncReplication.c @@ -64,10 +64,10 @@ int32_t syncNodeReplicate(SSyncNode* pNode) { } int32_t syncNodeReplicateWithoutLock(SSyncNode* pNode) { - if (pNode->state != TAOS_SYNC_STATE_LEADER || pNode->replicaNum == 1) { + if (pNode->state != TAOS_SYNC_STATE_LEADER || pNode->raftCfg.cfg.totalReplicaNum == 1) { return -1; } - for (int32_t i = 0; i < pNode->replicaNum; i++) { + for (int32_t i = 0; i < pNode->totalReplicaNum; i++) { if (syncUtilSameId(&pNode->replicasId[i], &pNode->myRaftId)) { continue; } diff --git a/source/libs/sync/src/syncSnapshot.c b/source/libs/sync/src/syncSnapshot.c index a9d0ddad17..763d4ec5d6 100644 --- a/source/libs/sync/src/syncSnapshot.c +++ b/source/libs/sync/src/syncSnapshot.c @@ -795,13 +795,18 @@ int32_t syncNodeOnSnapshot(SSyncNode *pSyncNode, const SRpcMsg *pRpcMsg) { return -1; } - if (pMsg->term > raftStoreGetTerm(pSyncNode)) { - syncNodeStepDown(pSyncNode, pMsg->term); + if(pSyncNode->raftCfg.cfg.nodeInfo[pSyncNode->raftCfg.cfg.myIndex].nodeRole != TAOS_SYNC_ROLE_LEARNER){ + if (pMsg->term > raftStoreGetTerm(pSyncNode)) { + syncNodeStepDown(pSyncNode, pMsg->term); + } + } + else{ + syncNodeUpdateTermWithoutStepDown(pSyncNode, pMsg->term); } // state, term, seq/ack int32_t code = 0; - if (pSyncNode->state == TAOS_SYNC_STATE_FOLLOWER) { + if (pSyncNode->state == TAOS_SYNC_STATE_FOLLOWER || pSyncNode->state == TAOS_SYNC_STATE_LEARNER) { if (pMsg->term == raftStoreGetTerm(pSyncNode)) { if (pMsg->seq == SYNC_SNAPSHOT_SEQ_PREP_SNAPSHOT) { syncLogRecvSyncSnapshotSend(pSyncNode, pMsg, "process seq pre-snapshot"); diff --git a/source/libs/sync/src/syncTimeout.c b/source/libs/sync/src/syncTimeout.c index a27be2853e..5ee67da9ab 100644 --- a/source/libs/sync/src/syncTimeout.c +++ b/source/libs/sync/src/syncTimeout.c @@ -58,7 +58,7 @@ static void syncNodeCleanConfigIndex(SSyncNode* ths) { static int32_t syncNodeTimerRoutine(SSyncNode* ths) { ths->tmrRoutineNum++; - if (ths->tmrRoutineNum % 60 == 0 && ths->replicaNum > 1) { + if (ths->tmrRoutineNum % 60 == 0 && ths->totalReplicaNum > 1) { sNInfo(ths, "timer routines"); } else { sNTrace(ths, "timer routines"); diff --git a/source/libs/sync/src/syncVoteMgr.c b/source/libs/sync/src/syncVoteMgr.c index 6bd9625276..83b0dde8e0 100644 --- a/source/libs/sync/src/syncVoteMgr.c +++ b/source/libs/sync/src/syncVoteMgr.c @@ -30,7 +30,7 @@ SVotesGranted *voteGrantedCreate(SSyncNode *pNode) { return NULL; } - pVotesGranted->replicas = &pNode->replicasId; + pVotesGranted->replicas = (void*)&pNode->replicasId; pVotesGranted->replicaNum = pNode->replicaNum; voteGrantedClearVotes(pVotesGranted); @@ -49,7 +49,7 @@ void voteGrantedDestroy(SVotesGranted *pVotesGranted) { } void voteGrantedUpdate(SVotesGranted *pVotesGranted, SSyncNode *pNode) { - pVotesGranted->replicas = &pNode->replicasId; + pVotesGranted->replicas = (void*)&pNode->replicasId; pVotesGranted->replicaNum = pNode->replicaNum; voteGrantedClearVotes(pVotesGranted); @@ -115,7 +115,7 @@ SVotesRespond *votesRespondCreate(SSyncNode *pNode) { return NULL; } - pVotesRespond->replicas = &pNode->replicasId; + pVotesRespond->replicas = (void*)&pNode->replicasId; pVotesRespond->replicaNum = pNode->replicaNum; pVotesRespond->term = 0; pVotesRespond->pNode = pNode; @@ -130,7 +130,7 @@ void votesRespondDestory(SVotesRespond *pVotesRespond) { } void votesRespondUpdate(SVotesRespond *pVotesRespond, SSyncNode *pNode) { - pVotesRespond->replicas = &pNode->replicasId; + pVotesRespond->replicas = (void*)&pNode->replicasId; pVotesRespond->replicaNum = pNode->replicaNum; pVotesRespond->term = 0; pVotesRespond->pNode = pNode; diff --git a/source/libs/wal/src/walRead.c b/source/libs/wal/src/walRead.c index 206cb75963..7ef7e010a5 100644 --- a/source/libs/wal/src/walRead.c +++ b/source/libs/wal/src/walRead.c @@ -108,6 +108,13 @@ int32_t walNextValidMsg(SWalReader *pReader) { int64_t walReaderGetCurrentVer(const SWalReader *pReader) { return pReader->curVersion; } int64_t walReaderGetValidFirstVer(const SWalReader *pReader) { return walGetFirstVer(pReader->pWal); } +void walReaderValidVersionRange(SWalReader *pReader, int64_t *sver, int64_t *ever) { + *sver = walGetFirstVer(pReader->pWal); + int64_t lastVer = walGetLastVer(pReader->pWal); + int64_t committedVer = walGetCommittedVer(pReader->pWal); + *ever = pReader->cond.scanUncommited ? lastVer : committedVer; +} + static int64_t walReadSeekFilePos(SWalReader *pReader, int64_t fileFirstVer, int64_t ver) { int64_t ret = 0; diff --git a/source/util/src/tcompare.c b/source/util/src/tcompare.c index dc57ed97b2..d6f0e898ef 100644 --- a/source/util/src/tcompare.c +++ b/source/util/src/tcompare.c @@ -1255,7 +1255,7 @@ int32_t taosArrayCompareString(const void *a, const void *b) { int32_t comparestrPatternMatch(const void *pLeft, const void *pRight) { SPatternCompareInfo pInfo = PATTERN_COMPARE_INFO_INITIALIZER; - ASSERT(varDataLen(pRight) <= TSDB_MAX_FIELD_LEN); + ASSERT(varDataTLen(pRight) <= TSDB_MAX_FIELD_LEN); size_t pLen = varDataLen(pRight); size_t sz = varDataLen(pLeft); diff --git a/source/util/src/terror.c b/source/util/src/terror.c index d9247a4e83..ee20f0f524 100644 --- a/source/util/src/terror.c +++ b/source/util/src/terror.c @@ -98,7 +98,8 @@ TAOS_DEFINE_ERROR(TSDB_CODE_NO_ENOUGH_DISKSPACE, "No enough disk space" TAOS_DEFINE_ERROR(TSDB_CODE_APP_IS_STARTING, "Database is starting up") TAOS_DEFINE_ERROR(TSDB_CODE_APP_IS_STOPPING, "Database is closing down") -TAOS_DEFINE_ERROR(TSDB_CODE_IVLD_DATA_FMT, "Invalid data format") +TAOS_DEFINE_ERROR(TSDB_CODE_INVALID_DATA_FMT, "Invalid data format") +TAOS_DEFINE_ERROR(TSDB_CODE_INVALID_CFG_VALUE, "Invalid configuration value") //client TAOS_DEFINE_ERROR(TSDB_CODE_TSC_INVALID_OPERATION, "Invalid operation") @@ -414,20 +415,14 @@ TAOS_DEFINE_ERROR(TSDB_CODE_GRANT_TABLE_LIMITED, "Table creation limite // sync TAOS_DEFINE_ERROR(TSDB_CODE_SYN_TIMEOUT, "Sync timeout") -TAOS_DEFINE_ERROR(TSDB_CODE_SYN_IS_LEADER, "Sync is leader") TAOS_DEFINE_ERROR(TSDB_CODE_SYN_NOT_LEADER, "Sync leader is unreachable") -TAOS_DEFINE_ERROR(TSDB_CODE_SYN_ONE_REPLICA, "Sync one replica") -TAOS_DEFINE_ERROR(TSDB_CODE_SYN_NOT_IN_NEW_CONFIG, "Sync not in new config") TAOS_DEFINE_ERROR(TSDB_CODE_SYN_NEW_CONFIG_ERROR, "Sync new config error") -TAOS_DEFINE_ERROR(TSDB_CODE_SYN_RECONFIG_NOT_READY, "Sync not ready for reconfig") -TAOS_DEFINE_ERROR(TSDB_CODE_SYN_PROPOSE_NOT_READY, "Sync not ready for propose") -TAOS_DEFINE_ERROR(TSDB_CODE_SYN_STANDBY_NOT_READY, "Sync not ready for standby") -TAOS_DEFINE_ERROR(TSDB_CODE_SYN_BATCH_ERROR, "Sync batch error") +TAOS_DEFINE_ERROR(TSDB_CODE_SYN_PROPOSE_NOT_READY, "Sync not ready to propose") TAOS_DEFINE_ERROR(TSDB_CODE_SYN_RESTORING, "Sync leader is restoring") TAOS_DEFINE_ERROR(TSDB_CODE_SYN_INVALID_SNAPSHOT_MSG, "Sync invalid snapshot msg") TAOS_DEFINE_ERROR(TSDB_CODE_SYN_BUFFER_FULL, "Sync buffer is full") TAOS_DEFINE_ERROR(TSDB_CODE_SYN_WRITE_STALL, "Sync write stall") -TAOS_DEFINE_ERROR(TSDB_CODE_SYN_NEGO_WIN_EXCEEDED, "Sync negotiation win exceeded") +TAOS_DEFINE_ERROR(TSDB_CODE_SYN_NEGOTIATION_WIN_FULL, "Sync negotiation win is full") TAOS_DEFINE_ERROR(TSDB_CODE_SYN_INTERNAL_ERROR, "Sync internal error") //tq @@ -445,8 +440,7 @@ TAOS_DEFINE_ERROR(TSDB_CODE_TQ_NO_COMMITTED_OFFSET, "TQ no committed offse // wal TAOS_DEFINE_ERROR(TSDB_CODE_WAL_FILE_CORRUPTED, "WAL file is corrupted") -TAOS_DEFINE_ERROR(TSDB_CODE_WAL_SIZE_LIMIT, "WAL size exceeds limit") -TAOS_DEFINE_ERROR(TSDB_CODE_WAL_INVALID_VER, "WAL use invalid version") +TAOS_DEFINE_ERROR(TSDB_CODE_WAL_INVALID_VER, "WAL invalid version") TAOS_DEFINE_ERROR(TSDB_CODE_WAL_LOG_NOT_EXIST, "WAL log not exist") TAOS_DEFINE_ERROR(TSDB_CODE_WAL_CHKSUM_MISMATCH, "WAL checksum mismatch") TAOS_DEFINE_ERROR(TSDB_CODE_WAL_LOG_INCOMPLETE, "WAL log incomplete") @@ -582,11 +576,9 @@ TAOS_DEFINE_ERROR(TSDB_CODE_FUNC_DUP_TIMESTAMP, "Duplicate timestamps TAOS_DEFINE_ERROR(TSDB_CODE_UDF_STOPPING, "udf is stopping") TAOS_DEFINE_ERROR(TSDB_CODE_UDF_PIPE_READ_ERR, "udf pipe read error") TAOS_DEFINE_ERROR(TSDB_CODE_UDF_PIPE_CONNECT_ERR, "udf pipe connect error") -TAOS_DEFINE_ERROR(TSDB_CODE_UDF_PIPE_NO_PIPE, "udf no pipe") +TAOS_DEFINE_ERROR(TSDB_CODE_UDF_PIPE_NOT_EXIST, "udf pipe not exist") TAOS_DEFINE_ERROR(TSDB_CODE_UDF_LOAD_UDF_FAILURE, "udf load failure") -TAOS_DEFINE_ERROR(TSDB_CODE_UDF_INVALID_STATE, "udf invalid state") TAOS_DEFINE_ERROR(TSDB_CODE_UDF_INVALID_INPUT, "udf invalid function input") -TAOS_DEFINE_ERROR(TSDB_CODE_UDF_NO_FUNC_HANDLE, "udf no function handle") TAOS_DEFINE_ERROR(TSDB_CODE_UDF_INVALID_BUFSIZE, "udf invalid bufsize") TAOS_DEFINE_ERROR(TSDB_CODE_UDF_INVALID_OUTPUT_TYPE, "udf invalid output type") TAOS_DEFINE_ERROR(TSDB_CODE_UDF_SCRIPT_NOT_SUPPORTED, "udf program language not supported") diff --git a/source/util/src/tlog.c b/source/util/src/tlog.c index d415379f92..c8ac15786f 100644 --- a/source/util/src/tlog.c +++ b/source/util/src/tlog.c @@ -23,10 +23,11 @@ #define LOG_MAX_LINE_SIZE (10024) #define LOG_MAX_LINE_BUFFER_SIZE (LOG_MAX_LINE_SIZE + 3) #define LOG_MAX_LINE_DUMP_SIZE (1024 * 1024) -#define LOG_MAX_LINE_DUMP_BUFFER_SIZE (LOG_MAX_LINE_DUMP_SIZE + 3) +#define LOG_MAX_LINE_DUMP_BUFFER_SIZE (LOG_MAX_LINE_DUMP_SIZE + 128) #define LOG_FILE_NAME_LEN 300 #define LOG_DEFAULT_BUF_SIZE (20 * 1024 * 1024) // 20MB +#define LOG_SLOW_BUF_SIZE (10 * 1024 * 1024) // 10MB #define LOG_DEFAULT_INTERVAL 25 #define LOG_INTERVAL_STEP 5 @@ -50,6 +51,8 @@ typedef struct { int32_t stop; TdThread asyncThread; TdThreadMutex buffMutex; + int32_t writeInterval; + int32_t lastDuration; } SLogBuff; typedef struct { @@ -61,6 +64,7 @@ typedef struct { pid_t pid; char logName[LOG_FILE_NAME_LEN]; SLogBuff *logHandle; + SLogBuff *slowHandle; TdThreadMutex logMutex; } SLogObj; @@ -68,7 +72,6 @@ extern SConfig *tsCfg; static int8_t tsLogInited = 0; static SLogObj tsLogObj = {.fileNum = 1}; static int64_t tsAsyncLogLostLines = 0; -static int32_t tsWriteInterval = LOG_DEFAULT_INTERVAL; static int32_t tsDaylightActive; /* Currently in daylight saving time. */ bool tsLogEmbedded = 0; @@ -81,6 +84,7 @@ int64_t tsNumOfErrorLogs = 0; int64_t tsNumOfInfoLogs = 0; int64_t tsNumOfDebugLogs = 0; int64_t tsNumOfTraceLogs = 0; +int64_t tsNumOfSlowLogs = 0; // log int32_t dDebugFlag = 131; @@ -135,6 +139,34 @@ static int32_t taosStartLog() { return 0; } +int32_t taosInitSlowLog() { + char fullName[PATH_MAX] = {0}; + char logFileName[64] = {0}; +#ifdef CUS_PROMPT + snprintf(logFileName, 64, "%sSlowLog", CUS_PROMPT); +#else + snprintf(logFileName, 64, "taosSlowLog"); +#endif + + if (strlen(tsLogDir) != 0) { + snprintf(fullName, PATH_MAX, "%s" TD_DIRSEP "%s", tsLogDir, logFileName); + } else { + snprintf(fullName, PATH_MAX, "%s", logFileName); + } + + tsLogObj.slowHandle = taosLogBuffNew(LOG_SLOW_BUF_SIZE); + if (tsLogObj.slowHandle == NULL) return -1; + + taosUmaskFile(0); + tsLogObj.slowHandle->pFile = taosOpenFile(fullName, TD_FILE_CREATE | TD_FILE_WRITE | TD_FILE_APPEND); + if (tsLogObj.slowHandle->pFile == NULL) { + printf("\nfailed to open slow log file:%s, reason:%s\n", fullName, strerror(errno)); + return -1; + } + + return 0; +} + int32_t taosInitLog(const char *logName, int32_t maxFiles) { if (atomic_val_compare_exchange_8(&tsLogInited, 0, 1) != 0) return 0; osUpdate(); @@ -150,6 +182,8 @@ int32_t taosInitLog(const char *logName, int32_t maxFiles) { tsLogObj.logHandle = taosLogBuffNew(LOG_DEFAULT_BUF_SIZE); if (tsLogObj.logHandle == NULL) return -1; if (taosOpenLogFile(fullName, tsNumOfLogLines, maxFiles) < 0) return -1; + + if (taosInitSlowLog() < 0) return -1; if (taosStartLog() < 0) return -1; return 0; } @@ -158,25 +192,34 @@ static void taosStopLog() { if (tsLogObj.logHandle) { tsLogObj.logHandle->stop = 1; } + if (tsLogObj.slowHandle) { + tsLogObj.slowHandle->stop = 1; + } } void taosCloseLog() { + taosStopLog(); + + if (tsLogObj.logHandle != NULL && taosCheckPthreadValid(tsLogObj.logHandle->asyncThread)) { + taosThreadJoin(tsLogObj.logHandle->asyncThread, NULL); + taosThreadClear(&tsLogObj.logHandle->asyncThread); + } + + if (tsLogObj.slowHandle != NULL) { + taosThreadMutexDestroy(&tsLogObj.slowHandle->buffMutex); + taosCloseFile(&tsLogObj.slowHandle->pFile); + taosMemoryFreeClear(tsLogObj.slowHandle->buffer); + taosMemoryFreeClear(tsLogObj.slowHandle); + } + if (tsLogObj.logHandle != NULL) { - taosStopLog(); - if (tsLogObj.logHandle != NULL && taosCheckPthreadValid(tsLogObj.logHandle->asyncThread)) { - taosThreadJoin(tsLogObj.logHandle->asyncThread, NULL); - taosThreadClear(&tsLogObj.logHandle->asyncThread); - } tsLogInited = 0; taosThreadMutexDestroy(&tsLogObj.logHandle->buffMutex); taosCloseFile(&tsLogObj.logHandle->pFile); taosMemoryFreeClear(tsLogObj.logHandle->buffer); - memset(&tsLogObj.logHandle->buffer, 0, sizeof(tsLogObj.logHandle->buffer)); taosThreadMutexDestroy(&tsLogObj.logMutex); taosMemoryFreeClear(tsLogObj.logHandle); - memset(&tsLogObj.logHandle, 0, sizeof(tsLogObj.logHandle)); - tsLogObj.logHandle = NULL; } } @@ -510,10 +553,9 @@ void taosPrintLongString(const char *flags, ELogLevel level, int32_t dflag, cons va_list argpointer; va_start(argpointer, format); - len += vsnprintf(buffer + len, LOG_MAX_LINE_DUMP_BUFFER_SIZE - len, format, argpointer); + len += vsnprintf(buffer + len, LOG_MAX_LINE_DUMP_BUFFER_SIZE - 2 - len, format, argpointer); va_end(argpointer); - if (len > LOG_MAX_LINE_DUMP_SIZE) len = LOG_MAX_LINE_DUMP_SIZE; buffer[len++] = '\n'; buffer[len] = 0; @@ -521,6 +563,31 @@ void taosPrintLongString(const char *flags, ELogLevel level, int32_t dflag, cons taosMemoryFree(buffer); } +void taosPrintSlowLog(const char *format, ...) { + if (!osLogSpaceAvailable()) return; + + char *buffer = taosMemoryMalloc(LOG_MAX_LINE_DUMP_BUFFER_SIZE); + int32_t len = taosBuildLogHead(buffer, ""); + + va_list argpointer; + va_start(argpointer, format); + len += vsnprintf(buffer + len, LOG_MAX_LINE_DUMP_BUFFER_SIZE - 2 - len, format, argpointer); + va_end(argpointer); + + buffer[len++] = '\n'; + buffer[len] = 0; + + atomic_add_fetch_64(&tsNumOfSlowLogs, 1); + + if (tsAsyncLog) { + taosPushLogBuffer(tsLogObj.slowHandle, buffer, len); + } else { + taosWriteFile(tsLogObj.slowHandle->pFile, buffer, len); + } + + taosMemoryFree(buffer); +} + void taosDumpData(unsigned char *msg, int32_t len) { if (!osLogSpaceAvailable()) return; taosUpdateLogNums(DEBUG_DUMP); @@ -565,6 +632,7 @@ static SLogBuff *taosLogBuffNew(int32_t bufSize) { LOG_BUF_SIZE(pLogBuf) = bufSize; pLogBuf->minBuffSize = bufSize / 10; pLogBuf->stop = 0; + pLogBuf->writeInterval = LOG_DEFAULT_INTERVAL; if (taosThreadMutexInit(&LOG_BUF_MUTEX(pLogBuf), NULL) < 0) goto _err; // tsem_init(&(pLogBuf->buffNotEmpty), 0, 0); @@ -648,83 +716,78 @@ static int32_t taosGetLogRemainSize(SLogBuff *pLogBuf, int32_t start, int32_t en } static void taosWriteLog(SLogBuff *pLogBuf) { - static int32_t lastDuration = 0; - int32_t remainChecked = 0; - int32_t start, end, pollSize; + int32_t start = LOG_BUF_START(pLogBuf); + int32_t end = LOG_BUF_END(pLogBuf); - do { - if (remainChecked == 0) { - start = LOG_BUF_START(pLogBuf); - end = LOG_BUF_END(pLogBuf); + if (start == end) { + dbgEmptyW++; + pLogBuf->writeInterval = LOG_MAX_INTERVAL; + return; + } - if (start == end) { - dbgEmptyW++; - tsWriteInterval = LOG_MAX_INTERVAL; - return; - } - - pollSize = taosGetLogRemainSize(pLogBuf, start, end); - if (pollSize < pLogBuf->minBuffSize) { - lastDuration += tsWriteInterval; - if (lastDuration < LOG_MAX_WAIT_MSEC) { - break; - } - } - - lastDuration = 0; + int32_t pollSize = taosGetLogRemainSize(pLogBuf, start, end); + if (pollSize < pLogBuf->minBuffSize) { + pLogBuf->lastDuration += pLogBuf->writeInterval; + if (pLogBuf->lastDuration < LOG_MAX_WAIT_MSEC) { + return; } + } - if (start < end) { - taosWriteFile(pLogBuf->pFile, LOG_BUF_BUFFER(pLogBuf) + start, pollSize); - } else { - int32_t tsize = LOG_BUF_SIZE(pLogBuf) - start; - taosWriteFile(pLogBuf->pFile, LOG_BUF_BUFFER(pLogBuf) + start, tsize); + pLogBuf->lastDuration = 0; - taosWriteFile(pLogBuf->pFile, LOG_BUF_BUFFER(pLogBuf), end); + if (start < end) { + taosWriteFile(pLogBuf->pFile, LOG_BUF_BUFFER(pLogBuf) + start, pollSize); + } else { + int32_t tsize = LOG_BUF_SIZE(pLogBuf) - start; + taosWriteFile(pLogBuf->pFile, LOG_BUF_BUFFER(pLogBuf) + start, tsize); + + taosWriteFile(pLogBuf->pFile, LOG_BUF_BUFFER(pLogBuf), end); + } + + dbgWN++; + dbgWSize += pollSize; + + if (pollSize < pLogBuf->minBuffSize) { + dbgSmallWN++; + if (pLogBuf->writeInterval < LOG_MAX_INTERVAL) { + pLogBuf->writeInterval += LOG_INTERVAL_STEP; } - - dbgWN++; - dbgWSize += pollSize; - - if (pollSize < pLogBuf->minBuffSize) { - dbgSmallWN++; - if (tsWriteInterval < LOG_MAX_INTERVAL) { - tsWriteInterval += LOG_INTERVAL_STEP; - } - } else if (pollSize > LOG_BUF_SIZE(pLogBuf) / 3) { - dbgBigWN++; - tsWriteInterval = LOG_MIN_INTERVAL; - } else if (pollSize > LOG_BUF_SIZE(pLogBuf) / 4) { - if (tsWriteInterval > LOG_MIN_INTERVAL) { - tsWriteInterval -= LOG_INTERVAL_STEP; - } + } else if (pollSize > LOG_BUF_SIZE(pLogBuf) / 3) { + dbgBigWN++; + pLogBuf->writeInterval = LOG_MIN_INTERVAL; + } else if (pollSize > LOG_BUF_SIZE(pLogBuf) / 4) { + if (pLogBuf->writeInterval > LOG_MIN_INTERVAL) { + pLogBuf->writeInterval -= LOG_INTERVAL_STEP; } + } - LOG_BUF_START(pLogBuf) = (LOG_BUF_START(pLogBuf) + pollSize) % LOG_BUF_SIZE(pLogBuf); + LOG_BUF_START(pLogBuf) = (LOG_BUF_START(pLogBuf) + pollSize) % LOG_BUF_SIZE(pLogBuf); - start = LOG_BUF_START(pLogBuf); - end = LOG_BUF_END(pLogBuf); + start = LOG_BUF_START(pLogBuf); + end = LOG_BUF_END(pLogBuf); - pollSize = taosGetLogRemainSize(pLogBuf, start, end); - if (pollSize < pLogBuf->minBuffSize) { - break; - } + pollSize = taosGetLogRemainSize(pLogBuf, start, end); + if (pollSize < pLogBuf->minBuffSize) { + return; + } - tsWriteInterval = LOG_MIN_INTERVAL; - - remainChecked = 1; - } while (1); + pLogBuf->writeInterval = 0; } static void *taosAsyncOutputLog(void *param) { - SLogBuff *pLogBuf = (SLogBuff *)param; + SLogBuff *pLogBuf = (SLogBuff *)tsLogObj.logHandle; + SLogBuff *pSlowBuf = (SLogBuff *)tsLogObj.slowHandle; + setThreadName("log"); int32_t count = 0; int32_t updateCron = 0; + int32_t writeInterval = 0; + while (1) { - count += tsWriteInterval; + writeInterval = TMIN(pLogBuf->writeInterval, pSlowBuf->writeInterval); + count += writeInterval; updateCron++; - taosMsleep(tsWriteInterval); + taosMsleep(writeInterval); if (count > 1000) { osUpdate(); count = 0; @@ -732,13 +795,14 @@ static void *taosAsyncOutputLog(void *param) { // Polling the buffer taosWriteLog(pLogBuf); + taosWriteLog(pSlowBuf); if (updateCron >= 3600 * 24 * 40 / 2) { taosUpdateDaylight(); updateCron = 0; } - if (pLogBuf->stop) break; + if (pLogBuf->stop || pSlowBuf->stop) break; } return NULL; diff --git a/tests/parallel_test/cases.task b/tests/parallel_test/cases.task index cbb62d2ae7..3ec1b03657 100644 --- a/tests/parallel_test/cases.task +++ b/tests/parallel_test/cases.task @@ -564,7 +564,7 @@ ,,y,system-test,./pytest.sh python3 ./test.py -f 1-insert/alter_replica.py -N 3 ,,y,system-test,./pytest.sh python3 ./test.py -f 1-insert/influxdb_line_taosc_insert.py ,,y,system-test,./pytest.sh python3 ./test.py -f 1-insert/opentsdb_telnet_line_taosc_insert.py -#,,n,system-test,./pytest.sh python3 ./test.py -f 1-insert/opentsdb_json_taosc_insert.py +#,,y,system-test,./pytest.sh python3 ./test.py -f 1-insert/opentsdb_json_taosc_insert.py ,,y,system-test,./pytest.sh python3 ./test.py -f 1-insert/test_stmt_muti_insert_query.py ,,y,system-test,./pytest.sh python3 ./test.py -f 1-insert/test_stmt_set_tbname_tag.py ,,y,system-test,./pytest.sh python3 ./test.py -f 1-insert/alter_stable.py @@ -769,6 +769,7 @@ ,,y,system-test,./pytest.sh python3 ./test.py -f 1-insert/delete_normaltable.py ,,y,system-test,./pytest.sh python3 ./test.py -f 1-insert/delete_systable.py ,,y,system-test,./pytest.sh python3 ./test.py -f 1-insert/keep_expired.py +,,y,system-test,./pytest.sh python3 ./test.py -f 1-insert/stmt_error.py ,,y,system-test,./pytest.sh python3 ./test.py -f 1-insert/drop.py ,,y,system-test,./pytest.sh python3 ./test.py -f 1-insert/drop.py -N 3 -M 3 -i False -n 3 ,,y,system-test,./pytest.sh python3 ./test.py -f 2-query/join2.py @@ -1320,6 +1321,7 @@ ,,y,script,./test.sh -f tsim/query/emptyTsRange.sim ,,y,script,./test.sh -f tsim/query/partitionby.sim ,,y,script,./test.sh -f tsim/query/tableCount.sim +,,y,script,./test.sh -f tsim/query/tag_scan.sim ,,y,script,./test.sh -f tsim/query/nullColSma.sim ,,y,script,./test.sh -f tsim/qnode/basic1.sim ,,y,script,./test.sh -f tsim/snode/basic1.sim diff --git a/tests/parallel_test/container_build.sh b/tests/parallel_test/container_build.sh index c6de1778a6..8a68e5f76d 100755 --- a/tests/parallel_test/container_build.sh +++ b/tests/parallel_test/container_build.sh @@ -97,7 +97,11 @@ docker run \ -v ${REP_REAL_PATH}/community/contrib/lz4/:${REP_DIR}/community/contrib/lz4 \ -v ${REP_REAL_PATH}/community/contrib/zlib/:${REP_DIR}/community/contrib/zlib \ -v ${REP_REAL_PATH}/community/contrib/jemalloc/:${REP_DIR}/community/contrib/jemalloc \ +<<<<<<< HEAD --rm --ulimit core=-1 taos_test:v1.0 sh -c "pip uninstall taospy -y;pip3 install taospy==2.7.2;cd $REP_DIR;rm -rf debug;mkdir -p debug;cd debug;cmake .. -DBUILD_HTTP=false -DBUILD_TOOLS=true -DBUILD_TEST=true -DWEBSOCKET=true -DBUILD_SANITIZER=1 -DTOOLS_SANITIZE=true -DTOOLS_BUILD_TYPE=Debug -DBUILD_TAOSX=true -DJEMALLOC_ENABLED=false -DPORTABLE=true -DWITH_GFLAGS=false;make -j || exit 1 " +======= + --rm --ulimit core=-1 taos_test:v1.0 sh -c "pip uninstall taospy -y;pip3 install taospy==2.7.2;cd $REP_DIR;rm -rf debug;mkdir -p debug;cd debug;cmake .. -DBUILD_HTTP=false -DBUILD_TOOLS=true -DBUILD_TEST=true -DWEBSOCKET=true -DBUILD_SANITIZER=1 -DTOOLS_SANITIZE=true -DTOOLS_BUILD_TYPE=Debug -DBUILD_TAOSX=true -DJEMALLOC_ENABLED=0;make -j || exit 1 " +>>>>>>> enh/3.0 mv ${REP_REAL_PATH}/debug ${WORKDIR}/debugSan diff --git a/tests/pytest/util/sql.py b/tests/pytest/util/sql.py index 783ee476cb..c041282bfc 100644 --- a/tests/pytest/util/sql.py +++ b/tests/pytest/util/sql.py @@ -431,7 +431,7 @@ class TDSql: time.sleep(1) continue - def execute(self, sql,queryTimes=10): + def execute(self, sql,queryTimes=30): self.sql = sql i=1 while i <= queryTimes: diff --git a/tests/script/api/makefile b/tests/script/api/makefile index 6739794cc8..6d55d8a75f 100644 --- a/tests/script/api/makefile +++ b/tests/script/api/makefile @@ -15,8 +15,11 @@ exe: gcc $(CFLAGS) ./stopquery.c -o $(ROOT)stopquery $(LFLAGS) gcc $(CFLAGS) ./dbTableRoute.c -o $(ROOT)dbTableRoute $(LFLAGS) gcc $(CFLAGS) ./insertSameTs.c -o $(ROOT)insertSameTs $(LFLAGS) + gcc $(CFLAGS) ./passwdTest.c -o $(ROOT)passwdTest $(LFLAGS) clean: rm $(ROOT)batchprepare rm $(ROOT)stopquery rm $(ROOT)dbTableRoute + rm $(ROOT)insertSameTs + rm $(ROOT)passwdTest diff --git a/tests/script/api/passwdTest.c b/tests/script/api/passwdTest.c new file mode 100644 index 0000000000..1bf4987689 --- /dev/null +++ b/tests/script/api/passwdTest.c @@ -0,0 +1,193 @@ +/* + * Copyright (c) 2019 TAOS Data, Inc. + * + * This program is free software: you can use, redistribute, and/or modify + * it under the terms of the GNU Affero General Public License, version 3 + * or later ("AGPL"), as published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. + * + * You should have received a copy of the GNU Affero General Public License + * along with this program. If not, see . + */ + +// TAOS standard API example. The same syntax as MySQL, but only a subset +// to compile: gcc -o demo demo.c -ltaos + +/** + * passwdTest.c + * - Run the test case in clear TDengine environment with default root passwd 'taosdata' + */ + +#include +#include +#include +#include +#include +#include "taos.h" // TAOS header file + +#define nDup 1 +#define nRoot 10 +#define nUser 10 +#define USER_LEN 24 + +void createUsers(TAOS *taos, const char *host, char *qstr); +void passVerTestMulti(const char *host, char *qstr); + +int nPassVerNotified = 0; +TAOS *taosu[nRoot] = {0}; +char users[nUser][USER_LEN] = {0}; + +void __taos_notify_cb(void *param, void *ext, int type) { + switch (type) { + case TAOS_NOTIFY_PASSVER: { + ++nPassVerNotified; + printf("%s:%d type:%d user:%s ver:%d\n", __func__, __LINE__, type, param ? (char *)param : "NULL", *(int *)ext); + break; + } + default: + printf("%s:%d unknown type:%d\n", __func__, __LINE__, type); + break; + } +} + +static void queryDB(TAOS *taos, char *command) { + int i; + TAOS_RES *pSql = NULL; + int32_t code = -1; + + for (i = 0; i < nDup; ++i) { + if (NULL != pSql) { + taos_free_result(pSql); + pSql = NULL; + } + + pSql = taos_query(taos, command); + code = taos_errno(pSql); + if (0 == code) { + break; + } + } + + if (code != 0) { + fprintf(stderr, "failed to run: %s, reason: %s\n", command, taos_errstr(pSql)); + taos_free_result(pSql); + taos_close(taos); + exit(EXIT_FAILURE); + } else { + fprintf(stderr, "success to run: %s\n", command); + } + + taos_free_result(pSql); +} + +int main(int argc, char *argv[]) { + char qstr[1024]; + + // connect to server + if (argc < 2) { + printf("please input server-ip \n"); + return 0; + } + + TAOS *taos = taos_connect(argv[1], "root", "taosdata", NULL, 0); + if (taos == NULL) { + printf("failed to connect to server, reason:%s\n", "null taos" /*taos_errstr(taos)*/); + exit(1); + } + createUsers(taos, argv[1], qstr); + passVerTestMulti(argv[1], qstr); + + taos_close(taos); + taos_cleanup(); +} + +void createUsers(TAOS *taos, const char *host, char *qstr) { + // users + for (int i = 0; i < nUser; ++i) { + sprintf(users[i], "user%d", i); + sprintf(qstr, "CREATE USER %s PASS 'taosdata'", users[i]); + queryDB(taos, qstr); + + taosu[i] = taos_connect(host, users[i], "taosdata", NULL, 0); + if (taosu[i] == NULL) { + printf("failed to connect to server, user:%s, reason:%s\n", users[i], "null taos" /*taos_errstr(taos)*/); + exit(1); + } + + int code = taos_set_notify_cb(taosu[i], __taos_notify_cb, users[i], TAOS_NOTIFY_PASSVER); + + if (code != 0) { + fprintf(stderr, "failed to run: taos_set_notify_cb for user:%s since %d\n", users[i], code); + } else { + fprintf(stderr, "success to run: taos_set_notify_cb for user:%s\n", users[i]); + } + + // alter pass for users + sprintf(qstr, "alter user %s pass 'taos'", users[i]); + queryDB(taos, qstr); + } +} + +void passVerTestMulti(const char *host, char *qstr) { + // root + TAOS *taos[nRoot] = {0}; + char userName[USER_LEN] = "root"; + + for (int i = 0; i < nRoot; ++i) { + taos[i] = taos_connect(host, "root", "taosdata", NULL, 0); + if (taos[i] == NULL) { + printf("failed to connect to server, reason:%s\n", "null taos" /*taos_errstr(taos)*/); + exit(1); + } + + int code = taos_set_notify_cb(taos[i], __taos_notify_cb, userName, TAOS_NOTIFY_PASSVER); + + if (code != 0) { + fprintf(stderr, "failed to run: taos_set_notify_cb since %d\n", code); + } else { + fprintf(stderr, "success to run: taos_set_notify_cb\n"); + } + } + + queryDB(taos[0], "create database if not exists demo1 vgroups 1 minrows 10"); + queryDB(taos[0], "create database if not exists demo2 vgroups 1 minrows 10"); + queryDB(taos[0], "create database if not exists demo3 vgroups 1 minrows 10"); + + queryDB(taos[0], "create table demo1.stb (ts timestamp, c1 int) tags(t1 int)"); + queryDB(taos[0], "create table demo2.stb (ts timestamp, c1 int) tags(t1 int)"); + queryDB(taos[0], "create table demo3.stb (ts timestamp, c1 int) tags(t1 int)"); + + strcpy(qstr, "alter user root pass 'taos'"); + queryDB(taos[0], qstr); + + // calculate the nPassVerNotified for root and users + int nConn = nRoot + nUser; + + for (int i = 0; i < 15; ++i) { + if (nPassVerNotified >= nConn) break; + sleep(1); + } + + // close the taos_conn + for (int i = 0; i < nRoot; ++i) { + taos_close(taos[i]); + printf("%s:%d close taos[%d]\n", __func__, __LINE__, i); + sleep(1); + } + + for (int i = 0; i < nUser; ++i) { + taos_close(taosu[i]); + printf("%s:%d close taosu[%d]\n", __func__, __LINE__, i); + sleep(1); + } + + if (nPassVerNotified >= nConn) { + fprintf(stderr, "succeed to get passVer notification since nNotify %d >= nConn %d\n", nPassVerNotified, nConn); + } else { + fprintf(stderr, "failed to get passVer notification since nNotify %d < nConn %d\n", nPassVerNotified, nConn); + } + // sleep(300); +} \ No newline at end of file diff --git a/tests/script/local.supp b/tests/script/local.supp new file mode 100644 index 0000000000..562cddba46 --- /dev/null +++ b/tests/script/local.supp @@ -0,0 +1,227 @@ +{ + + Memcheck:Leak + match-leak-kinds: reachable + fun:_Znwm + fun:_ZN9__gnu_cxx13new_allocatorINSt8__detail10_Hash_nodeISt4pairIKjPFvPvEELb0EEEE8allocateEmPKv + fun:_ZNSt16allocator_traitsISaINSt8__detail10_Hash_nodeISt4pairIKjPFvPvEELb0EEEEE8allocateERS9_m + fun:_ZNSt8__detail16_Hashtable_allocISaINS_10_Hash_nodeISt4pairIKjPFvPvEELb0EEEEE16_M_allocate_nodeIJRKSt21piecewise_construct_tSt5tupleIJRS3_EESF_IJEEEEEPS8_DpOT_ + fun:_ZNSt10_HashtableIjSt4pairIKjPFvPvEESaIS5_ENSt8__detail10_Select1stESt8equal_toIjESt4hashIjENS7_18_Mod_range_hashingENS7_20_Default_ranged_hashENS7_20_Prime_rehash_policyENS7_17_Hashtable_traitsILb0ELb0ELb1EEEE12_Scoped_nodeC1IJRKSt21piecewise_construct_tSt5tupleIJRS1_EESO_IJEEEEEPNS7_16_Hashtable_allocISaINS7_10_Hash_nodeIS5_Lb0EEEEEEDpOT_ + fun:_ZNSt8__detail9_Map_baseIjSt4pairIKjPFvPvEESaIS6_ENS_10_Select1stESt8equal_toIjESt4hashIjENS_18_Mod_range_hashingENS_20_Default_ranged_hashENS_20_Prime_rehash_policyENS_17_Hashtable_traitsILb0ELb0ELb1EEELb1EEixERS2_ + fun:_ZNSt13unordered_mapIjPFvPvESt4hashIjESt8equal_toIjESaISt4pairIKjS2_EEEixERS8_ + fun:_ZN7rocksdb14ThreadLocalPtr10StaticMeta10SetHandlerEjPFvPvE + fun:_ZN7rocksdb14ThreadLocalPtrC1EPFvPvE + fun:_ZN7rocksdb16ColumnFamilyDataC1EjRKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEEPNS_7VersionEPNS_5CacheEPNS_18WriteBufferManagerERKNS_19ColumnFamilyOptionsERKNS_18ImmutableDBOptionsERKNS_11FileOptionsEPNS_15ColumnFamilySetEPNS_16BlockCacheTracerERKSt10shared_ptrINS_8IOTracerEES8_ + fun:_ZN7rocksdb15ColumnFamilySetC1ERKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEEPKNS_18ImmutableDBOptionsERKNS_11FileOptionsEPNS_5CacheEPNS_18WriteBufferManagerEPNS_15WriteControllerEPNS_16BlockCacheTracerERKSt10shared_ptrINS_8IOTracerEES8_ + fun:_ZN7rocksdb10VersionSetC1ERKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEEPKNS_18ImmutableDBOptionsERKNS_11FileOptionsEPNS_5CacheEPNS_18WriteBufferManagerEPNS_15WriteControllerEPNS_16BlockCacheTracerERKSt10shared_ptrINS_8IOTracerEES8_ + fun:_ZN7rocksdb6DBImplC1ERKNS_9DBOptionsERKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEEbbb + fun:_ZN7rocksdb6DBImpl4OpenERKNS_9DBOptionsERKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEERKSt6vectorINS_22ColumnFamilyDescriptorESaISD_EEPSC_IPNS_18ColumnFamilyHandleESaISJ_EEPPNS_2DBEbb + fun:_ZN7rocksdb2DB4OpenERKNS_9DBOptionsERKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEERKSt6vectorINS_22ColumnFamilyDescriptorESaISD_EEPSC_IPNS_18ColumnFamilyHandleESaISJ_EEPPS0_ + fun:_ZN7rocksdb2DB4OpenERKNS_7OptionsERKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEEPPS0_ + fun:rocksdb_open + fun:tsdbOpenRocksCache + fun:tsdbOpenCache + fun:tsdbOpen +} +{ + + Memcheck:Leak + match-leak-kinds: reachable + fun:_Znwm + fun:_ZN9__gnu_cxx13new_allocatorINSt8__detail10_Hash_nodeISt4pairIKjPFvPvEELb0EEEE8allocateEmPKv + fun:_ZNSt16allocator_traitsISaINSt8__detail10_Hash_nodeISt4pairIKjPFvPvEELb0EEEEE8allocateERS9_m + fun:_ZNSt8__detail16_Hashtable_allocISaINS_10_Hash_nodeISt4pairIKjPFvPvEELb0EEEEE16_M_allocate_nodeIJRKSt21piecewise_construct_tSt5tupleIJRS3_EESF_IJEEEEEPS8_DpOT_ + fun:_ZNSt10_HashtableIjSt4pairIKjPFvPvEESaIS5_ENSt8__detail10_Select1stESt8equal_toIjESt4hashIjENS7_18_Mod_range_hashingENS7_20_Default_ranged_hashENS7_20_Prime_rehash_policyENS7_17_Hashtable_traitsILb0ELb0ELb1EEEE12_Scoped_nodeC1IJRKSt21piecewise_construct_tSt5tupleIJRS1_EESO_IJEEEEEPNS7_16_Hashtable_allocISaINS7_10_Hash_nodeIS5_Lb0EEEEEEDpOT_ + fun:_ZNSt8__detail9_Map_baseIjSt4pairIKjPFvPvEESaIS6_ENS_10_Select1stESt8equal_toIjESt4hashIjENS_18_Mod_range_hashingENS_20_Default_ranged_hashENS_20_Prime_rehash_policyENS_17_Hashtable_traitsILb0ELb0ELb1EEELb1EEixERS2_ + fun:_ZNSt13unordered_mapIjPFvPvESt4hashIjESt8equal_toIjESaISt4pairIKjS2_EEEixERS8_ + fun:_ZN7rocksdb14ThreadLocalPtr10StaticMeta10SetHandlerEjPFvPvE + fun:_ZN7rocksdb14ThreadLocalPtrC1EPFvPvE + fun:_ZN7rocksdb16ColumnFamilyDataC1EjRKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEEPNS_7VersionEPNS_5CacheEPNS_18WriteBufferManagerERKNS_19ColumnFamilyOptionsERKNS_18ImmutableDBOptionsERKNS_11FileOptionsEPNS_15ColumnFamilySetEPNS_16BlockCacheTracerERKSt10shared_ptrINS_8IOTracerEES8_ + fun:_ZN7rocksdb15ColumnFamilySet18CreateColumnFamilyERKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEEjPNS_7VersionERKNS_19ColumnFamilyOptionsE + fun:_ZN7rocksdb10VersionSet18CreateColumnFamilyERKNS_19ColumnFamilyOptionsEPKNS_11VersionEditE + fun:_ZN7rocksdb18VersionEditHandler15CreateCfAndInitERKNS_19ColumnFamilyOptionsERKNS_11VersionEditE + fun:_ZN7rocksdb18VersionEditHandler10InitializeEv + fun:_ZN7rocksdb22VersionEditHandlerBase7IterateERNS_3log6ReaderEPNS_6StatusE + fun:_ZN7rocksdb10VersionSet7RecoverERKSt6vectorINS_22ColumnFamilyDescriptorESaIS2_EEbPNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE + fun:_ZN7rocksdb6DBImpl7RecoverERKSt6vectorINS_22ColumnFamilyDescriptorESaIS2_EEbbbPm + fun:_ZN7rocksdb6DBImpl4OpenERKNS_9DBOptionsERKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEERKSt6vectorINS_22ColumnFamilyDescriptorESaISD_EEPSC_IPNS_18ColumnFamilyHandleESaISJ_EEPPNS_2DBEbb + fun:_ZN7rocksdb2DB4OpenERKNS_9DBOptionsERKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEERKSt6vectorINS_22ColumnFamilyDescriptorESaISD_EEPSC_IPNS_18ColumnFamilyHandleESaISJ_EEPPS0_ + fun:_ZN7rocksdb2DB4OpenERKNS_7OptionsERKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEEPPS0_ +} +{ + + Memcheck:Leak + match-leak-kinds: reachable + fun:calloc + fun:__cxa_thread_atexit_impl + fun:__cxa_thread_atexit + fun:__tls_init + fun:_ZTWN7rocksdb12perf_contextE + fun:_ZN7rocksdb17InstrumentedMutex4LockEv + fun:_ZN7rocksdb21InstrumentedMutexLockC1EPNS_17InstrumentedMutexE + fun:_ZN7rocksdb5Timer8ShutdownEv + fun:_ZN7rocksdb5TimerD1Ev + fun:_ZNKSt14default_deleteIN7rocksdb5TimerEEclEPS1_ + fun:_ZNSt10unique_ptrIN7rocksdb5TimerESt14default_deleteIS1_EED1Ev + fun:_ZN7rocksdb21PeriodicWorkSchedulerD1Ev + fun:__run_exit_handlers + fun:exit + fun:(below main) +} +{ + + Memcheck:Leak + match-leak-kinds: reachable + fun:_Znwm + fun:_ZN7rocksdb24CacheEntryStatsCollectorINS_13InternalStats19CacheEntryRoleStatsEE9GetSharedEPNS_5CacheEPNS_11SystemClockEPSt10shared_ptrIS3_E + fun:_ZN7rocksdb13InternalStatsC1EiPNS_11SystemClockEPNS_16ColumnFamilyDataE + fun:_ZN7rocksdb16ColumnFamilyDataC1EjRKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEEPNS_7VersionEPNS_5CacheEPNS_18WriteBufferManagerERKNS_19ColumnFamilyOptionsERKNS_18ImmutableDBOptionsERKNS_11FileOptionsEPNS_15ColumnFamilySetEPNS_16BlockCacheTracerERKSt10shared_ptrINS_8IOTracerEES8_ + fun:_ZN7rocksdb15ColumnFamilySet18CreateColumnFamilyERKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEEjPNS_7VersionERKNS_19ColumnFamilyOptionsE + fun:_ZN7rocksdb10VersionSet18CreateColumnFamilyERKNS_19ColumnFamilyOptionsEPKNS_11VersionEditE + fun:_ZN7rocksdb18VersionEditHandler15CreateCfAndInitERKNS_19ColumnFamilyOptionsERKNS_11VersionEditE + fun:_ZN7rocksdb18VersionEditHandler10InitializeEv + fun:_ZN7rocksdb22VersionEditHandlerBase7IterateERNS_3log6ReaderEPNS_6StatusE + fun:_ZN7rocksdb10VersionSet7RecoverERKSt6vectorINS_22ColumnFamilyDescriptorESaIS2_EEbPNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE + fun:_ZN7rocksdb6DBImpl7RecoverERKSt6vectorINS_22ColumnFamilyDescriptorESaIS2_EEbbbPm + fun:_ZN7rocksdb6DBImpl4OpenERKNS_9DBOptionsERKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEERKSt6vectorINS_22ColumnFamilyDescriptorESaISD_EEPSC_IPNS_18ColumnFamilyHandleESaISJ_EEPPNS_2DBEbb + fun:_ZN7rocksdb2DB4OpenERKNS_9DBOptionsERKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEERKSt6vectorINS_22ColumnFamilyDescriptorESaISD_EEPSC_IPNS_18ColumnFamilyHandleESaISJ_EEPPS0_ + fun:_ZN7rocksdb2DB4OpenERKNS_7OptionsERKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEEPPS0_ + fun:rocksdb_open + fun:tsdbOpenRocksCache + fun:tsdbOpenCache + fun:tsdbOpen + fun:vnodeOpen + fun:vmProcessCreateVnodeReq +} +{ + + Memcheck:Leak + match-leak-kinds: reachable + fun:_Znwm + fun:_ZN7rocksdb12_GLOBAL__N_111GetRegistryEv + fun:_ZN7rocksdb23CopyCacheDeleterRoleMapEv + fun:_ZN7rocksdb13InternalStats19CacheEntryRoleStats15BeginCollectionEPNS_5CacheEPNS_11SystemClockEm + fun:_ZN7rocksdb24CacheEntryStatsCollectorINS_13InternalStats19CacheEntryRoleStatsEE12CollectStatsEii + fun:_ZN7rocksdb13InternalStats22CollectCacheEntryStatsEb + fun:_ZN7rocksdb6DBImpl9DumpStatsEv + fun:_ZZN7rocksdb21PeriodicWorkScheduler8RegisterEPNS_6DBImplEjjENKUlvE_clEv + fun:_ZSt13__invoke_implIvRZN7rocksdb21PeriodicWorkScheduler8RegisterEPNS0_6DBImplEjjEUlvE_JEET_St14__invoke_otherOT0_DpOT1_ + fun:_ZSt10__invoke_rIvRZN7rocksdb21PeriodicWorkScheduler8RegisterEPNS0_6DBImplEjjEUlvE_JEENSt9enable_ifIXsrSt6__and_IJSt7is_voidIT_ESt14__is_invocableIT0_JDpT1_EEEE5valueES9_E4typeEOSC_DpOSD_ + fun:_ZNSt17_Function_handlerIFvvEZN7rocksdb21PeriodicWorkScheduler8RegisterEPNS1_6DBImplEjjEUlvE_E9_M_invokeERKSt9_Any_data + fun:_ZNKSt8functionIFvvEEclEv + fun:_ZN7rocksdb5Timer3RunEv + fun:_ZSt13__invoke_implIvMN7rocksdb5TimerEFvvEPS1_JEET_St21__invoke_memfun_derefOT0_OT1_DpOT2_ + fun:_ZSt8__invokeIMN7rocksdb5TimerEFvvEJPS1_EENSt15__invoke_resultIT_JDpT0_EE4typeEOS6_DpOS7_ + fun:_ZNSt6thread8_InvokerISt5tupleIJMN7rocksdb5TimerEFvvEPS3_EEE9_M_invokeIJLm0ELm1EEEEvSt12_Index_tupleIJXspT_EEE + fun:_ZNSt6thread8_InvokerISt5tupleIJMN7rocksdb5TimerEFvvEPS3_EEEclEv + fun:_ZNSt6thread11_State_implINS_8_InvokerISt5tupleIJMN7rocksdb5TimerEFvvEPS4_EEEEE6_M_runEv + obj:/usr/lib/x86_64-linux-gnu/libstdc++.so.6.0.28 + fun:start_thread +} +{ + + Memcheck:Leak + match-leak-kinds: reachable + fun:_Znwm + fun:_ZN9__gnu_cxx13new_allocatorIPNSt8__detail15_Hash_node_baseEE8allocateEmPKv + fun:_ZNSt16allocator_traitsISaIPNSt8__detail15_Hash_node_baseEEE8allocateERS3_m + fun:_ZNSt8__detail16_Hashtable_allocISaINS_10_Hash_nodeIPN7rocksdb16ThreadStatusDataELb0EEEEE19_M_allocate_bucketsEm + fun:_ZNSt10_HashtableIPN7rocksdb16ThreadStatusDataES2_SaIS2_ENSt8__detail9_IdentityESt8equal_toIS2_ESt4hashIS2_ENS4_18_Mod_range_hashingENS4_20_Default_ranged_hashENS4_20_Prime_rehash_policyENS4_17_Hashtable_traitsILb0ELb1ELb1EEEE19_M_allocate_bucketsEm + fun:_ZNSt10_HashtableIPN7rocksdb16ThreadStatusDataES2_SaIS2_ENSt8__detail9_IdentityESt8equal_toIS2_ESt4hashIS2_ENS4_18_Mod_range_hashingENS4_20_Default_ranged_hashENS4_20_Prime_rehash_policyENS4_17_Hashtable_traitsILb0ELb1ELb1EEEE13_M_rehash_auxEmSt17integral_constantIbLb1EE + fun:_ZNSt10_HashtableIPN7rocksdb16ThreadStatusDataES2_SaIS2_ENSt8__detail9_IdentityESt8equal_toIS2_ESt4hashIS2_ENS4_18_Mod_range_hashingENS4_20_Default_ranged_hashENS4_20_Prime_rehash_policyENS4_17_Hashtable_traitsILb0ELb1ELb1EEEE9_M_rehashEmRKm + fun:_ZNSt10_HashtableIPN7rocksdb16ThreadStatusDataES2_SaIS2_ENSt8__detail9_IdentityESt8equal_toIS2_ESt4hashIS2_ENS4_18_Mod_range_hashingENS4_20_Default_ranged_hashENS4_20_Prime_rehash_policyENS4_17_Hashtable_traitsILb0ELb1ELb1EEEE21_M_insert_unique_nodeERKS2_mmPNS4_10_Hash_nodeIS2_Lb0EEEm + fun:_ZNSt10_HashtableIPN7rocksdb16ThreadStatusDataES2_SaIS2_ENSt8__detail9_IdentityESt8equal_toIS2_ESt4hashIS2_ENS4_18_Mod_range_hashingENS4_20_Default_ranged_hashENS4_20_Prime_rehash_policyENS4_17_Hashtable_traitsILb0ELb1ELb1EEEE9_M_insertIRKS2_NS4_10_AllocNodeISaINS4_10_Hash_nodeIS2_Lb0EEEEEEEESt4pairINS4_14_Node_iteratorIS2_Lb1ELb0EEEbEOT_RKT0_St17integral_constantIbLb1EEm + fun:_ZNSt8__detail12_Insert_baseIPN7rocksdb16ThreadStatusDataES3_SaIS3_ENS_9_IdentityESt8equal_toIS3_ESt4hashIS3_ENS_18_Mod_range_hashingENS_20_Default_ranged_hashENS_20_Prime_rehash_policyENS_17_Hashtable_traitsILb0ELb1ELb1EEEE6insertERKS3_ + fun:_ZNSt13unordered_setIPN7rocksdb16ThreadStatusDataESt4hashIS2_ESt8equal_toIS2_ESaIS2_EE6insertERKS2_ + fun:_ZN7rocksdb19ThreadStatusUpdater14RegisterThreadENS_12ThreadStatus10ThreadTypeEm + fun:_ZN7rocksdb16ThreadStatusUtil14RegisterThreadEPKNS_3EnvENS_12ThreadStatus10ThreadTypeE + fun:_ZN7rocksdb14ThreadPoolImpl4Impl15BGThreadWrapperEPv + fun:_ZSt13__invoke_implIvPFvPvEJPN7rocksdb16BGThreadMetadataEEET_St14__invoke_otherOT0_DpOT1_ + fun:_ZSt8__invokeIPFvPvEJPN7rocksdb16BGThreadMetadataEEENSt15__invoke_resultIT_JDpT0_EE4typeEOS7_DpOS8_ + fun:_ZNSt6thread8_InvokerISt5tupleIJPFvPvEPN7rocksdb16BGThreadMetadataEEEE9_M_invokeIJLm0ELm1EEEEvSt12_Index_tupleIJXspT_EEE + fun:_ZNSt6thread8_InvokerISt5tupleIJPFvPvEPN7rocksdb16BGThreadMetadataEEEEclEv + fun:_ZNSt6thread11_State_implINS_8_InvokerISt5tupleIJPFvPvEPN7rocksdb16BGThreadMetadataEEEEEE6_M_runEv + obj:/usr/lib/x86_64-linux-gnu/libstdc++.so.6.0.28 +} +{ + + Memcheck:Leak + match-leak-kinds: reachable + fun:_Znwm + fun:_ZN9__gnu_cxx13new_allocatorIPNSt8__detail15_Hash_node_baseEE8allocateEmPKv + fun:_ZNSt16allocator_traitsISaIPNSt8__detail15_Hash_node_baseEEE8allocateERS3_m + fun:_ZNSt8__detail16_Hashtable_allocISaINS_10_Hash_nodeISt4pairIKjPFvPvEELb0EEEEE19_M_allocate_bucketsEm + fun:_ZNSt10_HashtableIjSt4pairIKjPFvPvEESaIS5_ENSt8__detail10_Select1stESt8equal_toIjESt4hashIjENS7_18_Mod_range_hashingENS7_20_Default_ranged_hashENS7_20_Prime_rehash_policyENS7_17_Hashtable_traitsILb0ELb0ELb1EEEE19_M_allocate_bucketsEm + fun:_ZNSt10_HashtableIjSt4pairIKjPFvPvEESaIS5_ENSt8__detail10_Select1stESt8equal_toIjESt4hashIjENS7_18_Mod_range_hashingENS7_20_Default_ranged_hashENS7_20_Prime_rehash_policyENS7_17_Hashtable_traitsILb0ELb0ELb1EEEE13_M_rehash_auxEmSt17integral_constantIbLb1EE + fun:_ZNSt10_HashtableIjSt4pairIKjPFvPvEESaIS5_ENSt8__detail10_Select1stESt8equal_toIjESt4hashIjENS7_18_Mod_range_hashingENS7_20_Default_ranged_hashENS7_20_Prime_rehash_policyENS7_17_Hashtable_traitsILb0ELb0ELb1EEEE9_M_rehashEmRKm + fun:_ZNSt10_HashtableIjSt4pairIKjPFvPvEESaIS5_ENSt8__detail10_Select1stESt8equal_toIjESt4hashIjENS7_18_Mod_range_hashingENS7_20_Default_ranged_hashENS7_20_Prime_rehash_policyENS7_17_Hashtable_traitsILb0ELb0ELb1EEEE21_M_insert_unique_nodeERS1_mmPNS7_10_Hash_nodeIS5_Lb0EEEm + fun:_ZNSt8__detail9_Map_baseIjSt4pairIKjPFvPvEESaIS6_ENS_10_Select1stESt8equal_toIjESt4hashIjENS_18_Mod_range_hashingENS_20_Default_ranged_hashENS_20_Prime_rehash_policyENS_17_Hashtable_traitsILb0ELb0ELb1EEELb1EEixERS2_ + fun:_ZNSt13unordered_mapIjPFvPvESt4hashIjESt8equal_toIjESaISt4pairIKjS2_EEEixERS8_ + fun:_ZN7rocksdb14ThreadLocalPtr10StaticMeta10SetHandlerEjPFvPvE + fun:_ZN7rocksdb14ThreadLocalPtrC1EPFvPvE + fun:_ZN7rocksdb16ColumnFamilyDataC1EjRKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEEPNS_7VersionEPNS_5CacheEPNS_18WriteBufferManagerERKNS_19ColumnFamilyOptionsERKNS_18ImmutableDBOptionsERKNS_11FileOptionsEPNS_15ColumnFamilySetEPNS_16BlockCacheTracerERKSt10shared_ptrINS_8IOTracerEES8_ + fun:_ZN7rocksdb15ColumnFamilySetC1ERKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEEPKNS_18ImmutableDBOptionsERKNS_11FileOptionsEPNS_5CacheEPNS_18WriteBufferManagerEPNS_15WriteControllerEPNS_16BlockCacheTracerERKSt10shared_ptrINS_8IOTracerEES8_ + fun:_ZN7rocksdb10VersionSetC1ERKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEEPKNS_18ImmutableDBOptionsERKNS_11FileOptionsEPNS_5CacheEPNS_18WriteBufferManagerEPNS_15WriteControllerEPNS_16BlockCacheTracerERKSt10shared_ptrINS_8IOTracerEES8_ + fun:_ZN7rocksdb6DBImplC1ERKNS_9DBOptionsERKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEEbbb + fun:_ZN7rocksdb6DBImpl4OpenERKNS_9DBOptionsERKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEERKSt6vectorINS_22ColumnFamilyDescriptorESaISD_EEPSC_IPNS_18ColumnFamilyHandleESaISJ_EEPPNS_2DBEbb + fun:_ZN7rocksdb2DB4OpenERKNS_9DBOptionsERKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEERKSt6vectorINS_22ColumnFamilyDescriptorESaISD_EEPSC_IPNS_18ColumnFamilyHandleESaISJ_EEPPS0_ + fun:_ZN7rocksdb2DB4OpenERKNS_7OptionsERKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEEPPS0_ + fun:rocksdb_open +} +{ + + Memcheck:Leak + match-leak-kinds: reachable + fun:_Znwm + fun:_ZN7rocksdb12_GLOBAL__N_125CreateThreadStatusUpdaterEv + fun:_ZN7rocksdb12_GLOBAL__N_18PosixEnvC1Ev + fun:_ZN7rocksdb3Env7DefaultEv + fun:_ZN7rocksdb9DBOptionsC1Ev + fun:_ZN7rocksdb7OptionsC1Ev + fun:_ZN7rocksdb18ImmutableCFOptionsC1Ev + fun:_Z41__static_initialization_and_destruction_0ii + fun:_GLOBAL__sub_I_cf_options.cc + fun:__libc_csu_init + fun:(below main) +} +{ + + Memcheck:Leak + match-leak-kinds: reachable + fun:_Znwm + fun:_ZN7rocksdb14ThreadLocalPtr8InstanceEv + fun:_ZN7rocksdb14ThreadLocalPtr14InitSingletonsEv + fun:_ZN7rocksdb3Env7DefaultEv + fun:_ZN7rocksdb9DBOptionsC1Ev + fun:_ZN7rocksdb7OptionsC1Ev + fun:_ZN7rocksdb18ImmutableCFOptionsC1Ev + fun:_Z41__static_initialization_and_destruction_0ii + fun:_GLOBAL__sub_I_cf_options.cc + fun:__libc_csu_init + fun:(below main) +} +{ + + Memcheck:Leak + match-leak-kinds: reachable + fun:_Znwm + fun:_ZN7rocksdb24CacheEntryStatsCollectorINS_13InternalStats19CacheEntryRoleStatsEE9GetSharedEPNS_5CacheEPNS_11SystemClockEPSt10shared_ptrIS3_E + fun:_ZN7rocksdb13InternalStatsC1EiPNS_11SystemClockEPNS_16ColumnFamilyDataE + fun:_ZN7rocksdb16ColumnFamilyDataC1EjRKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEEPNS_7VersionEPNS_5CacheEPNS_18WriteBufferManagerERKNS_19ColumnFamilyOptionsERKNS_18ImmutableDBOptionsERKNS_11FileOptionsEPNS_15ColumnFamilySetEPNS_16BlockCacheTracerERKSt10shared_ptrINS_8IOTracerEES8_ + fun:_ZN7rocksdb15ColumnFamilySet18CreateColumnFamilyERKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEEjPNS_7VersionERKNS_19ColumnFamilyOptionsE + fun:_ZN7rocksdb10VersionSet18CreateColumnFamilyERKNS_19ColumnFamilyOptionsEPKNS_11VersionEditE + fun:_ZN7rocksdb18VersionEditHandler15CreateCfAndInitERKNS_19ColumnFamilyOptionsERKNS_11VersionEditE + fun:_ZN7rocksdb18VersionEditHandler10InitializeEv + fun:_ZN7rocksdb22VersionEditHandlerBase7IterateERNS_3log6ReaderEPNS_6StatusE + fun:_ZN7rocksdb10VersionSet7RecoverERKSt6vectorINS_22ColumnFamilyDescriptorESaIS2_EEbPNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE + fun:_ZN7rocksdb6DBImpl7RecoverERKSt6vectorINS_22ColumnFamilyDescriptorESaIS2_EEbbbPm + fun:_ZN7rocksdb6DBImpl4OpenERKNS_9DBOptionsERKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEERKSt6vectorINS_22ColumnFamilyDescriptorESaISD_EEPSC_IPNS_18ColumnFamilyHandleESaISJ_EEPPNS_2DBEbb + fun:_ZN7rocksdb2DB4OpenERKNS_9DBOptionsERKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEERKSt6vectorINS_22ColumnFamilyDescriptorESaISD_EEPSC_IPNS_18ColumnFamilyHandleESaISJ_EEPPS0_ + fun:_ZN7rocksdb2DB4OpenERKNS_7OptionsERKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEEPPS0_ + fun:rocksdb_open + fun:tsdbOpenRocksCache + fun:tsdbOpenCache + fun:tsdbOpen + fun:vnodeOpen + fun:vmOpenVnodeInThread +} diff --git a/tests/script/sh/exec.sh b/tests/script/sh/exec.sh index f548a4cc41..5061688de2 100755 --- a/tests/script/sh/exec.sh +++ b/tests/script/sh/exec.sh @@ -109,8 +109,10 @@ if [ "$EXEC_OPTON" = "start" ]; then if [ "$VALGRIND_OPTION" = "true" ]; then TT=`date +%s` #mkdir ${LOG_DIR}/${TT} - echo "nohup valgrind --log-file=${LOG_DIR}/valgrind-taosd-${NODE_NAME}-${TT}.log --tool=memcheck --leak-check=full --show-reachable=no --track-origins=yes --show-leak-kinds=all --num-callers=20 -v -v --workaround-gcc296-bugs=yes $EXE_DIR/taosd -c $CFG_DIR > /dev/null 2>&1 &" - nohup valgrind --log-file=${LOG_DIR}/valgrind-taosd-${NODE_NAME}-${TT}.log --tool=memcheck --leak-check=full --show-reachable=no --track-origins=yes --show-leak-kinds=all --num-callers=20 -v -v --workaround-gcc296-bugs=yes $EXE_DIR/taosd -c $CFG_DIR > /dev/null 2>&1 & + #echo "nohup valgrind --log-file=${LOG_DIR}/valgrind-taosd-${NODE_NAME}-${TT}.log --tool=memcheck --leak-check=full --show-reachable=no --track-origins=yes --show-leak-kinds=all --num-callers=20 -v -v --workaround-gcc296-bugs=yes $EXE_DIR/taosd -c $CFG_DIR > /dev/null 2>&1 &" + #nohup valgrind --log-file=${LOG_DIR}/valgrind-taosd-${NODE_NAME}-${TT}.log --gen-suppressions=all --tool=memcheck --leak-check=full --show-reachable=no --track-origins=yes --show-leak-kinds=all --num-callers=20 -v -v --workaround-gcc296-bugs=yes $EXE_DIR/taosd -c $CFG_DIR > /dev/null 2>&1 & + echo "nohup valgrind --log-file=${LOG_DIR}/valgrind-taosd-${NODE_NAME}-${TT}.log --suppressions=${SCRIPT_DIR}/local.supp --tool=memcheck --leak-check=full --show-reachable=no --track-origins=yes --show-leak-kinds=all --num-callers=20 -v -v --workaround-gcc296-bugs=yes $EXE_DIR/taosd -c $CFG_DIR > /dev/null 2>&1 &" + nohup valgrind --log-file=${LOG_DIR}/valgrind-taosd-${NODE_NAME}-${TT}.log --suppressions=${SCRIPT_DIR}/local.supp --tool=memcheck --leak-check=full --show-reachable=no --track-origins=yes --show-leak-kinds=all --num-callers=20 -v -v --workaround-gcc296-bugs=yes $EXE_DIR/taosd -c $CFG_DIR > /dev/null 2>&1 & else echo "nohup $EXE_DIR/taosd -c $CFG_DIR > /dev/null 2>&1 &" nohup $EXE_DIR/taosd -c $CFG_DIR > /dev/null 2> $ASAN_DIR/$NODE_NAME.asan & diff --git a/tests/script/tsim/dnode/split_vgroup_replica1.sim b/tests/script/tsim/dnode/split_vgroup_replica1.sim index 51d63d25f4..135f7a579e 100644 --- a/tests/script/tsim/dnode/split_vgroup_replica1.sim +++ b/tests/script/tsim/dnode/split_vgroup_replica1.sim @@ -93,6 +93,19 @@ endi print =============== step4: split print split vgroup 2 sql split vgroup 2 +$wt = 0 + stepwt1: + $wt = $wt + 1 + sleep 1000 + if $wt == 200 then + print ====> split vgroup not completed! + return -1 + endi +sql show transactions +if $rows != 0 then + print wait 1 seconds to alter + goto stepwt1 +endi print =============== step5: check split result sql show d1.tables diff --git a/tests/script/tsim/query/partitionby.sim b/tests/script/tsim/query/partitionby.sim index 8babd1aa8d..4c221e02d3 100644 --- a/tests/script/tsim/query/partitionby.sim +++ b/tests/script/tsim/query/partitionby.sim @@ -36,4 +36,34 @@ if $rows != 0 then return -1 endi +sql insert into tb0 values (now, 0); +sql insert into tb1 values (now, 1); +sql insert into tb2 values (now, 2); +sql insert into tb3 values (now, 3); +sql insert into tb4 values (now, 4); +sql insert into tb5 values (now, 5); +sql insert into tb6 values (now, 6); +sql insert into tb7 values (now, 7); + +sql select * from (select 1 from $mt1 where ts is not null partition by tbname limit 1); +if $rows != 8 then + return -1 +endi + +sql select count(*) from (select ts from $mt1 where ts is not null partition by tbname slimit 2); +if $rows != 1 then + return -1 +endi +if $data00 != 2 then + return -1 +endi + +sql select count(*) from (select ts from $mt1 where ts is not null partition by tbname limit 2); +if $rows != 1 then + return -1 +endi +if $data00 != 8 then + return -1 +endi + system sh/exec.sh -n dnode1 -s stop -x SIGINT diff --git a/tests/script/tsim/query/tag_scan.sim b/tests/script/tsim/query/tag_scan.sim new file mode 100644 index 0000000000..03e3a20632 --- /dev/null +++ b/tests/script/tsim/query/tag_scan.sim @@ -0,0 +1,48 @@ +system sh/stop_dnodes.sh +system sh/deploy.sh -n dnode1 -i 1 +system sh/exec.sh -n dnode1 -s start +sql connect +sql drop database if exists test +sql create database test; +sql use test; + +sql create table st(ts timestamp, f int) tags (t int); +sql insert into ct1 using st tags(1) values(now, 1); +sql insert into ct2 using st tags(2) values(now, 2); +sql insert into ct3 using st tags(3) values(now, 3); +sql insert into ct4 using st tags(4) values(now, 4); + +sql create table st2(ts timestamp, f int) tags (t int); +sql insert into ct21 using st2 tags(1) values(now, 1); +sql insert into ct22 using st2 tags(2) values(now, 2); +sql insert into ct23 using st2 tags(3) values(now, 3); +sql insert into ct24 using st2 tags(4) values(now, 4); + +sql select tbname, 1 from st group by tbname order by tbname; +print $rows $data00 $data10 $data20 +if $rows != 4 then + return -1 +endi +if $data00 != @ct1@ then + return -1 +endi +if $data10 != @ct2@ then + return -1 +endi +sql select tbname, 1 from st group by tbname slimit 0, 1; +print $rows +if $rows != 1 then + return -1 +endi +sql select tbname, 1 from st group by tbname slimit 2, 2; +print $rows $data00 $data10 +if $rows != 2 then + return -1 +endi +sql select tbname, 1 from st group by tbname order by tbname slimit 0, 1; +print $rows $data00 $data10 $data20 +if $rows != 4 then + return -1 +endi + +system sh/exec.sh -n dnode1 -s stop -x SIGINT diff --git a/tests/system-test/1-insert/boundary.py b/tests/system-test/1-insert/boundary.py index d3742ef5f9..29dcbc7c46 100644 --- a/tests/system-test/1-insert/boundary.py +++ b/tests/system-test/1-insert/boundary.py @@ -166,6 +166,61 @@ class TDTestCase: else: tdLog.exit("error info is not true") tdSql.execute('drop database db') + + def row_col_tag_maxlen_check(self): + tdSql.prepare() + tdSql.execute('use db') + tdSql.execute('create table if not exists stb1 (ts timestamp, c1 int,c2 binary(1000)) tags (city binary(16382))') + tdSql.error('create table if not exists stb1 (ts timestamp, c1 int,c2 binary(1000)) tags (city binary(16383))') + tdSql.execute('create table if not exists stb2 (ts timestamp, c0 tinyint, c1 int, c2 nchar(16379)) tags (city binary(16382))') + tdSql.error('create table if not exists stb2 (ts timestamp, c0 smallint, c1 int, c2 nchar(16379)) tags (city binary(16382))') + tdSql.execute('create table if not exists stb3 (ts timestamp, c1 int, c2 binary(65517)) tags (city binary(16382))') + tdSql.error('create table if not exists stb3 (ts timestamp, c0 bool, c1 int, c2 binary(65517)) tags (city binary(16382))') + # prepare the column and tag data + char100='abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMN0123456789abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMN0123456789' + tag_max_16382='' + binary_max_65517 = '' + nchar_max_16379='' + for num in range(163): + nchar_max_16379 += char100 + for num in range(4): + binary_max_65517 += char100 + + nchar_max_16379 += '0123456789012345678901234567890123456789012345678901234567890123456789012345678' + tag_max_16382 = nchar_max_16379 + tag_max_16382 += '9ab' + + for num in range(3): + binary_max_65517 += char100 + binary_max_65517 += '01234567890123456' + + # insert/query and check + tdSql.execute(f"create table ct1 using stb1 tags('{tag_max_16382}')") + tdSql.execute(f"create table ct2 using stb2 tags('{tag_max_16382}')") + tdSql.execute(f"create table ct3 using stb3 tags('{tag_max_16382}')") + tdSql.execute(f"insert into ct1 values (now,1,'nchar_max_16379')") + tdSql.execute(f"insert into ct2 values (now,1,1,'{nchar_max_16379}')") + tdSql.execute(f"insert into ct3 values (now,1,'{binary_max_65517}')") + + tdSql.query("select * from stb1") + tdSql.checkEqual(tdSql.queryResult[0][3],tag_max_16382) + + tdSql.query("select * from ct2") + tdSql.checkEqual(tdSql.queryResult[0][3],nchar_max_16379) + + tdSql.query("select * from stb2") + tdSql.checkEqual(tdSql.queryResult[0][3],nchar_max_16379) + tdSql.checkEqual(tdSql.queryResult[0][4],tag_max_16382) + + tdSql.query("select * from ct3") + tdSql.checkEqual(tdSql.queryResult[0][2],binary_max_65517) + + tdSql.query("select * from stb3") + tdSql.checkEqual(tdSql.queryResult[0][2],binary_max_65517) + tdSql.checkEqual(tdSql.queryResult[0][3],tag_max_16382) + + tdSql.execute('drop database db') + def run(self): self.dbname_length_check() self.tbname_length_check() @@ -174,6 +229,7 @@ class TDTestCase: self.username_length_check() self.password_length_check() self.sql_length_check() + self.row_col_tag_maxlen_check() def stop(self): tdSql.close() diff --git a/tests/system-test/1-insert/influxdb_line_taosc_insert.py b/tests/system-test/1-insert/influxdb_line_taosc_insert.py index 6372502484..3e9bad4efd 100644 --- a/tests/system-test/1-insert/influxdb_line_taosc_insert.py +++ b/tests/system-test/1-insert/influxdb_line_taosc_insert.py @@ -439,7 +439,7 @@ class TDTestCase: for input_sql in [self.genLongSql(127, 1)[0], self.genLongSql(1, 4093)[0]]: tdCom.cleanTb(dbname="test") self._conn.schemaless_insert([input_sql], TDSmlProtocolType.LINE.value, TDSmlTimestampType.NANO_SECOND.value) - for input_sql in [self.genLongSql(129, 1)[0], self.genLongSql(1, 4095)[0]]: + for input_sql in [self.genLongSql(128, 1)[0], self.genLongSql(1, 4094)[0]]: tdCom.cleanTb(dbname="test") try: self._conn.schemaless_insert([input_sql], TDSmlProtocolType.LINE.value, TDSmlTimestampType.NANO_SECOND.value) @@ -578,10 +578,16 @@ class TDTestCase: # binary stb_name = tdCom.getLongName(7, "letters") - input_sql = f'{stb_name},t0=t,t1="{tdCom.getLongName(16374, "letters")}" c0=f 1626006833639000000' + input_sql = f'{stb_name},t0=t,t1="{tdCom.getLongName(4091, "letters")}" c0=f 1626006833639000000' self._conn.schemaless_insert([input_sql], TDSmlProtocolType.LINE.value, TDSmlTimestampType.NANO_SECOND.value) - input_sql = f'{stb_name},t0=t,t1="{tdCom.getLongName(16375, "letters")}" c0=f 1626006833639000000' + input_sql = f'{stb_name},t0="a",t1="{tdCom.getLongName(4088, "letters")}" c0=f 1626006833639000000' + try: + self._conn.schemaless_insert([input_sql], TDSmlProtocolType.LINE.value, TDSmlTimestampType.NANO_SECOND.value) + except SchemalessError as err: + tdSql.checkNotEqual(err.errno, 0) + + input_sql = f'{stb_name},t0=t,t1="{tdCom.getLongName(4092, "letters")}" c0=f 1626006833639000000' try: self._conn.schemaless_insert([input_sql], TDSmlProtocolType.LINE.value, TDSmlTimestampType.NANO_SECOND.value) except SchemalessError as err: @@ -590,10 +596,10 @@ class TDTestCase: # nchar # * legal nchar could not be larger than 16374/4 stb_name = tdCom.getLongName(7, "letters") - input_sql = f'{stb_name},t0=t,t1=L"{tdCom.getLongName(4093, "letters")}" c0=f 1626006833639000000' + input_sql = f'{stb_name},t0=t,t1=L"{tdCom.getLongName(4090, "letters")}" c0=f 1626006833639000000' self._conn.schemaless_insert([input_sql], TDSmlProtocolType.LINE.value, TDSmlTimestampType.NANO_SECOND.value) - input_sql = f'{stb_name},t0=t,t1=L"{tdCom.getLongName(4094, "letters")}" c0=f 1626006833639000000' + input_sql = f'{stb_name},t0=t,t1=L"{tdCom.getLongName(4091, "letters")}" c0=f 1626006833639000000' try: self._conn.schemaless_insert([input_sql], TDSmlProtocolType.LINE.value, TDSmlTimestampType.NANO_SECOND.value) except SchemalessError as err: @@ -672,28 +678,34 @@ class TDTestCase: except SchemalessError as err: tdSql.checkNotEqual(err.errno, 0) - # # # binary - # stb_name = tdCom.getLongName(7, "letters") - # input_sql = f'{stb_name},t0=t c0=f,c1="{tdCom.getLongName(16374, "letters")}" 1626006833639000000' - # self._conn.schemaless_insert([input_sql], TDSmlProtocolType.LINE.value, TDSmlTimestampType.NANO_SECOND.value) + # binary + stb_name = tdCom.getLongName(7, "letters") + input_sql = f'{stb_name},t0=t c0=1i32,c1="{tdCom.getLongName(65517, "letters")}" 1626006833639000000' + self._conn.schemaless_insert([input_sql], TDSmlProtocolType.LINE.value, TDSmlTimestampType.NANO_SECOND.value) - # input_sql = f'{stb_name},t0=t c0=f,c1="{tdCom.getLongName(16375, "letters")}" 1626006833639000000' - # try: - # self._conn.schemaless_insert([input_sql], TDSmlProtocolType.LINE.value, TDSmlTimestampType.NANO_SECOND.value) - # except SchemalessError as err: - # tdSql.checkNotEqual(err.errno, 0) + input_sql = f'{stb_name},t0=t c0=1i32,c1="{tdCom.getLongName(65517, "letters")},c2=f" 1626006833639000000' + try: + self._conn.schemaless_insert([input_sql], TDSmlProtocolType.LINE.value, TDSmlTimestampType.NANO_SECOND.value) + except SchemalessError as err: + tdSql.checkNotEqual(err.errno, 0) - # # nchar - # # * legal nchar could not be larger than 16374/4 - # stb_name = tdCom.getLongName(7, "letters") - # input_sql = f'{stb_name},t0=t c0=f,c1=L"{tdCom.getLongName(4093, "letters")}" 1626006833639000000' - # self._conn.schemaless_insert([input_sql], TDSmlProtocolType.LINE.value, TDSmlTimestampType.NANO_SECOND.value) + input_sql = f'{stb_name},t0=t c0=f,c1="{tdCom.getLongName(65518, "letters")}" 1626006833639000000' + try: + self._conn.schemaless_insert([input_sql], TDSmlProtocolType.LINE.value, TDSmlTimestampType.NANO_SECOND.value) + except SchemalessError as err: + tdSql.checkNotEqual(err.errno, 0) - # input_sql = f'{stb_name},t0=t c0=f,c1=L"{tdCom.getLongName(4094, "letters")}" 1626006833639000000' - # try: - # self._conn.schemaless_insert([input_sql], TDSmlProtocolType.LINE.value, TDSmlTimestampType.NANO_SECOND.value) - # except SchemalessError as err: - # tdSql.checkNotEqual(err.errno, 0) + # nchar + # * legal nchar could not be larger than 16374/4 + stb_name = tdCom.getLongName(7, "letters") + input_sql = f'{stb_name},t0=t c0=1i32,c1=L"{tdCom.getLongName(16379, "letters")}",c2=f 1626006833639000000' + self._conn.schemaless_insert([input_sql], TDSmlProtocolType.LINE.value, TDSmlTimestampType.NANO_SECOND.value) + + input_sql = f'{stb_name},t0=t c0=1i32,c1=L"{tdCom.getLongName(16380, "letters")}",c2=1i16 1626006833639000000' + try: + self._conn.schemaless_insert([input_sql], TDSmlProtocolType.LINE.value, TDSmlTimestampType.NANO_SECOND.value) + except SchemalessError as err: + tdSql.checkNotEqual(err.errno, 0) def tagColIllegalValueCheckCase(self): @@ -884,7 +896,7 @@ class TDTestCase: tdSql.checkRows(2) tdSql.checkNotEqual(tb_name1, tb_name3) - # * tag binary max is 16384, col+ts binary max 49151 + # * tag binary max is 16384-2, col+ts binary max 65531 def tagColBinaryMaxLengthCheckCase(self): """ every binary and nchar must be length+2 @@ -896,35 +908,59 @@ class TDTestCase: self._conn.schemaless_insert([input_sql], TDSmlProtocolType.LINE.value, TDSmlTimestampType.NANO_SECOND.value) # * every binary and nchar must be length+2, so here is two tag, max length could not larger than 16384-2*2 - input_sql = f'{stb_name},t0=t,t1="{tdCom.getLongName(16374, "letters")}",t2="{tdCom.getLongName(5, "letters")}" c0=f 1626006833639000000' + stb_name = tdCom.getLongName(8, "letters") + input_sql = f'{stb_name},t0=f,t1="{tdCom.getLongName(4091, "letters")}", c0=f 1626006833639000000' self._conn.schemaless_insert([input_sql], TDSmlProtocolType.LINE.value, TDSmlTimestampType.NANO_SECOND.value) tdSql.query(f"select * from {stb_name}") - tdSql.checkRows(2) - input_sql = f'{stb_name},t0=t,t1="{tdCom.getLongName(16374, "letters")}",t2="{tdCom.getLongName(6, "letters")}" c0=f 1626006833639000000' + tdSql.checkRows(1) + + input_sql = f'{stb_name},t0=t,t1="{tdCom.getLongName(4092, "letters")}", c0=f 1626006833639000000' try: self._conn.schemaless_insert([input_sql], TDSmlProtocolType.LINE.value, TDSmlTimestampType.NANO_SECOND.value) raise Exception("should not reach here") except SchemalessError as err: tdSql.checkNotEqual(err.errno, 0) tdSql.query(f"select * from {stb_name}") - tdSql.checkRows(2) + tdSql.checkRows(1) + + stb_name = tdCom.getLongName(9, "letters") # # * check col，col+ts max in describe ---> 16143 - input_sql = f'{stb_name},t0=t c0=f,c1="{tdCom.getLongName(16374, "letters")}",c2="{tdCom.getLongName(16374, "letters")}",c3="{tdCom.getLongName(16374, "letters")}",c4="{tdCom.getLongName(12, "letters")}" 1626006833639000000' + input_sql = f'{stb_name},t0=t c0=1i32,c1="{tdCom.getLongName(65517, "letters")}" 1626006833639000000' self._conn.schemaless_insert([input_sql], TDSmlProtocolType.LINE.value, TDSmlTimestampType.NANO_SECOND.value) + input_sql = f'{stb_name},t0=t c0=1i32,c1="{tdCom.getLongName(65517, "letters")}",c2=f 1626006833639000000' + try: + self._conn.schemaless_insert([input_sql], TDSmlProtocolType.LINE.value, TDSmlTimestampType.NANO_SECOND.value) + except SchemalessError as err: + tdSql.checkNotEqual(err.errno, 0) + tdSql.query(f"select * from {stb_name}") - tdSql.checkRows(3) + tdSql.checkRows(1) + + + stb_name = tdCom.getLongName(10, "letters") + input_sql = f'{stb_name},t0=t c0=1i16,c1="{tdCom.getLongName(49133, "letters")}",c2="{tdCom.getLongName(16384, "letters")}" 1626006833639000000' + self._conn.schemaless_insert([input_sql], TDSmlProtocolType.LINE.value, TDSmlTimestampType.NANO_SECOND.value) + + input_sql = f'{stb_name},t0=t c0=1i16,c1="{tdCom.getLongName(49133, "letters")}",c2="{tdCom.getLongName(16384, "letters")},c3=t" 1626006833639000000' + try: + self._conn.schemaless_insert([input_sql], TDSmlProtocolType.LINE.value, TDSmlTimestampType.NANO_SECOND.value) + except SchemalessError as err: + tdSql.checkNotEqual(err.errno, 0) + + tdSql.query(f"select * from {stb_name}") + tdSql.checkRows(1) + input_sql = f'{stb_name},t0=t c0=f,c1="{tdCom.getLongName(16374, "letters")}",c2="{tdCom.getLongName(16374, "letters")}",c3="{tdCom.getLongName(16374, "letters")}",c4="{tdCom.getLongName(13, "letters")}" 1626006833639000000' try: self._conn.schemaless_insert([input_sql], TDSmlProtocolType.LINE.value, TDSmlTimestampType.NANO_SECOND.value) except SchemalessError as err: tdSql.checkNotEqual(err.errno, 0) - tdSql.query(f"select * from {stb_name}") - tdSql.checkRows(3) - # * tag nchar max is 16374/4, col+ts nchar max 49151 + + # * tag nchar max is (16384-2)/4, col+ts nchar max 65531 def tagColNcharMaxLengthCheckCase(self): """ check nchar length limit @@ -935,30 +971,31 @@ class TDTestCase: input_sql = f'{stb_name},id="{tb_name}",t0=t c0=f 1626006833639000000' code = self._conn.schemaless_insert([input_sql], TDSmlProtocolType.LINE.value, TDSmlTimestampType.NANO_SECOND.value) - # * legal nchar could not be larger than 16374/4 - input_sql = f'{stb_name},t0=t,t1=L"{tdCom.getLongName(4093, "letters")}",t2=L"{tdCom.getLongName(1, "letters")}" c0=f 1626006833639000000' - self._conn.schemaless_insert([input_sql], TDSmlProtocolType.LINE.value, TDSmlTimestampType.NANO_SECOND.value) - tdSql.query(f"select * from {stb_name}") - tdSql.checkRows(2) - input_sql = f'{stb_name},t0=t,t1=L"{tdCom.getLongName(4093, "letters")}",t2=L"{tdCom.getLongName(2, "letters")}" c0=f 1626006833639000000' - try: - self._conn.schemaless_insert([input_sql], TDSmlProtocolType.LINE.value, TDSmlTimestampType.NANO_SECOND.value) - except SchemalessError as err: - tdSql.checkNotEqual(err.errno, 0) - tdSql.query(f"select * from {stb_name}") - tdSql.checkRows(2) + # * legal tag nchar could not be larger than (16384-2)/4 + # input_sql = f'{stb_name},t0=t,t1=L"{tdCom.getLongName(4093, "letters")}",t2=L"{tdCom.getLongName(1, "letters")}" c0=f 1626006833639000000' + # self._conn.schemaless_insert([input_sql], TDSmlProtocolType.LINE.value, TDSmlTimestampType.NANO_SECOND.value) + # tdSql.query(f"select * from {stb_name}") + # tdSql.checkRows(2) + # input_sql = f'{stb_name},t0=t,t1=L"{tdCom.getLongName(4093, "letters")}",t2=L"{tdCom.getLongName(2, "letters")}" c0=f 1626006833639000000' + # try: + # self._conn.schemaless_insert([input_sql], TDSmlProtocolType.LINE.value, TDSmlTimestampType.NANO_SECOND.value) + # except SchemalessError as err: + # tdSql.checkNotEqual(err.errno, 0) + # tdSql.query(f"select * from {stb_name}") + # tdSql.checkRows(2) input_sql = f'{stb_name},t0=t c0=f,c1=L"{tdCom.getLongName(4093, "letters")}",c2=L"{tdCom.getLongName(4093, "letters")}",c3=L"{tdCom.getLongName(4093, "letters")}",c4=L"{tdCom.getLongName(4, "letters")}" 1626006833639000000' self._conn.schemaless_insert([input_sql], TDSmlProtocolType.LINE.value, TDSmlTimestampType.NANO_SECOND.value) tdSql.query(f"select * from {stb_name}") - tdSql.checkRows(3) + tdSql.checkRows(2) + input_sql = f'{stb_name},t0=t c0=f,c1=L"{tdCom.getLongName(4093, "letters")}",c2=L"{tdCom.getLongName(4093, "letters")}",c3=L"{tdCom.getLongName(4093, "letters")}",c4=L"{tdCom.getLongName(5, "letters")}" 1626006833639000000' try: self._conn.schemaless_insert([input_sql], TDSmlProtocolType.LINE.value, TDSmlTimestampType.NANO_SECOND.value) except SchemalessError as err: tdSql.checkNotEqual(err.errno, 0) tdSql.query(f"select * from {stb_name}") - tdSql.checkRows(3) + tdSql.checkRows(2) def batchInsertCheckCase(self): """ @@ -1274,13 +1311,13 @@ class TDTestCase: self.idSeqCheckCase() self.idUpperCheckCase() self.noIdCheckCase() - # self.maxColTagCheckCase() + self.maxColTagCheckCase() self.idIllegalNameCheckCase() self.idStartWithNumCheckCase() self.nowTsCheckCase() self.dateFormatTsCheckCase() self.illegalTsCheckCase() - # self.tagValueLengthCheckCase() + self.tagValueLengthCheckCase() self.colValueLengthCheckCase() self.tagColIllegalValueCheckCase() self.duplicateIdTagColInsertCheckCase() @@ -1290,8 +1327,8 @@ class TDTestCase: self.tagColAddDupIDCheckCase() self.tagColAddCheckCase() self.tagMd5Check() - # self.tagColBinaryMaxLengthCheckCase() - # self.tagColNcharMaxLengthCheckCase() + self.tagColBinaryMaxLengthCheckCase() + self.tagColNcharMaxLengthCheckCase() self.batchInsertCheckCase() self.multiInsertCheckCase(10) self.batchErrorInsertCheckCase() diff --git a/tests/system-test/1-insert/stmt_error.py b/tests/system-test/1-insert/stmt_error.py new file mode 100644 index 0000000000..c6d747c317 --- /dev/null +++ b/tests/system-test/1-insert/stmt_error.py @@ -0,0 +1,224 @@ +# encoding:UTF-8 +from taos import * + +from ctypes import * +from datetime import datetime +import taos + +import taos +import time + +from util.log import * +from util.cases import * +from util.sql import * +from util.dnodes import * + +class TDTestCase: + def __init__(self): + self.err_case = 0 + self.curret_case = 0 + + def caseDescription(self): + + ''' + case1 : [TD-11899] : this is an test case for check stmt error use . + ''' + return + + def init(self, conn, logSql, replicaVar=1): + self.replicaVar = int(replicaVar) + tdLog.debug("start to execute %s" % __file__) + tdSql.init(conn.cursor(), logSql) + + def conn(self): + # type: () -> taos.TaosConnection + return connect() + + def test_stmt_insert(self,conn): + # type: (TaosConnection) -> None + + dbname = "pytest_taos_stmt" + try: + conn.execute("drop database if exists %s" % dbname) + conn.execute("create database if not exists %s" % dbname) + conn.select_db(dbname) + + conn.execute( + "create table if not exists log(ts timestamp, bo bool, nil tinyint, ti tinyint, si smallint, ii int,\ + bi bigint, tu tinyint unsigned, su smallint unsigned, iu int unsigned, bu bigint unsigned, \ + ff float, dd double, bb binary(65059), nn nchar(100), tt timestamp)", + ) + conn.load_table_info("log") + + + stmt = conn.statement("insert into log values(?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?)") + params = new_bind_params(16) + params[0].timestamp(1626861392589, PrecisionEnum.Milliseconds) + params[1].bool(True) + params[2].tinyint(None) + params[3].tinyint(2) + params[4].smallint(3) + params[5].int(4) + params[6].bigint(5) + params[7].tinyint_unsigned(6) + params[8].smallint_unsigned(7) + params[9].int_unsigned(8) + params[10].bigint_unsigned(9) + params[11].float(10.1) + params[12].double(10.11) + binaryStr = '123456789' + for i in range(1301): + binaryStr += "1234567890abcdefghij1234567890abcdefghij12345hello" + params[13].binary(binaryStr) + params[14].nchar("stmt") + params[15].timestamp(1626861392589, PrecisionEnum.Milliseconds) + + stmt.bind_param(params) + stmt.execute() + + assert stmt.affected_rows == 1 + stmt.close() + + querystmt=conn.statement("select ?, bo, nil, ti, si, ii,bi, tu, su, iu, bu, ff, dd, bb, nn, tt from log") + queryparam=new_bind_params(1) + print(type(queryparam)) + queryparam[0].binary("ts") + querystmt.bind_param(queryparam) + querystmt.execute() + result=querystmt.use_result() + + row=result.fetch_all() + print(row) + + assert row[0][1] == True + assert row[0][2] == None + for i in range(3, 10): + assert row[0][i] == i - 1 + #float == may not work as expected + # assert row[0][11] == c_float(10.1) + assert row[0][12] == 10.11 + assert row[0][13][65054:] == "hello" + assert row[0][14] == "stmt" + + conn.execute("drop database if exists %s" % dbname) + conn.close() + + except Exception as err: + conn.execute("drop database if exists %s" % dbname) + conn.close() + raise err + + def test_stmt_insert_error(self,conn): + # type: (TaosConnection) -> None + + dbname = "pytest_taos_stmt_error" + try: + conn.execute("drop database if exists %s" % dbname) + conn.execute("create database if not exists %s" % dbname) + conn.select_db(dbname) + + conn.execute( + "create table if not exists log(ts timestamp, bo bool, nil tinyint, ti tinyint, si smallint, ii int,\ + bi bigint, tu tinyint unsigned, su smallint unsigned, iu int unsigned, bu bigint unsigned, \ + ff float, dd double, bb binary(100), nn nchar(100), tt timestamp , error_data int )", + ) + conn.load_table_info("log") + + + stmt = conn.statement("insert into log values(?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,1000)") + params = new_bind_params(16) + params[0].timestamp(1626861392589, PrecisionEnum.Milliseconds) + params[1].bool(True) + params[2].tinyint(None) + params[3].tinyint(2) + params[4].smallint(3) + params[5].int(4) + params[6].bigint(5) + params[7].tinyint_unsigned(6) + params[8].smallint_unsigned(7) + params[9].int_unsigned(8) + params[10].bigint_unsigned(9) + params[11].float(10.1) + params[12].double(10.11) + params[13].binary("hello") + params[14].nchar("stmt") + params[15].timestamp(1626861392589, PrecisionEnum.Milliseconds) + + stmt.bind_param(params) + stmt.execute() + + conn.close() + + except Exception as err: + conn.execute("drop database if exists %s" % dbname) + conn.close() + raise err + + def test_stmt_insert_error_null_timestamp(self,conn): + + dbname = "pytest_taos_stmt_error_null_ts" + try: + conn.execute("drop database if exists %s" % dbname) + conn.execute("create database if not exists %s" % dbname) + conn.execute("alter database %s keep 36500" % dbname) + conn.select_db(dbname) + + conn.execute("create stable STB(ts timestamp, n int) tags(b int)") + + stmt = conn.statement("insert into ? using STB tags(?) values(?, ?)") + params = new_bind_params(1) + params[0].int(4); + stmt.set_tbname_tags("ct", params); + + multi_params = new_multi_binds(2); + multi_params[0].timestamp([9223372036854775808]) + multi_params[1].int([123]) + stmt.bind_param_batch(multi_params) + + stmt.execute() + result = stmt.use_result() + + result.close() + stmt.close() + + stmt = conn.statement("select * from STB") + stmt.execute() + result = stmt.use_result() + print(result.affected_rows) + row = result.next() + print(row) + + result.close() + stmt.close() + conn.close() + + except Exception as err: + conn.close() + raise err + + def run(self): + + self.test_stmt_insert(self.conn()) + try: + self.test_stmt_insert_error(self.conn()) + except Exception as error : + + if str(error)=='[0x0200]: no mix usage for ? and values': + tdLog.info('=========stmt error occured for bind part column ==============') + else: + tdLog.exit("expect error(%s) not occured" % str(error)) + + try: + self.test_stmt_insert_error_null_timestamp(self.conn()) + tdLog.exit("expect error not occured - 1") + except Exception as error : + if str(error)=='[0x060b]: Timestamp data out of range': + tdLog.info('=========stmt error occured for bind part column(NULL Timestamp) ==============') + else: + tdLog.exit("expect error(%s) not occured - 2" % str(error)) + def stop(self): + tdSql.close() + tdLog.success("%s successfully executed" % __file__) + +tdCases.addWindows(__file__, TDTestCase()) +tdCases.addLinux(__file__, TDTestCase()) \ No newline at end of file diff --git a/tests/system-test/runAllOne.sh b/tests/system-test/runAllOne.sh index 58548f43db..a870e36935 100644 --- a/tests/system-test/runAllOne.sh +++ b/tests/system-test/runAllOne.sh @@ -287,6 +287,7 @@ python3 ./test.py -f 1-insert/tb_100w_data_order.py -P python3 ./test.py -f 1-insert/delete_childtable.py -P python3 ./test.py -f 1-insert/delete_normaltable.py -P python3 ./test.py -f 1-insert/keep_expired.py -P +python3 ./test.py -f 1-insert/stmt_error.py -P python3 ./test.py -f 1-insert/drop.py -P python3 ./test.py -f 2-query/join2.py -P python3 ./test.py -f 2-query/union1.py -P diff --git a/tests/system-test/win-test-file b/tests/system-test/win-test-file index 6baa82ac46..96ffa63707 100644 --- a/tests/system-test/win-test-file +++ b/tests/system-test/win-test-file @@ -218,6 +218,7 @@ python3 ./test.py -f 1-insert/delete_stable.py python3 ./test.py -f 1-insert/delete_childtable.py python3 ./test.py -f 1-insert/delete_normaltable.py python3 ./test.py -f 1-insert/keep_expired.py +python3 ./test.py -f 1-insert/stmt_error.py python3 ./test.py -f 1-insert/drop.py python3 ./test.py -f 1-insert/drop.py -N 3 -M 3 -i False -n 3 python3 ./test.py -f 2-query/join2.py diff --git a/tools/shell/src/shellEngine.c b/tools/shell/src/shellEngine.c index 0f91bdeeda..e2752f7899 100644 --- a/tools/shell/src/shellEngine.c +++ b/tools/shell/src/shellEngine.c @@ -716,7 +716,7 @@ int32_t shellCalcColWidth(TAOS_FIELD *field, int32_t precision) { case TSDB_DATA_TYPE_NCHAR: case TSDB_DATA_TYPE_JSON: { - int16_t bytes = field->bytes * TSDB_NCHAR_SIZE; + uint16_t bytes = field->bytes * TSDB_NCHAR_SIZE; if (bytes > shell.args.displayWidth) { return TMAX(shell.args.displayWidth, width); } else { @@ -1118,6 +1118,7 @@ int32_t shellExecute() { } if (shell.conn == NULL) { + printf("failed to connect to server, reason: %s\n", taos_errstr(NULL)); fflush(stdout); return -1; }