diff --git a/docs/en/25-application/03-immigrate.md b/docs/en/25-application/03-immigrate.md index 4d47aec1d7..fe67f97389 100644 --- a/docs/en/25-application/03-immigrate.md +++ b/docs/en/25-application/03-immigrate.md @@ -379,11 +379,11 @@ We still use the hypothetical environment from Chapter 4. There are three measur ### Storage resource estimation -Assuming that the number of sensor devices that generate data and need to be stored is `n`, the frequency of data generation is `t` per second, and the length of each record is `L` bytes, the scale of data generated per day is `n * t * L` bytes. Assuming the compression ratio is `C`, the daily data size is `(n * t * L)/C` bytes. The storage resources are estimated to accommodate the data scale for 1.5 years. In the production environment, the compression ratio C of TDengine is generally between 5 and 7. +Assuming that the number of sensor devices that generate data and need to be stored is `n`, the frequency of data generation is `t` per second, and the length of each record is `L` bytes, the scale of data generated per day is `86400 * n * t * L` bytes. Assuming the compression ratio is `C`, the daily data size is `(86400 * n * t * L)/C` bytes. The storage resources are estimated to accommodate the data scale for 1.5 years. In the production environment, the compression ratio C of TDengine is generally between 5 and 7. With additional 20% redundancy, you can calculate the required storage resources: ```matlab -(n * t * L) * (365 * 1.5) * (1+20%)/C +(86400 * n * t * L) * (365 * 1.5) * (1+20%)/C ```` Substituting in the above formula, the raw data generated every year is 11.8TB without considering the label information. Note that tag information is associated with each timeline in TDengine, not every record. The amount of data to be recorded is somewhat reduced relative to the generated data, and label data can be ignored as a whole. Assuming a compression ratio of 5, the size of the retained data ends up being 2.56 TB. diff --git a/docs/zh/25-application/03-immigrate.md b/docs/zh/25-application/03-immigrate.md index 9d8946bc4a..d1c9caea09 100644 --- a/docs/zh/25-application/03-immigrate.md +++ b/docs/zh/25-application/03-immigrate.md @@ -367,10 +367,10 @@ WHERE ts>=1510560000 AND ts<=1515000009 ### 存储资源估算 -假设产生数据并需要存储的传感器设备数量为 `n`,数据生成的频率为`t`条/秒,每条记录的长度为 `L` bytes,则每天产生的数据规模为 `n×t×L` bytes。假设压缩比为 C,则每日产生数据规模为 `(n×t×L)/C` bytes。存储资源预估为能够容纳 1.5 年的数据规模,生产环境下 TDengine 的压缩比 C 一般在 5 ~ 7 之间,同时为最后结果增加 20% 的冗余,可计算得到需要存储资源: +假设产生数据并需要存储的传感器设备数量为 `n`,数据生成的频率为`t`条/秒,每条记录的长度为 `L` bytes,则每天产生的数据规模为 `86400×n×t×L` bytes。假设压缩比为 C,则每日产生数据规模为 `(86400×n×t×L)/C` bytes。存储资源预估为能够容纳 1.5 年的数据规模,生产环境下 TDengine 的压缩比 C 一般在 5 ~ 7 之间,同时为最后结果增加 20% 的冗余,可计算得到需要存储资源: ```matlab -(n×t×L)×(365×1.5)×(1+20%)/C +(86400×n×t×L)×(365×1.5)×(1+20%)/C ``` 结合以上的计算公式,将参数带入计算公式,在不考虑标签信息的情况下,每年产生的原始数据规模是 11.8TB。需要注意的是,由于标签信息在 TDengine 中关联到每个时间线,并不是每条记录。所以需要记录的数据量规模相对于产生的数据有一定的降低,而这部分标签数据整体上可以忽略不记。假设压缩比为 5,则保留的数据规模最终为 2.56 TB。 diff --git a/include/common/tcommon.h b/include/common/tcommon.h index fd4ed6b180..779453f714 100644 --- a/include/common/tcommon.h +++ b/include/common/tcommon.h @@ -88,8 +88,6 @@ typedef struct { #pragma pack(push, 1) typedef struct SColumnDataAgg { int16_t colId; - int16_t minIndex; - int16_t maxIndex; int16_t numOfNull; int64_t sum; int64_t max; @@ -117,6 +115,20 @@ typedef struct SSDataBlock { SDataBlockInfo info; } SSDataBlock; +enum { + FETCH_TYPE__DATA = 1, + FETCH_TYPE__META, + FETCH_TYPE__NONE, +}; + +typedef struct { + int8_t fetchType; + union { + SSDataBlock data; + void* meta; + }; +} SFetchRet; + typedef struct SVarColAttr { int32_t* offset; // start position for each entry in the list uint32_t length; // used buffer size that contain the valid data diff --git a/include/libs/executor/executor.h b/include/libs/executor/executor.h index 7b4565a99f..0094afef12 100644 --- a/include/libs/executor/executor.h +++ b/include/libs/executor/executor.h @@ -30,7 +30,7 @@ struct SRpcMsg; struct SSubplan; typedef struct SReadHandle { - void* streamReader; + void* tqReader; void* meta; void* config; void* vnode; @@ -38,7 +38,7 @@ typedef struct SReadHandle { SMsgCb* pMsgCb; bool initMetaReader; bool initTableReader; - bool initStreamReader; + bool initTqReader; } SReadHandle; typedef enum { @@ -52,7 +52,7 @@ typedef enum { * @param streamReadHandle * @return */ -qTaskInfo_t qCreateStreamExecTaskInfo(void* msg, void* streamReadHandle); +qTaskInfo_t qCreateStreamExecTaskInfo(void* msg, SReadHandle* readers); /** * Switch the stream scan to snapshot mode @@ -176,6 +176,9 @@ int32_t qGetStreamScanStatus(qTaskInfo_t tinfo, uint64_t* uid, int64_t* ts); int32_t qStreamPrepareScan(qTaskInfo_t tinfo, uint64_t uid, int64_t ts); +void* qExtractReaderFromStreamScanner(void* scanner); +int32_t qExtractStreamScanner(qTaskInfo_t tinfo, void** scanner); + #ifdef __cplusplus } #endif diff --git a/include/libs/stream/tstream.h b/include/libs/stream/tstream.h index 52f671e176..d6cb2c27b0 100644 --- a/include/libs/stream/tstream.h +++ b/include/libs/stream/tstream.h @@ -235,11 +235,6 @@ typedef struct SStreamTask { int8_t taskStatus; int8_t execStatus; - // exec info - int64_t enqueueVer; - int64_t processedVer; - int64_t checkpointVer; - // node info int32_t selfChildId; int32_t nodeId; diff --git a/include/libs/sync/sync.h b/include/libs/sync/sync.h index 8a6d6b7722..bef26cb310 100644 --- a/include/libs/sync/sync.h +++ b/include/libs/sync/sync.h @@ -163,9 +163,6 @@ typedef struct SSyncLogStore { // return commit index of log SyncIndex (*getCommitIndex)(struct SSyncLogStore* pLogStore); - // refactor, log[0 .. n] ==> log[m .. n] - // int32_t (*syncLogSetBeginIndex)(struct SSyncLogStore* pLogStore, SyncIndex beginIndex); - SyncIndex (*syncLogBeginIndex)(struct SSyncLogStore* pLogStore); SyncIndex (*syncLogEndIndex)(struct SSyncLogStore* pLogStore); bool (*syncLogIsEmpty)(struct SSyncLogStore* pLogStore); diff --git a/source/client/src/TMQConnector.c b/source/client/src/TMQConnector.c index c3d1e60782..1d84dcf7a2 100644 --- a/source/client/src/TMQConnector.c +++ b/source/client/src/TMQConnector.c @@ -324,9 +324,7 @@ JNIEXPORT jint JNICALL Java_com_taosdata_jdbc_tmq_TMQConnector_fetchRawBlockImp( (*env)->CallVoidMethod(env, rowobj, g_blockdataSetNumOfRowsFp, (jint)numOfRows); (*env)->CallVoidMethod(env, rowobj, g_blockdataSetNumOfColsFp, (jint)numOfFields); - char *chars = (char *)data; - int32_t len = chars[0] + (chars[1] << 8) + (chars[2] << 16) + (chars[3] << 24); - (*env)->CallVoidMethod(env, rowobj, g_blockdataSetByteArrayFp, len, jniFromNCharToByteArray(env, (char *)data, len)); - + int32_t len = *(int32_t *)data; + (*env)->CallVoidMethod(env, rowobj, g_blockdataSetByteArrayFp, jniFromNCharToByteArray(env, (char *)data, len)); return JNI_SUCCESS; } diff --git a/source/client/src/TSDBJNIConnector.c b/source/client/src/TSDBJNIConnector.c index 246785b570..227c2fff18 100644 --- a/source/client/src/TSDBJNIConnector.c +++ b/source/client/src/TSDBJNIConnector.c @@ -592,8 +592,7 @@ JNIEXPORT jint JNICALL Java_com_taosdata_jdbc_TSDBJNIConnector_fetchBlockImp(JNI (*env)->CallVoidMethod(env, rowobj, g_blockdataSetNumOfRowsFp, (jint)numOfRows); (*env)->CallVoidMethod(env, rowobj, g_blockdataSetNumOfColsFp, (jint)numOfFields); - char *chars = (char *)data; - int32_t len = chars[0] + (chars[1] << 8) + (chars[2] << 16) + (chars[3] << 24); + int32_t len = *(int32_t *)data; (*env)->CallVoidMethod(env, rowobj, g_blockdataSetByteArrayFp, jniFromNCharToByteArray(env, (char *)data, len)); return JNI_SUCCESS; diff --git a/source/client/test/clientTests.cpp b/source/client/test/clientTests.cpp index e8e3237b67..7927c1e008 100644 --- a/source/client/test/clientTests.cpp +++ b/source/client/test/clientTests.cpp @@ -750,7 +750,6 @@ TEST(testCase, projection_query_stables) { taos_close(pConn); } - TEST(testCase, agg_query_tables) { TAOS* pConn = taos_connect("localhost", "root", "taosdata", NULL, 0); ASSERT_NE(pConn, nullptr); @@ -763,7 +762,7 @@ TEST(testCase, agg_query_tables) { } taos_free_result(pRes); - pRes = taos_query(pConn, "show table distributed st1"); + pRes = taos_query(pConn, "show table distributed tup"); if (taos_errno(pRes) != 0) { printf("failed to select from table, reason:%s\n", taos_errstr(pRes)); taos_free_result(pRes); @@ -822,13 +821,29 @@ TEST(testCase, async_api_test) { } #endif + TEST(testCase, update_test) { TAOS* pConn = taos_connect("localhost", "root", "taosdata", NULL, 0); ASSERT_NE(pConn, nullptr); - taos_query(pConn, "use abc1"); + TAOS_RES* pRes = taos_query(pConn, "create database if not exists abc1"); + if (taos_errno(pRes) != TSDB_CODE_SUCCESS) { + printf("failed to create database, code:%s", taos_errstr(pRes)); + taos_free_result(pRes); + return; + } - TAOS_RES* pRes = taos_query(pConn, "create table tup (ts timestamp, k int);"); + taos_free_result(pRes); + + pRes = taos_query(pConn, "use abc1"); + if (taos_errno(pRes) != TSDB_CODE_SUCCESS) { + printf("failed to use db, code:%s", taos_errstr(pRes)); + taos_free_result(pRes); + return; + } + taos_free_result(pRes); + + pRes = taos_query(pConn, "create table tup (ts timestamp, k int);"); if (taos_errno(pRes) != 0) { printf("failed to create table, reason:%s", taos_errstr(pRes)); } @@ -836,11 +851,10 @@ TEST(testCase, update_test) { taos_free_result(pRes); char s[256] = {0}; - for(int32_t i = 0; i < 7000; ++i) { - sprintf(s, "insert into tup values('2020-1-1 1:1:1', %d)", i); + for(int32_t i = 0; i < 17000; ++i) { + sprintf(s, "insert into tup values(now+%da, %d)", i, i); pRes = taos_query(pConn, s); taos_free_result(pRes); } } - #pragma GCC diagnostic pop diff --git a/source/common/src/tdata.c b/source/common/src/tdata.c deleted file mode 100644 index 6dea4a4e57..0000000000 --- a/source/common/src/tdata.c +++ /dev/null @@ -1,14 +0,0 @@ -/* - * 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 . - */ \ No newline at end of file diff --git a/source/common/src/tmsgtype.c b/source/common/src/tmsgtype.c deleted file mode 100644 index 3ca0f00fc2..0000000000 --- a/source/common/src/tmsgtype.c +++ /dev/null @@ -1,18 +0,0 @@ -/* - * 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 . - */ - -#define _DEFAULT_SOURCE -#define TSDB_SQL_C -#include "tmsgtype.h" diff --git a/source/dnode/mgmt/mgmt_mnode/src/mmHandle.c b/source/dnode/mgmt/mgmt_mnode/src/mmHandle.c index d60b69daba..4e3449feb7 100644 --- a/source/dnode/mgmt/mgmt_mnode/src/mmHandle.c +++ b/source/dnode/mgmt/mgmt_mnode/src/mmHandle.c @@ -238,10 +238,12 @@ SArray *mmGetMsgHandles() { if (dmSetMgmtHandle(pArray, TDMT_SYNC_PING, mmPutMsgToSyncQueue, 1) == NULL) goto _OVER; if (dmSetMgmtHandle(pArray, TDMT_SYNC_PING_REPLY, mmPutMsgToSyncQueue, 1) == NULL) goto _OVER; if (dmSetMgmtHandle(pArray, TDMT_SYNC_CLIENT_REQUEST, mmPutMsgToSyncQueue, 1) == NULL) goto _OVER; + if (dmSetMgmtHandle(pArray, TDMT_SYNC_CLIENT_REQUEST_BATCH, mmPutMsgToSyncQueue, 1) == NULL) goto _OVER; if (dmSetMgmtHandle(pArray, TDMT_SYNC_CLIENT_REQUEST_REPLY, mmPutMsgToSyncQueue, 1) == NULL) goto _OVER; if (dmSetMgmtHandle(pArray, TDMT_SYNC_REQUEST_VOTE, mmPutMsgToSyncQueue, 1) == NULL) goto _OVER; if (dmSetMgmtHandle(pArray, TDMT_SYNC_REQUEST_VOTE_REPLY, mmPutMsgToSyncQueue, 1) == NULL) goto _OVER; if (dmSetMgmtHandle(pArray, TDMT_SYNC_APPEND_ENTRIES, mmPutMsgToSyncQueue, 1) == NULL) goto _OVER; + if (dmSetMgmtHandle(pArray, TDMT_SYNC_APPEND_ENTRIES_BATCH, mmPutMsgToSyncQueue, 1) == NULL) goto _OVER; if (dmSetMgmtHandle(pArray, TDMT_SYNC_APPEND_ENTRIES_REPLY, mmPutMsgToSyncQueue, 1) == NULL) goto _OVER; if (dmSetMgmtHandle(pArray, TDMT_SYNC_SNAPSHOT_SEND, mmPutMsgToSyncQueue, 1) == NULL) goto _OVER; if (dmSetMgmtHandle(pArray, TDMT_SYNC_SNAPSHOT_RSP, mmPutMsgToSyncQueue, 1) == NULL) goto _OVER; diff --git a/source/dnode/mgmt/mgmt_vnode/src/vmHandle.c b/source/dnode/mgmt/mgmt_vnode/src/vmHandle.c index a3df32a08c..993a604b51 100644 --- a/source/dnode/mgmt/mgmt_vnode/src/vmHandle.c +++ b/source/dnode/mgmt/mgmt_vnode/src/vmHandle.c @@ -377,10 +377,12 @@ SArray *vmGetMsgHandles() { if (dmSetMgmtHandle(pArray, TDMT_SYNC_PING, vmPutMsgToSyncQueue, 0) == NULL) goto _OVER; if (dmSetMgmtHandle(pArray, TDMT_SYNC_PING_REPLY, vmPutMsgToSyncQueue, 0) == NULL) goto _OVER; if (dmSetMgmtHandle(pArray, TDMT_SYNC_CLIENT_REQUEST, vmPutMsgToSyncQueue, 0) == NULL) goto _OVER; + if (dmSetMgmtHandle(pArray, TDMT_SYNC_CLIENT_REQUEST_BATCH, vmPutMsgToSyncQueue, 0) == NULL) goto _OVER; if (dmSetMgmtHandle(pArray, TDMT_SYNC_CLIENT_REQUEST_REPLY, vmPutMsgToSyncQueue, 0) == NULL) goto _OVER; if (dmSetMgmtHandle(pArray, TDMT_SYNC_REQUEST_VOTE, vmPutMsgToSyncQueue, 0) == NULL) goto _OVER; if (dmSetMgmtHandle(pArray, TDMT_SYNC_REQUEST_VOTE_REPLY, vmPutMsgToSyncQueue, 0) == NULL) goto _OVER; if (dmSetMgmtHandle(pArray, TDMT_SYNC_APPEND_ENTRIES, vmPutMsgToSyncQueue, 0) == NULL) goto _OVER; + if (dmSetMgmtHandle(pArray, TDMT_SYNC_APPEND_ENTRIES_BATCH, vmPutMsgToSyncQueue, 0) == NULL) goto _OVER; if (dmSetMgmtHandle(pArray, TDMT_SYNC_APPEND_ENTRIES_REPLY, vmPutMsgToSyncQueue, 0) == NULL) goto _OVER; if (dmSetMgmtHandle(pArray, TDMT_SYNC_SET_VNODE_STANDBY, vmPutMsgToSyncQueue, 0) == NULL) goto _OVER; diff --git a/source/dnode/vnode/CMakeLists.txt b/source/dnode/vnode/CMakeLists.txt index 24c4f2912c..897e4f26e2 100644 --- a/source/dnode/vnode/CMakeLists.txt +++ b/source/dnode/vnode/CMakeLists.txt @@ -9,7 +9,6 @@ target_sources( "src/vnd/vnodeCfg.c" "src/vnd/vnodeCommit.c" "src/vnd/vnodeQuery.c" - "src/vnd/vnodeStateMgr.c" "src/vnd/vnodeModule.c" "src/vnd/vnodeSvr.c" "src/vnd/vnodeSync.c" @@ -32,7 +31,6 @@ target_sources( "src/sma/smaUtil.c" "src/sma/smaOpen.c" "src/sma/smaCommit.c" - "src/sma/smaSnapshot.c" "src/sma/smaRollup.c" "src/sma/smaTimeRange.c" @@ -43,7 +41,6 @@ target_sources( "src/tsdb/tsdbOpen.c" "src/tsdb/tsdbMemTable.c" "src/tsdb/tsdbRead.c" - "src/tsdb/tsdbReadImpl.c" "src/tsdb/tsdbCache.c" "src/tsdb/tsdbWrite.c" "src/tsdb/tsdbReaderWriter.c" diff --git a/source/dnode/vnode/inc/vnode.h b/source/dnode/vnode/inc/vnode.h index 722087ea45..11a20285c0 100644 --- a/source/dnode/vnode/inc/vnode.h +++ b/source/dnode/vnode/inc/vnode.h @@ -64,8 +64,8 @@ int32_t vnodeSnapshotRead(SVSnapshotReader *pReader, const void **ppData, uint32 int32_t vnodeProcessCreateTSma(SVnode *pVnode, void *pCont, uint32_t contLen); int32_t vnodeGetAllTableList(SVnode *pVnode, uint64_t uid, SArray *list); int32_t vnodeGetCtbIdList(SVnode *pVnode, int64_t suid, SArray *list); -void * vnodeGetIdx(SVnode *pVnode); -void * vnodeGetIvtIdx(SVnode *pVnode); +void *vnodeGetIdx(SVnode *pVnode); +void *vnodeGetIvtIdx(SVnode *pVnode); int32_t vnodeGetLoad(SVnode *pVnode, SVnodeLoad *pLoad); int32_t vnodeValidateTableHash(SVnode *pVnode, char *tableFName); @@ -95,7 +95,7 @@ typedef struct SMetaFltParam { tb_uid_t suid; int16_t cid; int16_t type; - char * val; + char *val; bool reverse; int (*filterFunc)(void *a, void *b, int16_t type); @@ -131,13 +131,13 @@ int32_t tsdbReaderOpen(SVnode *pVnode, SQueryTableDataCond *pCond, SArray *pTabl void tsdbReaderClose(STsdbReader *pReader); bool tsdbNextDataBlock(STsdbReader *pReader); void tsdbRetrieveDataBlockInfo(STsdbReader *pReader, SDataBlockInfo *pDataBlockInfo); -int32_t tsdbRetrieveDataBlockStatisInfo(STsdbReader *pReader, SColumnDataAgg ***pBlockStatis, bool *allHave); +int32_t tsdbRetrieveDatablockSMA(STsdbReader *pReader, SColumnDataAgg ***pBlockStatis, bool *allHave); SArray *tsdbRetrieveDataBlock(STsdbReader *pTsdbReadHandle, SArray *pColumnIdList); int32_t tsdbReaderReset(STsdbReader *pReader, SQueryTableDataCond *pCond, int32_t tWinIdx); int32_t tsdbGetFileBlocksDistInfo(STsdbReader *pReader, STableBlockDistInfo *pTableBlockInfo); int64_t tsdbGetNumOfRowsInMemTable(STsdbReader *pHandle); -void * tsdbGetIdx(SMeta *pMeta); -void * tsdbGetIvtIdx(SMeta *pMeta); +void *tsdbGetIdx(SMeta *pMeta); +void *tsdbGetIvtIdx(SMeta *pMeta); int32_t tsdbLastRowReaderOpen(void *pVnode, int32_t type, SArray *pTableIdList, int32_t *colId, int32_t numOfCols, void **pReader); @@ -146,19 +146,37 @@ int32_t tsdbLastrowReaderClose(void *pReader); // tq -typedef struct STqReadHandle SStreamReader; +typedef struct STqReader { + int64_t ver; + const SSubmitReq *pMsg; + SSubmitBlk *pBlock; + SSubmitMsgIter msgIter; + SSubmitBlkIter blkIter; -SStreamReader *tqInitSubmitMsgScanner(SMeta *pMeta); + SWalReader *pWalReader; -void tqReadHandleSetColIdList(SStreamReader *pReadHandle, SArray *pColIdList); -int32_t tqReadHandleSetTbUidList(SStreamReader *pHandle, const SArray *tbUidList); -int32_t tqReadHandleAddTbUidList(SStreamReader *pHandle, const SArray *tbUidList); -int32_t tqReadHandleRemoveTbUidList(SStreamReader *pHandle, const SArray *tbUidList); + SMeta *pVnodeMeta; + SHashObj *tbIdHash; + SArray *pColIdList; // SArray -int32_t tqReadHandleSetMsg(SStreamReader *pHandle, SSubmitReq *pMsg, int64_t ver); -bool tqNextDataBlock(SStreamReader *pHandle); -bool tqNextDataBlockFilterOut(SStreamReader *pHandle, SHashObj *filterOutUids); -int32_t tqRetrieveDataBlock(SSDataBlock *pBlock, SStreamReader *pHandle); + int32_t cachedSchemaVer; + int64_t cachedSchemaSuid; + SSchemaWrapper *pSchemaWrapper; + STSchema *pSchema; +} STqReader; + +STqReader *tqOpenReader(SVnode *pVnode); +void tqCloseReader(STqReader *); + +void tqReaderSetColIdList(STqReader *pReader, SArray *pColIdList); +int32_t tqReaderSetTbUidList(STqReader *pReader, const SArray *tbUidList); +int32_t tqReaderAddTbUidList(STqReader *pReader, const SArray *tbUidList); +int32_t tqReaderRemoveTbUidList(STqReader *pReader, const SArray *tbUidList); + +int32_t tqReaderSetDataMsg(STqReader *pReader, SSubmitReq *pMsg, int64_t ver); +bool tqNextDataBlock(STqReader *pReader); +bool tqNextDataBlockFilterOut(STqReader *pReader, SHashObj *filterOutUids); +int32_t tqRetrieveDataBlock(SSDataBlock *pBlock, STqReader *pReader); // sma int32_t smaGetTSmaDays(SVnodeCfg *pCfg, void *pCont, uint32_t contLen, int32_t *days); @@ -214,7 +232,7 @@ struct SMetaEntry { int8_t type; int8_t flags; // TODO: need refactor? tb_uid_t uid; - char * name; + char *name; union { struct { SSchemaWrapper schemaRow; @@ -225,7 +243,7 @@ struct SMetaEntry { int64_t ctime; int32_t ttlDays; int32_t commentLen; - char * comment; + char *comment; tb_uid_t suid; uint8_t *pTags; } ctbEntry; @@ -233,7 +251,7 @@ struct SMetaEntry { int64_t ctime; int32_t ttlDays; int32_t commentLen; - char * comment; + char *comment; int32_t ncid; // next column id SSchemaWrapper schemaRow; } ntbEntry; @@ -247,17 +265,17 @@ struct SMetaEntry { struct SMetaReader { int32_t flags; - SMeta * pMeta; + SMeta *pMeta; SDecoder coder; SMetaEntry me; - void * pBuf; + void *pBuf; int32_t szBuf; }; struct SMTbCursor { - TBC * pDbc; - void * pKey; - void * pVal; + TBC *pDbc; + void *pKey; + void *pVal; int32_t kLen; int32_t vLen; SMetaReader mr; diff --git a/source/dnode/vnode/src/inc/tq.h b/source/dnode/vnode/src/inc/tq.h index 12fb500ba3..9c3bd85c71 100644 --- a/source/dnode/vnode/src/inc/tq.h +++ b/source/dnode/vnode/src/inc/tq.h @@ -44,25 +44,6 @@ extern "C" { typedef struct STqOffsetStore STqOffsetStore; -// tqRead - -struct STqReadHandle { - int64_t ver; - const SSubmitReq* pMsg; - SSubmitBlk* pBlock; - SSubmitMsgIter msgIter; - SSubmitBlkIter blkIter; - - SMeta* pVnodeMeta; - SHashObj* tbIdHash; - SArray* pColIdList; // SArray - - int32_t cachedSchemaVer; - int64_t cachedSchemaSuid; - SSchemaWrapper* pSchemaWrapper; - STSchema* pSchema; -}; - // tqPush typedef struct { @@ -102,7 +83,7 @@ typedef struct { typedef struct { int8_t subType; - SStreamReader* pExecReader[5]; + STqReader* pExecReader[5]; union { STqExecCol execCol; STqExecTb execTb; @@ -118,7 +99,7 @@ typedef struct { int32_t epoch; int8_t fetchMeta; - // reader + // TODO remove SWalReader* pWalReader; // push diff --git a/source/dnode/vnode/src/inc/tsdb.h b/source/dnode/vnode/src/inc/tsdb.h index 0f65a536e0..cd2dfd3351 100644 --- a/source/dnode/vnode/src/inc/tsdb.h +++ b/source/dnode/vnode/src/inc/tsdb.h @@ -245,6 +245,8 @@ int32_t tsdbCacheRelease(SLRUCache *pCache, LRUHandle *h); int32_t tsdbCacheDelete(SLRUCache *pCache, tb_uid_t uid, TSKEY eKey); +int32_t tsdbCacheLastArray2Row(SArray *pLastArray, STSRow **ppRow, STSchema *pSchema); + // structs ======================= typedef struct { int minFid; diff --git a/source/dnode/vnode/src/meta/metaTable.c b/source/dnode/vnode/src/meta/metaTable.c index 74076d323c..341173103c 100644 --- a/source/dnode/vnode/src/meta/metaTable.c +++ b/source/dnode/vnode/src/meta/metaTable.c @@ -16,6 +16,7 @@ #include "meta.h" static int metaSaveJsonVarToIdx(SMeta *pMeta, const SMetaEntry *pCtbEntry, const SSchema *pSchema); +static int metaDelJsonVarFromIdx(SMeta *pMeta, const SMetaEntry *pCtbEntry, const SSchema *pSchema); static int metaHandleEntry(SMeta *pMeta, const SMetaEntry *pME); static int metaSaveToTbDb(SMeta *pMeta, const SMetaEntry *pME); static int metaUpdateUidIdx(SMeta *pMeta, const SMetaEntry *pME); @@ -92,7 +93,65 @@ static int metaSaveJsonVarToIdx(SMeta *pMeta, const SMetaEntry *pCtbEntry, const } else if (type == TSDB_DATA_TYPE_BOOL) { int val = *(int *)(&pTagVal->i64); int len = sizeof(val); - term = indexTermCreate(suid, ADD_VALUE, TSDB_DATA_TYPE_INT, key, nKey, (const char *)&val, len); + term = indexTermCreate(suid, ADD_VALUE, TSDB_DATA_TYPE_BOOL, key, nKey, (const char *)&val, len); + } + if (term != NULL) { + indexMultiTermAdd(terms, term); + } + } + indexJsonPut(pMeta->pTagIvtIdx, terms, tuid); + indexMultiTermDestroy(terms); +#endif + return 0; +} +int metaDelJsonVarFromIdx(SMeta *pMeta, const SMetaEntry *pCtbEntry, const SSchema *pSchema) { +#ifdef USE_INVERTED_INDEX + if (pMeta->pTagIvtIdx == NULL || pCtbEntry == NULL) { + return -1; + } + void * data = pCtbEntry->ctbEntry.pTags; + const char *tagName = pSchema->name; + + tb_uid_t suid = pCtbEntry->ctbEntry.suid; + tb_uid_t tuid = pCtbEntry->uid; + const void *pTagData = pCtbEntry->ctbEntry.pTags; + int32_t nTagData = 0; + + SArray *pTagVals = NULL; + if (tTagToValArray((const STag *)data, &pTagVals) != 0) { + return -1; + } + + SIndexMultiTerm *terms = indexMultiTermCreate(); + int16_t nCols = taosArrayGetSize(pTagVals); + for (int i = 0; i < nCols; i++) { + STagVal *pTagVal = (STagVal *)taosArrayGet(pTagVals, i); + char type = pTagVal->type; + + char * key = pTagVal->pKey; + int32_t nKey = strlen(key); + + SIndexTerm *term = NULL; + if (type == TSDB_DATA_TYPE_NULL) { + term = indexTermCreate(suid, DEL_VALUE, TSDB_DATA_TYPE_VARCHAR, key, nKey, NULL, 0); + } else if (type == TSDB_DATA_TYPE_NCHAR) { + if (pTagVal->nData > 0) { + char * val = taosMemoryCalloc(1, pTagVal->nData + VARSTR_HEADER_SIZE); + int32_t len = taosUcs4ToMbs((TdUcs4 *)pTagVal->pData, pTagVal->nData, val + VARSTR_HEADER_SIZE); + memcpy(val, (uint16_t *)&len, VARSTR_HEADER_SIZE); + type = TSDB_DATA_TYPE_VARCHAR; + term = indexTermCreate(suid, DEL_VALUE, type, key, nKey, val, len); + } else if (pTagVal->nData == 0) { + term = indexTermCreate(suid, DEL_VALUE, TSDB_DATA_TYPE_VARCHAR, key, nKey, pTagVal->pData, 0); + } + } else if (type == TSDB_DATA_TYPE_DOUBLE) { + double val = *(double *)(&pTagVal->i64); + int len = sizeof(val); + term = indexTermCreate(suid, DEL_VALUE, type, key, nKey, (const char *)&val, len); + } else if (type == TSDB_DATA_TYPE_BOOL) { + int val = *(int *)(&pTagVal->i64); + int len = sizeof(val); + term = indexTermCreate(suid, DEL_VALUE, TSDB_DATA_TYPE_BOOL, key, nKey, (const char *)&val, len); } if (term != NULL) { indexMultiTermAdd(terms, term); @@ -434,9 +493,33 @@ static int metaDropTableByUid(SMeta *pMeta, tb_uid_t uid, int *type) { if (type) *type = e.type; + if (e.type == TSDB_CHILD_TABLE) { + void *tData = NULL; + int tLen = 0; + + if (tdbTbGet(pMeta->pUidIdx, &e.ctbEntry.suid, sizeof(tb_uid_t), &tData, &tLen) == 0) { + version = *(int64_t *)tData; + STbDbKey tbDbKey = {.uid = e.ctbEntry.suid, .version = version}; + if (tdbTbGet(pMeta->pTbDb, &tbDbKey, sizeof(tbDbKey), &tData, &tLen) == 0) { + SDecoder tdc = {0}; + SMetaEntry stbEntry = {0}; + + tDecoderInit(&tdc, tData, tLen); + metaDecodeEntry(&tdc, &stbEntry); + const SSchema *pTagColumn = &stbEntry.stbEntry.schemaTag.pSchema[0]; + if (pTagColumn->type == TSDB_DATA_TYPE_JSON) { + metaDelJsonVarFromIdx(pMeta, &e, pTagColumn); + } + tDecoderClear(&tdc); + } + tdbFree(tData); + } + } + tdbTbDelete(pMeta->pTbDb, &(STbDbKey){.version = version, .uid = uid}, sizeof(STbDbKey), &pMeta->txn); tdbTbDelete(pMeta->pNameIdx, e.name, strlen(e.name) + 1, &pMeta->txn); tdbTbDelete(pMeta->pUidIdx, &uid, sizeof(uid), &pMeta->txn); + if (e.type != TSDB_SUPER_TABLE) metaDeleteTtlIdx(pMeta, &e); if (e.type == TSDB_CHILD_TABLE) { diff --git a/source/dnode/vnode/src/sma/smaRollup.c b/source/dnode/vnode/src/sma/smaRollup.c index 92168a9b92..4c2e606e0e 100644 --- a/source/dnode/vnode/src/sma/smaRollup.c +++ b/source/dnode/vnode/src/sma/smaRollup.c @@ -277,6 +277,9 @@ static int32_t tdSetRSmaInfoItemParams(SSma *pSma, SRSmaParam *param, SRSmaInfo pItem->maxDelay = TSDB_MAX_ROLLUP_MAX_DELAY; } pItem->level = (idx == 0 ? TSDB_RETENTION_L1 : TSDB_RETENTION_L2); + smaInfo("vgId:%d table:%" PRIi64 " level:%" PRIi8 " maxdelay:%" PRIi64 " watermark:%" PRIi64 + ", finally maxdelay:%" PRIi32, + SMA_VID(pSma), pRSmaInfo->suid, idx + 1, param->maxdelay[idx], param->watermark[idx], pItem->maxDelay); } return TSDB_CODE_SUCCESS; _err: @@ -325,14 +328,14 @@ int32_t tdProcessRSmaCreateImpl(SSma *pSma, SRSmaParam *param, int64_t suid, con return TSDB_CODE_FAILED; } - SStreamReader *pReadHandle = tqInitSubmitMsgScanner(pMeta); - if (!pReadHandle) { + STqReader *pReader = tqOpenReader(pVnode); + if (!pReader) { terrno = TSDB_CODE_OUT_OF_MEMORY; goto _err; } SReadHandle handle = { - .streamReader = pReadHandle, + .tqReader = pReader, .meta = pMeta, .pMsgCb = pMsgCb, .vnode = pVnode, @@ -364,7 +367,7 @@ int32_t tdProcessRSmaCreateImpl(SSma *pSma, SRSmaParam *param, int64_t suid, con return TSDB_CODE_SUCCESS; _err: tdFreeRSmaInfo(pRSmaInfo); - taosMemoryFree(pReadHandle); + taosMemoryFree(pReader); return TSDB_CODE_FAILED; } @@ -570,11 +573,15 @@ static int32_t tdFetchAndSubmitRSmaResult(SRSmaInfoItem *pItem, int8_t blkType) SSubmitReq *pReq = NULL; // TODO: the schema update should be handled if (buildSubmitReqFromDataBlock(&pReq, pResult, pRSmaInfo->pTSchema, SMA_VID(pSma), pRSmaInfo->suid) < 0) { + smaError("vgId:%d, build submit req for rsma table %" PRIi64 "l evel %" PRIi8 " failed since %s", SMA_VID(pSma), + pRSmaInfo->suid, pItem->level, terrstr()); goto _err; } if (pReq && tdProcessSubmitReq(sinkTsdb, atomic_add_fetch_64(&pRSmaInfo->pStat->submitVer, 1), pReq) < 0) { taosMemoryFreeClear(pReq); + smaError("vgId:%d, process submit req for rsma table %" PRIi64 " level %" PRIi8 " failed since %s", SMA_VID(pSma), + pRSmaInfo->suid, pItem->level, terrstr()); goto _err; } diff --git a/source/dnode/vnode/src/sma/smaSnapshot.c b/source/dnode/vnode/src/sma/smaSnapshot.c deleted file mode 100644 index b2c85642b9..0000000000 --- a/source/dnode/vnode/src/sma/smaSnapshot.c +++ /dev/null @@ -1,16 +0,0 @@ -/* - * 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 . - */ - -#include "sma.h" \ No newline at end of file diff --git a/source/dnode/vnode/src/tq/tq.c b/source/dnode/vnode/src/tq/tq.c index 2dd80c5d52..e21b0fe9e8 100644 --- a/source/dnode/vnode/src/tq/tq.c +++ b/source/dnode/vnode/src/tq/tq.c @@ -447,26 +447,38 @@ int32_t tqProcessVgChangeReq(STQ* pTq, char* msg, int32_t msgLen) { pHandle->fetchMeta = req.withMeta; pHandle->pWalReader = walOpenReader(pTq->pVnode->pWal, NULL); - for (int32_t i = 0; i < 5; i++) { - pHandle->execHandle.pExecReader[i] = tqInitSubmitMsgScanner(pTq->pVnode->pMeta); - } + /*for (int32_t i = 0; i < 5; i++) {*/ + /*pHandle->execHandle.pExecReader[i] = tqOpenReader(pTq->pVnode);*/ + /*}*/ if (pHandle->execHandle.subType == TOPIC_SUB_TYPE__COLUMN) { pHandle->execHandle.execCol.qmsg = req.qmsg; req.qmsg = NULL; for (int32_t i = 0; i < 5; i++) { SReadHandle handle = { - .streamReader = pHandle->execHandle.pExecReader[i], + .tqReader = pHandle->execHandle.pExecReader[i], .meta = pTq->pVnode->pMeta, .vnode = pTq->pVnode, .initTableReader = true, + .initTqReader = true, }; pHandle->execHandle.execCol.task[i] = qCreateStreamExecTaskInfo(pHandle->execHandle.execCol.qmsg, &handle); ASSERT(pHandle->execHandle.execCol.task[i]); + void* scanner = NULL; + qExtractStreamScanner(pHandle->execHandle.execCol.task[i], &scanner); + ASSERT(scanner); + pHandle->execHandle.pExecReader[i] = qExtractReaderFromStreamScanner(scanner); + ASSERT(pHandle->execHandle.pExecReader[i]); } } else if (pHandle->execHandle.subType == TOPIC_SUB_TYPE__DB) { + for (int32_t i = 0; i < 5; i++) { + pHandle->execHandle.pExecReader[i] = tqOpenReader(pTq->pVnode); + } pHandle->execHandle.execDb.pFilterOutTbUid = taosHashInit(64, taosGetDefaultHashFunction(TSDB_DATA_TYPE_BIGINT), false, HASH_NO_LOCK); } else if (pHandle->execHandle.subType == TOPIC_SUB_TYPE__TABLE) { + for (int32_t i = 0; i < 5; i++) { + pHandle->execHandle.pExecReader[i] = tqOpenReader(pTq->pVnode); + } pHandle->execHandle.execTb.suid = req.suid; SArray* tbUidList = taosArrayInit(0, sizeof(int64_t)); vnodeGetCtbIdList(pTq->pVnode, req.suid, tbUidList); @@ -476,7 +488,7 @@ int32_t tqProcessVgChangeReq(STQ* pTq, char* msg, int32_t msgLen) { tqDebug("vg %d, idx %d, uid: %ld", TD_VID(pTq->pVnode), i, tbUid); } for (int32_t i = 0; i < 5; i++) { - tqReadHandleSetTbUidList(pHandle->execHandle.pExecReader[i], tbUidList); + tqReaderSetTbUidList(pHandle->execHandle.pExecReader[i], tbUidList); } taosArrayDestroy(tbUidList); } @@ -532,7 +544,7 @@ int32_t tqProcessTaskDeployReq(STQ* pTq, char* msg, int32_t msgLen) { SReadHandle handle = { .meta = pTq->pVnode->pMeta, .vnode = pTq->pVnode, - .initStreamReader = 1, + .initTqReader = 1, }; pTask->exec.executor = qCreateStreamExecTaskInfo(pTask->exec.qmsg, &handle); } else { diff --git a/source/dnode/vnode/src/tq/tqExec.c b/source/dnode/vnode/src/tq/tqExec.c index 9dd2a0258f..16822e6003 100644 --- a/source/dnode/vnode/src/tq/tqExec.c +++ b/source/dnode/vnode/src/tq/tqExec.c @@ -135,8 +135,8 @@ int32_t tqLogScanExec(STQ* pTq, STqExecHandle* pExec, SSubmitReq* pReq, SMqDataR } } else if (pExec->subType == TOPIC_SUB_TYPE__TABLE) { pRsp->withSchema = 1; - SStreamReader* pReader = pExec->pExecReader[workerId]; - tqReadHandleSetMsg(pReader, pReq, 0); + STqReader* pReader = pExec->pExecReader[workerId]; + tqReaderSetDataMsg(pReader, pReq, 0); while (tqNextDataBlock(pReader)) { SSDataBlock block = {0}; if (tqRetrieveDataBlock(&block, pReader) < 0) { @@ -153,8 +153,8 @@ int32_t tqLogScanExec(STQ* pTq, STqExecHandle* pExec, SSubmitReq* pReq, SMqDataR } } else if (pExec->subType == TOPIC_SUB_TYPE__DB) { pRsp->withSchema = 1; - SStreamReader* pReader = pExec->pExecReader[workerId]; - tqReadHandleSetMsg(pReader, pReq, 0); + STqReader* pReader = pExec->pExecReader[workerId]; + tqReaderSetDataMsg(pReader, pReq, 0); while (tqNextDataBlockFilterOut(pReader, pExec->execDb.pFilterOutTbUid)) { SSDataBlock block = {0}; if (tqRetrieveDataBlock(&block, pReader) < 0) { diff --git a/source/dnode/vnode/src/tq/tqMeta.c b/source/dnode/vnode/src/tq/tqMeta.c index c7e9c8eed6..67fa4ed166 100644 --- a/source/dnode/vnode/src/tq/tqMeta.c +++ b/source/dnode/vnode/src/tq/tqMeta.c @@ -79,12 +79,12 @@ int32_t tqMetaOpen(STQ* pTq) { tDecodeSTqHandle(&decoder, &handle); handle.pWalReader = walOpenReader(pTq->pVnode->pWal, NULL); for (int32_t i = 0; i < 5; i++) { - handle.execHandle.pExecReader[i] = tqInitSubmitMsgScanner(pTq->pVnode->pMeta); + handle.execHandle.pExecReader[i] = tqOpenReader(pTq->pVnode); } if (handle.execHandle.subType == TOPIC_SUB_TYPE__COLUMN) { for (int32_t i = 0; i < 5; i++) { SReadHandle reader = { - .streamReader = handle.execHandle.pExecReader[i], + .tqReader = handle.execHandle.pExecReader[i], .meta = pTq->pVnode->pMeta, .pMsgCb = &pTq->pVnode->msgCb, .vnode = pTq->pVnode, diff --git a/source/dnode/vnode/src/tq/tqRead.c b/source/dnode/vnode/src/tq/tqRead.c index bc992b2211..53ef17a6ba 100644 --- a/source/dnode/vnode/src/tq/tqRead.c +++ b/source/dnode/vnode/src/tq/tqRead.c @@ -15,6 +15,11 @@ #include "tq.h" +int64_t tqScanLog(STQ* pTq, const STqExecHandle* pExec, SMqDataRsp* pRsp, STqOffsetVal offset) { + /*if ()*/ + return 0; +} + int64_t tqFetchLog(STQ* pTq, STqHandle* pHandle, int64_t* fetchOffset, SWalCkHead** ppCkHead) { int32_t code = 0; taosThreadMutexLock(&pHandle->pWalReader->mutex); @@ -73,53 +78,107 @@ END: return code; } -SStreamReader* tqInitSubmitMsgScanner(SMeta* pMeta) { - SStreamReader* pReadHandle = taosMemoryMalloc(sizeof(SStreamReader)); - if (pReadHandle == NULL) { +STqReader* tqOpenReader(SVnode* pVnode) { + STqReader* pReader = taosMemoryMalloc(sizeof(STqReader)); + if (pReader == NULL) { return NULL; } - pReadHandle->pVnodeMeta = pMeta; - pReadHandle->pMsg = NULL; - pReadHandle->ver = -1; - pReadHandle->pColIdList = NULL; - pReadHandle->cachedSchemaVer = 0; - pReadHandle->cachedSchemaSuid = 0; - pReadHandle->pSchema = NULL; - pReadHandle->pSchemaWrapper = NULL; - pReadHandle->tbIdHash = NULL; - return pReadHandle; + + // TODO open + /*pReader->pWalReader = walOpenReader(pVnode->pWal, NULL);*/ + + pReader->pVnodeMeta = pVnode->pMeta; + pReader->pMsg = NULL; + pReader->ver = -1; + pReader->pColIdList = NULL; + pReader->cachedSchemaVer = 0; + pReader->cachedSchemaSuid = 0; + pReader->pSchema = NULL; + pReader->pSchemaWrapper = NULL; + pReader->tbIdHash = NULL; + return pReader; } -int32_t tqReadHandleSetMsg(SStreamReader* pReadHandle, SSubmitReq* pMsg, int64_t ver) { - pReadHandle->pMsg = pMsg; +void tqCloseReader(STqReader* pReader) { + // close wal reader + // free cached schema + // free hash + taosMemoryFree(pReader); +} - if (tInitSubmitMsgIter(pMsg, &pReadHandle->msgIter) < 0) return -1; +int32_t tqNextBlock(STqReader* pReader, SFetchRet* ret) { + bool fromProcessedMsg = pReader->pMsg != NULL; + + while (1) { + if (!fromProcessedMsg) { + if (walNextValidMsg(pReader->pWalReader) < 0) { + ret->fetchType = FETCH_TYPE__NONE; + return -1; + } + void* body = pReader->pWalReader->pHead->head.body; + if (pReader->pWalReader->pHead->head.msgType != TDMT_VND_SUBMIT) { + // TODO do filter + ret->fetchType = FETCH_TYPE__META; + ret->meta = pReader->pWalReader->pHead->head.body; + return 0; + } else { + tqReaderSetDataMsg(pReader, body, pReader->pWalReader->pHead->head.version); + } + } + + while (tqNextDataBlock(pReader)) { + memset(&ret->data, 0, sizeof(SSDataBlock)); + int32_t code = tqRetrieveDataBlock(&ret->data, pReader); + if (code != 0 || ret->data.info.rows == 0) { + if (fromProcessedMsg) { + ret->fetchType = FETCH_TYPE__NONE; + return 0; + } else { + break; + } + } + + ret->fetchType = FETCH_TYPE__DATA; + return 0; + } + + if (fromProcessedMsg) { + ret->fetchType = FETCH_TYPE__NONE; + return 0; + } + } +} + +int32_t tqReaderSetDataMsg(STqReader* pReader, SSubmitReq* pMsg, int64_t ver) { + pReader->pMsg = pMsg; + + if (tInitSubmitMsgIter(pMsg, &pReader->msgIter) < 0) return -1; while (true) { - if (tGetSubmitMsgNext(&pReadHandle->msgIter, &pReadHandle->pBlock) < 0) return -1; - if (pReadHandle->pBlock == NULL) break; + if (tGetSubmitMsgNext(&pReader->msgIter, &pReader->pBlock) < 0) return -1; + if (pReader->pBlock == NULL) break; } - if (tInitSubmitMsgIter(pMsg, &pReadHandle->msgIter) < 0) return -1; - pReadHandle->ver = ver; - memset(&pReadHandle->blkIter, 0, sizeof(SSubmitBlkIter)); + if (tInitSubmitMsgIter(pMsg, &pReader->msgIter) < 0) return -1; + pReader->ver = ver; + memset(&pReader->blkIter, 0, sizeof(SSubmitBlkIter)); return 0; } -bool tqNextDataBlock(SStreamReader* pHandle) { - if (pHandle->pMsg == NULL) return false; +bool tqNextDataBlock(STqReader* pReader) { + if (pReader->pMsg == NULL) return false; while (1) { - if (tGetSubmitMsgNext(&pHandle->msgIter, &pHandle->pBlock) < 0) { + if (tGetSubmitMsgNext(&pReader->msgIter, &pReader->pBlock) < 0) { return false; } - if (pHandle->pBlock == NULL) { - pHandle->pMsg = NULL; + if (pReader->pBlock == NULL) { + pReader->pMsg = NULL; return false; } - if (pHandle->tbIdHash == NULL) { + if (pReader->tbIdHash == NULL) { return true; } - void* ret = taosHashGet(pHandle->tbIdHash, &pHandle->msgIter.uid, sizeof(int64_t)); + void* ret = taosHashGet(pReader->tbIdHash, &pReader->msgIter.uid, sizeof(int64_t)); /*tqDebug("search uid %ld", pHandle->msgIter.uid);*/ if (ret != NULL) { /*tqDebug("find uid %ld", pHandle->msgIter.uid);*/ @@ -129,7 +188,7 @@ bool tqNextDataBlock(SStreamReader* pHandle) { return false; } -bool tqNextDataBlockFilterOut(SStreamReader* pHandle, SHashObj* filterOutUids) { +bool tqNextDataBlockFilterOut(STqReader* pHandle, SHashObj* filterOutUids) { while (1) { if (tGetSubmitMsgNext(&pHandle->msgIter, &pHandle->pBlock) < 0) { return false; @@ -145,38 +204,38 @@ bool tqNextDataBlockFilterOut(SStreamReader* pHandle, SHashObj* filterOutUids) { return false; } -int32_t tqRetrieveDataBlock(SSDataBlock* pBlock, SStreamReader* pHandle) { +int32_t tqRetrieveDataBlock(SSDataBlock* pBlock, STqReader* pReader) { // TODO: cache multiple schema - int32_t sversion = htonl(pHandle->pBlock->sversion); - if (pHandle->cachedSchemaSuid == 0 || pHandle->cachedSchemaVer != sversion || - pHandle->cachedSchemaSuid != pHandle->msgIter.suid) { - if (pHandle->pSchema) taosMemoryFree(pHandle->pSchema); - pHandle->pSchema = metaGetTbTSchema(pHandle->pVnodeMeta, pHandle->msgIter.uid, sversion); - if (pHandle->pSchema == NULL) { + int32_t sversion = htonl(pReader->pBlock->sversion); + if (pReader->cachedSchemaSuid == 0 || pReader->cachedSchemaVer != sversion || + pReader->cachedSchemaSuid != pReader->msgIter.suid) { + if (pReader->pSchema) taosMemoryFree(pReader->pSchema); + pReader->pSchema = metaGetTbTSchema(pReader->pVnodeMeta, pReader->msgIter.uid, sversion); + if (pReader->pSchema == NULL) { tqWarn("cannot found tsschema for table: uid: %ld (suid: %ld), version %d, possibly dropped table", - pHandle->msgIter.uid, pHandle->msgIter.suid, pHandle->cachedSchemaVer); + pReader->msgIter.uid, pReader->msgIter.suid, pReader->cachedSchemaVer); /*ASSERT(0);*/ terrno = TSDB_CODE_TQ_TABLE_SCHEMA_NOT_FOUND; return -1; } - if (pHandle->pSchemaWrapper) tDeleteSSchemaWrapper(pHandle->pSchemaWrapper); - pHandle->pSchemaWrapper = metaGetTableSchema(pHandle->pVnodeMeta, pHandle->msgIter.uid, sversion, true); - if (pHandle->pSchemaWrapper == NULL) { + if (pReader->pSchemaWrapper) tDeleteSSchemaWrapper(pReader->pSchemaWrapper); + pReader->pSchemaWrapper = metaGetTableSchema(pReader->pVnodeMeta, pReader->msgIter.uid, sversion, true); + if (pReader->pSchemaWrapper == NULL) { tqWarn("cannot found schema wrapper for table: suid: %ld, version %d, possibly dropped table", - pHandle->msgIter.uid, pHandle->cachedSchemaVer); + pReader->msgIter.uid, pReader->cachedSchemaVer); /*ASSERT(0);*/ terrno = TSDB_CODE_TQ_TABLE_SCHEMA_NOT_FOUND; return -1; } - pHandle->cachedSchemaVer = sversion; - pHandle->cachedSchemaSuid = pHandle->msgIter.suid; + pReader->cachedSchemaVer = sversion; + pReader->cachedSchemaSuid = pReader->msgIter.suid; } - STSchema* pTschema = pHandle->pSchema; - SSchemaWrapper* pSchemaWrapper = pHandle->pSchemaWrapper; + STSchema* pTschema = pReader->pSchema; + SSchemaWrapper* pSchemaWrapper = pReader->pSchemaWrapper; - int32_t colNumNeed = taosArrayGetSize(pHandle->pColIdList); + int32_t colNumNeed = taosArrayGetSize(pReader->pColIdList); if (colNumNeed == 0) { int32_t colMeta = 0; @@ -199,7 +258,7 @@ int32_t tqRetrieveDataBlock(SSDataBlock* pBlock, SStreamReader* pHandle) { while (colMeta < pSchemaWrapper->nCols && colNeed < colNumNeed) { SSchema* pColSchema = &pSchemaWrapper->pSchema[colMeta]; col_id_t colIdSchema = pColSchema->colId; - col_id_t colIdNeed = *(col_id_t*)taosArrayGet(pHandle->pColIdList, colNeed); + col_id_t colIdNeed = *(col_id_t*)taosArrayGet(pReader->pColIdList, colNeed); if (colIdSchema < colIdNeed) { colMeta++; } else if (colIdSchema > colIdNeed) { @@ -216,7 +275,7 @@ int32_t tqRetrieveDataBlock(SSDataBlock* pBlock, SStreamReader* pHandle) { } } - if (blockDataEnsureCapacity(pBlock, pHandle->msgIter.numOfRows) < 0) { + if (blockDataEnsureCapacity(pBlock, pReader->msgIter.numOfRows) < 0) { goto FAIL; } @@ -227,13 +286,12 @@ int32_t tqRetrieveDataBlock(SSDataBlock* pBlock, SStreamReader* pHandle) { STSRow* row; int32_t curRow = 0; - tInitSubmitBlkIter(&pHandle->msgIter, pHandle->pBlock, &pHandle->blkIter); + tInitSubmitBlkIter(&pReader->msgIter, pReader->pBlock, &pReader->blkIter); - pBlock->info.groupId = 0; - pBlock->info.uid = pHandle->msgIter.uid; - pBlock->info.rows = pHandle->msgIter.numOfRows; + pBlock->info.uid = pReader->msgIter.uid; + pBlock->info.rows = pReader->msgIter.numOfRows; - while ((row = tGetSubmitBlkNext(&pHandle->blkIter)) != NULL) { + while ((row = tGetSubmitBlkNext(&pReader->blkIter)) != NULL) { tdSTSRowIterReset(&iter, row); // get all wanted col of that block for (int32_t i = 0; i < colActual; i++) { @@ -255,9 +313,9 @@ FAIL: return -1; } -void tqReadHandleSetColIdList(SStreamReader* pReadHandle, SArray* pColIdList) { pReadHandle->pColIdList = pColIdList; } +void tqReaderSetColIdList(STqReader* pReadHandle, SArray* pColIdList) { pReadHandle->pColIdList = pColIdList; } -int tqReadHandleSetTbUidList(SStreamReader* pHandle, const SArray* tbUidList) { +int tqReaderSetTbUidList(STqReader* pHandle, const SArray* tbUidList) { if (pHandle->tbIdHash) { taosHashClear(pHandle->tbIdHash); } @@ -276,7 +334,7 @@ int tqReadHandleSetTbUidList(SStreamReader* pHandle, const SArray* tbUidList) { return 0; } -int tqReadHandleAddTbUidList(SStreamReader* pHandle, const SArray* tbUidList) { +int tqReaderAddTbUidList(STqReader* pHandle, const SArray* tbUidList) { if (pHandle->tbIdHash == NULL) { pHandle->tbIdHash = taosHashInit(64, taosGetDefaultHashFunction(TSDB_DATA_TYPE_BIGINT), true, HASH_NO_LOCK); if (pHandle->tbIdHash == NULL) { @@ -293,7 +351,7 @@ int tqReadHandleAddTbUidList(SStreamReader* pHandle, const SArray* tbUidList) { return 0; } -int tqReadHandleRemoveTbUidList(SStreamReader* pHandle, const SArray* tbUidList) { +int tqReaderRemoveTbUidList(STqReader* pHandle, const SArray* tbUidList) { ASSERT(pHandle->tbIdHash != NULL); for (int32_t i = 0; i < taosArrayGetSize(tbUidList); i++) { diff --git a/source/dnode/vnode/src/tsdb/tsdbCache.c b/source/dnode/vnode/src/tsdb/tsdbCache.c index 2086d94099..ea1dfc057f 100644 --- a/source/dnode/vnode/src/tsdb/tsdbCache.c +++ b/source/dnode/vnode/src/tsdb/tsdbCache.c @@ -59,6 +59,8 @@ static void getTableCacheKey(tb_uid_t uid, int cacheType, char *key, int *len) { static void deleteTableCacheLastrow(const void *key, size_t keyLen, void *value) { taosMemoryFree(value); } +static void deleteTableCacheLast(const void *key, size_t keyLen, void *value) { taosArrayDestroy(value); } + static int32_t tsdbCacheDeleteLastrow(SLRUCache *pCache, tb_uid_t uid, TSKEY eKey) { int32_t code = 0; char key[32] = {0}; @@ -761,7 +763,6 @@ static int32_t mergeLastRow(tb_uid_t uid, STsdb *pTsdb, bool *dup, STSRow **ppRo for (int i = 0; i < nMax; ++i) { TSDBKEY maxKey = TSDBROW_KEY(max[i]); - // bool deleted = false; bool deleted = tsdbKeyDeleted(&maxKey, pSkyline, &iSkyline); if (!deleted) { // iMerge[nMerge] = i; @@ -818,12 +819,22 @@ _err: return code; } -static int32_t mergeLast(tb_uid_t uid, STsdb *pTsdb, STSRow **ppRow) { - int32_t code = 0; +typedef struct { + TSKEY ts; + SColVal colVal; +} SLastCol; + +// static int32_t mergeLast(tb_uid_t uid, STsdb *pTsdb, STSRow **ppRow) { +static int32_t mergeLast(tb_uid_t uid, STsdb *pTsdb, SArray **ppLastArray) { + int32_t code = 0; + SArray *pSkyline = NULL; + STSRow *pRow = NULL; + STSRow **ppRow = &pRow; STSchema *pTSchema = metaGetTbTSchema(pTsdb->pVnode->pMeta, uid, -1); int16_t nCol = pTSchema->numOfCols; - SArray *pColArray = taosArrayInit(nCol, sizeof(SColVal)); + // SArray *pColArray = taosArrayInit(nCol, sizeof(SColVal)); + SArray *pColArray = taosArrayInit(nCol, sizeof(SLastCol)); tb_uid_t suid = getTableSuidByUid(uid, pTsdb); @@ -837,9 +848,9 @@ static int32_t mergeLast(tb_uid_t uid, STsdb *pTsdb, STSRow **ppRow) { tsdbGetTbDataFromMemTable(pTsdb->imem, suid, uid, &pIMem); } - *ppRow = NULL; + *ppLastArray = NULL; - SArray *pSkyline = taosArrayInit(32, sizeof(TSDBKEY)); + pSkyline = taosArrayInit(32, sizeof(TSDBKEY)); SDelIdx delIdx; @@ -943,7 +954,6 @@ static int32_t mergeLast(tb_uid_t uid, STsdb *pTsdb, STSRow **ppRow) { for (int i = 0; i < nMax; ++i) { TSDBKEY maxKey = TSDBROW_KEY(max[i]); - // bool deleted = false; bool deleted = tsdbKeyDeleted(&maxKey, pSkyline, &iSkyline); if (!deleted) { iMerge[nMerge] = iMax[i]; @@ -970,8 +980,9 @@ static int32_t mergeLast(tb_uid_t uid, STsdb *pTsdb, STSRow **ppRow) { tRowMergerClear(&merger); } } else { - *ppRow = NULL; - return code; + /* *ppRow = NULL; */ + /* return code; */ + continue; } if (iCol == 0) { @@ -980,7 +991,8 @@ static int32_t mergeLast(tb_uid_t uid, STsdb *pTsdb, STSRow **ppRow) { *pColVal = COL_VAL_VALUE(pTColumn->colId, pTColumn->type, (SValue){.ts = maxKey}); - if (taosArrayPush(pColArray, pColVal) == NULL) { + // if (taosArrayPush(pColArray, pColVal) == NULL) { + if (taosArrayPush(pColArray, &(SLastCol){.ts = TSKEY_MAX, .colVal = *pColVal}) == NULL) { code = TSDB_CODE_OUT_OF_MEMORY; goto _err; } @@ -991,7 +1003,8 @@ static int32_t mergeLast(tb_uid_t uid, STsdb *pTsdb, STSRow **ppRow) { for (int16_t i = iCol; i < nCol; ++i) { // tsdbRowGetColVal(*ppRow, pTSchema, i, pColVal); tTSRowGetVal(*ppRow, pTSchema, i, pColVal); - if (taosArrayPush(pColArray, pColVal) == NULL) { + // if (taosArrayPush(pColArray, pColVal) == NULL) { + if (taosArrayPush(pColArray, &(SLastCol){.ts = maxKey, .colVal = *pColVal}) == NULL) { code = TSDB_CODE_OUT_OF_MEMORY; goto _err; } @@ -1012,11 +1025,11 @@ static int32_t mergeLast(tb_uid_t uid, STsdb *pTsdb, STSRow **ppRow) { --nilColCount; } } - /* + if (*ppRow) { taosMemoryFreeClear(*ppRow); } - */ + continue; } @@ -1024,12 +1037,16 @@ static int32_t mergeLast(tb_uid_t uid, STsdb *pTsdb, STSRow **ppRow) { for (int16_t i = iCol; i < nCol; ++i) { SColVal colVal = {0}; tTSRowGetVal(*ppRow, pTSchema, i, &colVal); + TSKEY rowTs = (*ppRow)->ts; - SColVal *tColVal = (SColVal *)taosArrayGet(pColArray, i); + // SColVal *tColVal = (SColVal *)taosArrayGet(pColArray, i); + SLastCol *tTsVal = (SLastCol *)taosArrayGet(pColArray, i); + SColVal *tColVal = &tTsVal->colVal; if (!colVal.isNone && !colVal.isNull) { if (tColVal->isNull || tColVal->isNone) { - taosArraySet(pColArray, i, &colVal); + // taosArraySet(pColArray, i, &colVal); + taosArraySet(pColArray, i, &(SLastCol){.ts = rowTs, .colVal = colVal}); --nilColCount; } } else { @@ -1054,16 +1071,45 @@ static int32_t mergeLast(tb_uid_t uid, STsdb *pTsdb, STSRow **ppRow) { } while (nilColCount > 0); // if () new ts row from pColArray if non empty - if (taosArrayGetSize(pColArray) == nCol) { - code = tdSTSRowNew(pColArray, pTSchema, ppRow); - if (code) goto _err; + /* if (taosArrayGetSize(pColArray) == nCol) { */ + /* code = tdSTSRowNew(pColArray, pTSchema, ppRow); */ + /* if (code) goto _err; */ + /* } */ + /* taosArrayDestroy(pColArray); */ + if (taosArrayGetSize(pColArray) <= 0) { + *ppLastArray = NULL; + taosArrayDestroy(pColArray); + } else { + *ppLastArray = pColArray; + } + if (*ppRow) { + taosMemoryFreeClear(*ppRow); + } + + for (int i = 0; i < 3; ++i) { + if (input[i].nextRowClearFn) { + input[i].nextRowClearFn(input[i].iter); + } + } + if (pSkyline) { + taosArrayDestroy(pSkyline); } - taosArrayDestroy(pColArray); taosMemoryFreeClear(pTSchema); return code; _err: taosArrayDestroy(pColArray); + if (*ppRow) { + taosMemoryFreeClear(*ppRow); + } + for (int i = 0; i < 3; ++i) { + if (input[i].nextRowClearFn) { + input[i].nextRowClearFn(input[i].iter); + } + } + if (pSkyline) { + taosArrayDestroy(pSkyline); + } taosMemoryFreeClear(pTSchema); tsdbError("vgId:%d merge last_row failed since %s", TD_VID(pTsdb->pVnode), tstrerror(code)); return code; @@ -1103,6 +1149,30 @@ int32_t tsdbCacheGetLastrowH(SLRUCache *pCache, tb_uid_t uid, STsdb *pTsdb, LRUH return code; } +int32_t tsdbCacheLastArray2Row(SArray *pLastArray, STSRow **ppRow, STSchema *pTSchema) { + int32_t code = 0; + int16_t nCol = taosArrayGetSize(pLastArray); + SArray *pColArray = taosArrayInit(nCol, sizeof(SColVal)); + + for (int16_t iCol = 0; iCol < nCol; ++iCol) { + SLastCol *tTsVal = (SLastCol *)taosArrayGet(pLastArray, iCol); + SColVal *tColVal = &tTsVal->colVal; + taosArrayPush(pColArray, tColVal); + } + + code = tdSTSRowNew(pColArray, pTSchema, ppRow); + if (code) goto _err; + + taosArrayDestroy(pColArray); + + return code; + +_err: + taosArrayDestroy(pColArray); + + return code; +} + int32_t tsdbCacheGetLastH(SLRUCache *pCache, tb_uid_t uid, STsdb *pTsdb, LRUHandle **handle) { int32_t code = 0; char key[32] = {0}; @@ -1115,17 +1185,20 @@ int32_t tsdbCacheGetLastH(SLRUCache *pCache, tb_uid_t uid, STsdb *pTsdb, LRUHand //*ppRow = (STSRow *)taosLRUCacheValue(pCache, h); } else { - STSRow *pRow = NULL; - code = mergeLast(uid, pTsdb, &pRow); + // STSRow *pRow = NULL; + // code = mergeLast(uid, pTsdb, &pRow); + SArray *pLastArray = NULL; + code = mergeLast(uid, pTsdb, &pLastArray); // if table's empty or error, return code of -1 - if (code < 0 || pRow == NULL) { + // if (code < 0 || pRow == NULL) { + if (code < 0 || pLastArray == NULL) { *handle = NULL; return 0; } - _taos_lru_deleter_t deleter = deleteTableCacheLastrow; + _taos_lru_deleter_t deleter = deleteTableCacheLast; LRUStatus status = - taosLRUCacheInsert(pCache, key, keyLen, pRow, TD_ROW_LEN(pRow), deleter, NULL, TAOS_LRU_PRIORITY_LOW); + taosLRUCacheInsert(pCache, key, keyLen, pLastArray, pLastArray->capacity, deleter, NULL, TAOS_LRU_PRIORITY_LOW); if (status != TAOS_LRU_STATUS_OK) { code = -1; } diff --git a/source/dnode/vnode/src/tsdb/tsdbCacheRead.c b/source/dnode/vnode/src/tsdb/tsdbCacheRead.c index 30bc603b30..150ed620bf 100644 --- a/source/dnode/vnode/src/tsdb/tsdbCacheRead.c +++ b/source/dnode/vnode/src/tsdb/tsdbCacheRead.c @@ -128,6 +128,8 @@ int32_t tsdbRetrieveLastRow(void* pReader, SSDataBlock* pResBlock, const int32_t } pRow = (STSRow*)taosLRUCacheValue(lruCache, h); + // SArray* pLast = (SArray*)taosLRUCacheValue(lruCache, h); + // tsdbCacheLastArray2Row(pLast, &pRow, pr->pSchema); if (pRow->ts > lastKey) { // Set result row into the same rowIndex repeatly, so we need to check if the internal result row has already // appended or not. @@ -140,6 +142,7 @@ int32_t tsdbRetrieveLastRow(void* pReader, SSDataBlock* pResBlock, const int32_t lastKey = pRow->ts; } + // taosMemoryFree(pRow); tsdbCacheRelease(lruCache, h); } } else if (pr->type == LASTROW_RETRIEVE_TYPE_ALL) { @@ -158,8 +161,12 @@ int32_t tsdbRetrieveLastRow(void* pReader, SSDataBlock* pResBlock, const int32_t } pRow = (STSRow*)taosLRUCacheValue(lruCache, h); + // SArray* pLast = (SArray*)taosLRUCacheValue(lruCache, h); + // tsdbCacheLastArray2Row(pLast, &pRow, pr->pSchema); + saveOneRow(pRow, pResBlock, pr, slotIds); + // taosMemoryFree(pRow); tsdbCacheRelease(lruCache, h); pr->tableIndex += 1; diff --git a/source/dnode/vnode/src/tsdb/tsdbCommit.c b/source/dnode/vnode/src/tsdb/tsdbCommit.c index b60a4697fa..8a189d98b3 100644 --- a/source/dnode/vnode/src/tsdb/tsdbCommit.c +++ b/source/dnode/vnode/src/tsdb/tsdbCommit.c @@ -602,7 +602,7 @@ static int32_t tsdbGetOvlpNRow(STbDataIter *pIter, SBlock *pBlock) { iter.pRow = NULL; while (true) { - pRow = tsdbTbDataIterGet(pIter); + pRow = tsdbTbDataIterGet(&iter); if (pRow == NULL) break; key = TSDBROW_KEY(pRow); @@ -610,7 +610,7 @@ static int32_t tsdbGetOvlpNRow(STbDataIter *pIter, SBlock *pBlock) { c = tBlockCmprFn(&(SBlock){.maxKey = key, .minKey = key}, pBlock); if (c == 0) { nRow++; - tsdbTbDataIterNext(pIter); + tsdbTbDataIterNext(&iter); } else if (c > 0) { break; } else { @@ -635,7 +635,7 @@ static int32_t tsdbMergeAsSubBlock(SCommitter *pCommitter, STbDataIter *pIter, S code = tsdbCommitterUpdateRowSchema(pCommitter, pBlockIdx->suid, pBlockIdx->uid, TSDBROW_SVERSION(pRow)); if (code) goto _err; while (true) { - if (pRow) break; + if (pRow == NULL) break; code = tBlockDataAppendRow(pBlockData, pRow, pCommitter->skmRow.pTSchema); if (code) goto _err; diff --git a/source/dnode/vnode/src/tsdb/tsdbRead.c b/source/dnode/vnode/src/tsdb/tsdbRead.c index 8e5b135858..47aec548a1 100644 --- a/source/dnode/vnode/src/tsdb/tsdbRead.c +++ b/source/dnode/vnode/src/tsdb/tsdbRead.c @@ -48,14 +48,15 @@ typedef struct SBlockOrderSupporter { typedef struct SIOCostSummary { int64_t blockLoadTime; - int64_t statisInfoLoadTime; + int64_t smaLoadTime; int64_t checkForNextTime; int64_t headFileLoad; int64_t headFileLoadTime; } SIOCostSummary; typedef struct SBlockLoadSuppInfo { - SColumnDataAgg* pstatis; + SArray* pColAgg; + SColumnDataAgg tsColAgg; SColumnDataAgg** plist; int16_t* colIds; // column ids for loading file block data int32_t* slotIds; // colId to slotId @@ -63,10 +64,10 @@ typedef struct SBlockLoadSuppInfo { } SBlockLoadSuppInfo; typedef struct SFilesetIter { - int32_t numOfFiles; // number of total files - int32_t index; // current accessed index in the list - SArray* pFileList; // data file list - int32_t order; + int32_t numOfFiles; // number of total files + int32_t index; // current accessed index in the list + SArray* pFileList; // data file list + int32_t order; } SFilesetIter; typedef struct SFileDataBlockInfo { @@ -119,24 +120,10 @@ struct STsdbReader { int32_t type; // query type: 1. retrieve all data blocks, 2. retrieve direct prev|next rows SBlockLoadSuppInfo suppInfo; - SIOCostSummary cost; - STSchema* pSchema; - SDataFReader* pFileReader; - SVersionRange verRange; -#if 0 - SArray* prev; // previous row which is before than time window - SArray* next; // next row which is after the query time window - SFileBlockInfo* pDataBlockInfo; - SDataCols* pDataCols; // in order to hold current file data block - int32_t allocSize; // allocated data block size - SDataBlockLoadInfo dataBlockLoadInfo; /* record current block load information */ - SLoadCompBlockInfo compBlockLoadInfo; /* record current compblock information in SQueryAttr */ - // SDFileSet* pFileGroup; - // SFSIter fileIter; - // SReadH rhelper; - // SColumnDataAgg* statis; // query level statistics, only one table block statistics info exists at any time - // SColumnDataAgg** pstatis;// the ptr array list to return to caller -#endif + SIOCostSummary cost; + STSchema* pSchema; + SDataFReader* pFileReader; + SVersionRange verRange; }; static SFileDataBlockInfo* getCurrentBlockInfo(SDataBlockIter* pBlockIter); @@ -150,7 +137,7 @@ static int32_t doMergeRowsInBuf(SIterInfo* pIter, int64_t ts, SArray* pDelList, static int32_t doAppendOneRow(SSDataBlock* pBlock, STsdbReader* pReader, STSRow* pTSRow); static void setComposedBlockFlag(STsdbReader* pReader, bool composed); static void updateSchema(TSDBROW* pRow, uint64_t uid, STsdbReader* pReader); -static bool hasBeenDropped(const SArray* pDelList, int32_t* index, TSDBKEY* pKey); +static bool hasBeenDropped(const SArray* pDelList, int32_t* index, TSDBKEY* pKey, int32_t order); static void doMergeMultiRows(TSDBROW* pRow, uint64_t uid, SIterInfo* pIter, SArray* pDelList, STSRow** pTSRow, STsdbReader* pReader); @@ -252,33 +239,6 @@ static STimeWindow updateQueryTimeWindow(STsdb* pTsdb, STimeWindow* pWindow) { return win; } -// todo remove this -static void setQueryTimewindow(STsdbReader* pReader, SQueryTableDataCond* pCond, int32_t tWinIdx) { - // pReader->window = pCond->twindows[tWinIdx]; - - // bool updateTs = false; - // int64_t startTs = updateQueryTimeWindow(pReader->pTsdb); - // if (ASCENDING_TRAVERSE(pReader->order)) { - // if (startTs > pReader->window.skey) { - // pReader->window.skey = startTs; - // pCond->twindows[tWinIdx].skey = startTs; - // updateTs = true; - // } - // } else { - // if (startTs > pReader->window.ekey) { - // pReader->window.ekey = startTs; - // pCond->twindows[tWinIdx].ekey = startTs; - // updateTs = true; - // } - // } - - // if (updateTs) { - // tsdbDebug("%p update the query time window, old:%" PRId64 " - %" PRId64 ", new:%" PRId64 " - %" PRId64 ", %s", - // pReader, pCond->twindows[tWinIdx].skey, pCond->twindows[tWinIdx].ekey, pReader->window.skey, - // pReader->window.ekey, pReader->idStr); - // } -} - static void limitOutputBufferSize(const SQueryTableDataCond* pCond, int32_t* capacity) { int32_t rowLen = 0; for (int32_t i = 0; i < pCond->numOfCols; ++i) { @@ -405,21 +365,21 @@ static int32_t tsdbReaderCreate(SVnode* pVnode, SQueryTableDataCond* pCond, STsd pReader->type = pCond->type; pReader->window = updateQueryTimeWindow(pVnode->pTsdb, pCond->twindows); - // todo remove this - setQueryTimewindow(pReader, pCond, 0); ASSERT(pCond->numOfCols > 0); limitOutputBufferSize(pCond, &pReader->capacity); // allocate buffer in order to load data blocks from file SBlockLoadSuppInfo* pSup = &pReader->suppInfo; - pSup->pstatis = taosMemoryCalloc(pCond->numOfCols, sizeof(SColumnDataAgg)); + pSup->pColAgg = taosArrayInit(4, sizeof(SColumnDataAgg)); pSup->plist = taosMemoryCalloc(pCond->numOfCols, POINTER_BYTES); - if (pSup->pstatis == NULL || pSup->plist == NULL) { + if (pSup->pColAgg == NULL || pSup->plist == NULL) { code = TSDB_CODE_OUT_OF_MEMORY; goto _end; } + pSup->tsColAgg.colId = PRIMARYKEY_TIMESTAMP_COL_ID; + pReader->pResBlock = createResBlock(pCond, pReader->capacity); if (pReader->pResBlock == NULL) { code = terrno; @@ -920,200 +880,6 @@ _error: // return midPos; // } -// static int32_t mergeTwoRowFromMem(STsdbReader* pTsdbReadHandle, int32_t capacity, int32_t* curRow, STSRow* row1, -// STSRow* row2, int32_t numOfCols, uint64_t uid, STSchema* pSchema1, STSchema* -// pSchema2, bool update, TSKEY* lastRowKey) { -// #if 1 -// STSchema* pSchema; -// STSRow* row; -// int16_t colId; -// int16_t offset; - -// bool isRow1DataRow = TD_IS_TP_ROW(row1); -// bool isRow2DataRow; -// bool isChosenRowDataRow; -// int32_t chosen_itr; -// SCellVal sVal = {0}; -// TSKEY rowKey = TSKEY_INITIAL_VAL; -// int32_t nResult = 0; -// int32_t mergeOption = 0; // 0 discard 1 overwrite 2 merge - -// // the schema version info is embeded in STSRow -// int32_t numOfColsOfRow1 = 0; - -// if (pSchema1 == NULL) { -// pSchema1 = metaGetTbTSchema(REPO_META(pTsdbReadHandle->pTsdb), uid, TD_ROW_SVER(row1)); -// } - -// #ifdef TD_DEBUG_PRINT_ROW -// char flags[70] = {0}; -// STsdb* pTsdb = pTsdbReadHandle->rhelper.pRepo; -// snprintf(flags, 70, "%s:%d vgId:%d dir:%s row1%s=NULL,row2%s=NULL", __func__, __LINE__, TD_VID(pTsdb->pVnode), -// pTsdb->dir, row1 ? "!" : "", row2 ? "!" : ""); -// tdSRowPrint(row1, pSchema1, flags); -// #endif - -// if (isRow1DataRow) { -// numOfColsOfRow1 = schemaNCols(pSchema1); -// } else { -// numOfColsOfRow1 = tdRowGetNCols(row1); -// } - -// int32_t numOfColsOfRow2 = 0; -// if (row2) { -// isRow2DataRow = TD_IS_TP_ROW(row2); -// if (pSchema2 == NULL) { -// pSchema2 = metaGetTbTSchema(REPO_META(pTsdbReadHandle->pTsdb), uid, TD_ROW_SVER(row2)); -// } -// if (isRow2DataRow) { -// numOfColsOfRow2 = schemaNCols(pSchema2); -// } else { -// numOfColsOfRow2 = tdRowGetNCols(row2); -// } -// } - -// int32_t i = 0, j = 0, k = 0; -// while (i < numOfCols && (j < numOfColsOfRow1 || k < numOfColsOfRow2)) { -// SColumnInfoData* pColInfo = taosArrayGet(pTsdbReadHandle->pColumns, i); - -// int32_t colIdOfRow1; -// if (j >= numOfColsOfRow1) { -// colIdOfRow1 = INT32_MAX; -// } else if (isRow1DataRow) { -// colIdOfRow1 = pSchema1->columns[j].colId; -// } else { -// colIdOfRow1 = tdKvRowColIdAt(row1, j); -// } - -// int32_t colIdOfRow2; -// if (k >= numOfColsOfRow2) { -// colIdOfRow2 = INT32_MAX; -// } else if (isRow2DataRow) { -// colIdOfRow2 = pSchema2->columns[k].colId; -// } else { -// colIdOfRow2 = tdKvRowColIdAt(row2, k); -// } - -// if (colIdOfRow1 < colIdOfRow2) { // the most probability -// if (colIdOfRow1 < pColInfo->info.colId) { -// ++j; -// continue; -// } -// row = row1; -// pSchema = pSchema1; -// isChosenRowDataRow = isRow1DataRow; -// chosen_itr = j; -// } else if (colIdOfRow1 == colIdOfRow2) { -// if (colIdOfRow1 < pColInfo->info.colId) { -// ++j; -// ++k; -// continue; -// } -// row = row1; -// pSchema = pSchema1; -// isChosenRowDataRow = isRow1DataRow; -// chosen_itr = j; -// } else { -// if (colIdOfRow2 < pColInfo->info.colId) { -// ++k; -// continue; -// } -// row = row2; -// pSchema = pSchema2; -// chosen_itr = k; -// isChosenRowDataRow = isRow2DataRow; -// } - -// if (isChosenRowDataRow) { -// colId = pSchema->columns[chosen_itr].colId; -// offset = pSchema->columns[chosen_itr].offset; -// // TODO: use STSRowIter -// tdSTpRowGetVal(row, colId, pSchema->columns[chosen_itr].type, pSchema->flen, offset, chosen_itr - 1, &sVal); -// if (colId == PRIMARYKEY_TIMESTAMP_COL_ID) { -// rowKey = *(TSKEY*)sVal.val; -// if (rowKey != *lastRowKey) { -// mergeOption = 1; -// if (*lastRowKey != TSKEY_INITIAL_VAL) { -// ++(*curRow); -// } -// *lastRowKey = rowKey; -// ++nResult; -// } else if (update) { -// mergeOption = 2; -// } else { -// mergeOption = 0; -// break; -// } -// } -// } else { -// // TODO: use STSRowIter -// if (chosen_itr == 0) { -// colId = PRIMARYKEY_TIMESTAMP_COL_ID; -// tdSKvRowGetVal(row, PRIMARYKEY_TIMESTAMP_COL_ID, -1, -1, &sVal); -// rowKey = *(TSKEY*)sVal.val; -// if (rowKey != *lastRowKey) { -// mergeOption = 1; -// if (*lastRowKey != TSKEY_INITIAL_VAL) { -// ++(*curRow); -// } -// *lastRowKey = rowKey; -// ++nResult; -// } else if (update) { -// mergeOption = 2; -// } else { -// mergeOption = 0; -// break; -// } -// } else { -// SKvRowIdx* pColIdx = tdKvRowColIdxAt(row, chosen_itr - 1); -// colId = pColIdx->colId; -// offset = pColIdx->offset; -// tdSKvRowGetVal(row, colId, offset, chosen_itr - 1, &sVal); -// } -// } - -// ASSERT(rowKey != TSKEY_INITIAL_VAL); - -// if (colId == pColInfo->info.colId) { -// if (tdValTypeIsNorm(sVal.valType)) { -// colDataAppend(pColInfo, *curRow, sVal.val, false); -// } else if (tdValTypeIsNull(sVal.valType)) { -// colDataAppend(pColInfo, *curRow, NULL, true); -// } else if (tdValTypeIsNone(sVal.valType)) { -// // TODO: Set null if nothing append for this row -// if (mergeOption == 1) { -// colDataAppend(pColInfo, *curRow, NULL, true); -// } -// } else { -// ASSERT(0); -// } - -// ++i; - -// if (row == row1) { -// ++j; -// } else { -// ++k; -// } -// } else { -// if (mergeOption == 1) { -// colDataAppend(pColInfo, *curRow, NULL, true); -// } -// ++i; -// } -// } - -// if (mergeOption == 1) { -// while (i < numOfCols) { // the remain columns are all null data -// SColumnInfoData* pColInfo = taosArrayGet(pTsdbReadHandle->pColumns, i); -// colDataAppend(pColInfo, *curRow, NULL, true); -// ++i; -// } -// } - -// return nResult; -// #endif -// } // static void doCheckGeneratedBlockRange(STsdbReader* pTsdbReadHandle) { // SQueryFilePos* cur = &pTsdbReadHandle->cur; @@ -1393,66 +1159,6 @@ _error: // pTsdbReadHandle->idStr); // } -// int32_t binarySearchForKey(char* pValue, int num, TSKEY key, int order) { -// int firstPos, lastPos, midPos = -1; -// int numOfRows; -// TSKEY* keyList; - -// if (num <= 0) return -1; - -// keyList = (TSKEY*)pValue; -// firstPos = 0; -// lastPos = num - 1; - -// if (order == TSDB_ORDER_DESC) { -// // find the first position which is smaller than the key -// while (1) { -// if (key >= keyList[lastPos]) return lastPos; -// if (key == keyList[firstPos]) return firstPos; -// if (key < keyList[firstPos]) return firstPos - 1; - -// numOfRows = lastPos - firstPos + 1; -// midPos = (numOfRows >> 1) + firstPos; - -// if (key < keyList[midPos]) { -// lastPos = midPos - 1; -// } else if (key > keyList[midPos]) { -// firstPos = midPos + 1; -// } else { -// break; -// } -// } - -// } else { -// // find the first position which is bigger than the key -// while (1) { -// if (key <= keyList[firstPos]) return firstPos; -// if (key == keyList[lastPos]) return lastPos; - -// if (key > keyList[lastPos]) { -// lastPos = lastPos + 1; -// if (lastPos >= num) -// return -1; -// else -// return lastPos; -// } - -// numOfRows = lastPos - firstPos + 1; -// midPos = (numOfRows >> 1) + firstPos; - -// if (key < keyList[midPos]) { -// lastPos = midPos - 1; -// } else if (key > keyList[midPos]) { -// firstPos = midPos + 1; -// } else { -// break; -// } -// } -// } - -// return midPos; -// } - static void cleanupBlockOrderSupporter(SBlockOrderSupporter* pSup) { taosMemoryFreeClear(pSup->numOfBlocksPerTable); taosMemoryFreeClear(pSup->indexPerTable); @@ -1710,29 +1416,60 @@ static bool keyOverlapFileBlock(TSDBKEY key, SBlock* pBlock, SVersionRange* pVer (pBlock->minVersion <= pVerRange->maxVer); } +static bool overlapWithDelSkyline(STableBlockScanInfo* pBlockScanInfo, const SBlock* pBlock, int32_t order) { + if (pBlockScanInfo->delSkyline == NULL) { + return false; + } + + TSDBKEY* pFirst = taosArrayGet(pBlockScanInfo->delSkyline, 0); + TSDBKEY* pLast = taosArrayGetLast(pBlockScanInfo->delSkyline); + + // ts is not overlap + if (pBlock->minKey.ts > pLast->ts || pBlock->maxKey.ts < pFirst->ts) { + return false; + } + + int32_t step = ASCENDING_TRAVERSE(order)? 1:-1; + + // version is not overlap + size_t num = taosArrayGetSize(pBlockScanInfo->delSkyline); + for(int32_t i = pBlockScanInfo->fileDelIndex; i < num; i += step) { + TSDBKEY* p = taosArrayGet(pBlockScanInfo->delSkyline, i); + if (p->ts >= pBlock->minKey.ts && p->ts <= pBlock->maxKey.ts) { + if (p->version >= pBlock->minVersion) { + return true; + } + } else if (p->ts > pBlock->maxKey.ts) { + return false; + } + } + + ASSERT(0); + return false; +} + // 1. the version of all rows should be less than the endVersion // 2. current block should not overlap with next neighbor block // 3. current timestamp should not be overlap with each other // 4. output buffer should be large enough to hold all rows in current block +// 5. delete info should not overlap with current block data static bool fileBlockShouldLoad(STsdbReader* pReader, SFileDataBlockInfo* pFBlock, SBlock* pBlock, STableBlockScanInfo* pScanInfo, TSDBKEY key) { int32_t neighborIndex = 0; SBlock* pNeighbor = getNeighborBlockOfSameTable(pFBlock, pScanInfo, &neighborIndex, pReader->order); + // overlap with neighbor bool overlapWithNeighbor = false; if (pNeighbor) { overlapWithNeighbor = overlapWithNeighborBlock(pBlock, pNeighbor, pReader->order); } - bool hasDup = false; - if (pBlock->nSubBlock == 1) { - hasDup = pBlock->hasDup; - } else { - hasDup = true; - } + // has duplicated ts of different version in this block + bool hasDup = (pBlock->nSubBlock == 1)? pBlock->hasDup:true; + bool overlapWithDel= overlapWithDelSkyline(pScanInfo, pBlock, pReader->order); return (overlapWithNeighbor || hasDup || dataBlockPartiallyRequired(&pReader->window, &pReader->verRange, pBlock) || - keyOverlapFileBlock(key, pBlock, &pReader->verRange) || (pBlock->nRow > pReader->capacity)); + keyOverlapFileBlock(key, pBlock, &pReader->verRange) || (pBlock->nRow > pReader->capacity) || overlapWithDel); } static int32_t buildDataBlockFromBuf(STsdbReader* pReader, STableBlockScanInfo* pBlockScanInfo, int64_t endKey) { @@ -1957,7 +1694,7 @@ static bool isValidFileBlockRow(SBlockData* pBlockData, SFileBlockDumpInfo* pDum } TSDBKEY k = {.ts = ts, .version = ver}; - if (hasBeenDropped(pBlockScanInfo->delSkyline, &pBlockScanInfo->fileDelIndex, &k)) { + if (hasBeenDropped(pBlockScanInfo->delSkyline, &pBlockScanInfo->fileDelIndex, &k, pReader->order)) { return false; } @@ -2486,41 +2223,104 @@ static SVersionRange getQueryVerRange(SVnode* pVnode, SQueryTableDataCond* pCond // taosArrayPush(pTsdbReadHandle->pTableCheckInfo, &info); // } -bool hasBeenDropped(const SArray* pDelList, int32_t* index, TSDBKEY* pKey) { +bool hasBeenDropped(const SArray* pDelList, int32_t* index, TSDBKEY* pKey, int32_t order) { ASSERT(pKey != NULL); if (pDelList == NULL) { return false; } + size_t num = taosArrayGetSize(pDelList); + bool asc = ASCENDING_TRAVERSE(order); + int32_t step = asc? 1:-1; - if (*index >= taosArrayGetSize(pDelList) - 1) { - TSDBKEY* last = taosArrayGetLast(pDelList); - if (pKey->ts > last->ts) { - return false; - } else if (pKey->ts == last->ts) { - size_t size = taosArrayGetSize(pDelList); - TSDBKEY* prev = taosArrayGet(pDelList, size - 2); - if (prev->version >= pKey->version) { - return true; - } else { + if (asc) { + if (*index >= num - 1) { + TSDBKEY* last = taosArrayGetLast(pDelList); + ASSERT(pKey->ts >= last->ts); + + if (pKey->ts > last->ts) { return false; + } else if (pKey->ts == last->ts) { + TSDBKEY* prev = taosArrayGet(pDelList, num - 2); + return (prev->version >= pKey->version); } } else { - ASSERT(0); + TSDBKEY* pCurrent = taosArrayGet(pDelList, *index); + TSDBKEY* pNext = taosArrayGet(pDelList, (*index) + 1); + + if (pKey->ts < pCurrent->ts) { + return false; + } + + if (pCurrent->ts <= pKey->ts && pNext->ts >= pKey->ts && pCurrent->version >= pKey->version) { + return true; + } + + while (pNext->ts <= pKey->ts && (*index) < num - 1) { + (*index) += 1; + + if ((*index) < num - 1) { + pCurrent = taosArrayGet(pDelList, *index); + pNext = taosArrayGet(pDelList, (*index) + 1); + + // it is not a consecutive deletion range, ignore it + if (pCurrent->version == 0 && pNext->version > 0) { + continue; + } + + if (pCurrent->ts <= pKey->ts && pNext->ts >= pKey->ts && pCurrent->version >= pKey->version) { + return true; + } + } + } + + return false; } } else { - TSDBKEY* pCurrent = taosArrayGet(pDelList, *index); - TSDBKEY* pNext = taosArrayGet(pDelList, (*index) + 1); + if (*index <= 0) { + TSDBKEY* pFirst = taosArrayGet(pDelList, 0); - if (pCurrent->ts <= pKey->ts && pNext->ts >= pKey->ts && pCurrent->version >= pKey->version) { - return true; + if (pKey->ts < pFirst->ts) { + return false; + } else if (pKey->ts == pFirst->ts) { + return pFirst->version >= pKey->version; + } else { + ASSERT(0); + } } else { - while (pNext->ts < pKey->ts && (*index) < taosArrayGetSize(pDelList) - 1) { - (*index) += 1; + TSDBKEY* pCurrent = taosArrayGet(pDelList, *index); + TSDBKEY* pPrev = taosArrayGet(pDelList, (*index) - 1); + + if (pKey->ts > pCurrent->ts) { + return false; + } + + if (pPrev->ts <= pKey->ts && pCurrent->ts >= pKey->ts && pPrev->version >= pKey->version) { + return true; + } + + while (pPrev->ts >= pKey->ts && (*index) > 1) { + (*index) += step; + + if ((*index) >= 1) { + pCurrent = taosArrayGet(pDelList, *index); + pPrev = taosArrayGet(pDelList, (*index) - 1); + + // it is not a consecutive deletion range, ignore it + if (pCurrent->version > 0 && pPrev->version == 0) { + continue; + } + + if (pPrev->ts <= pKey->ts && pCurrent->ts >= pKey->ts && pPrev->version >= pKey->version) { + return true; + } + } } return false; } } + + return false; } TSDBROW* getValidRow(SIterInfo* pIter, const SArray* pDelList, STsdbReader* pReader) { @@ -2537,7 +2337,7 @@ TSDBROW* getValidRow(SIterInfo* pIter, const SArray* pDelList, STsdbReader* pRea // it is a valid data version if ((key.version <= pReader->verRange.maxVer && key.version >= pReader->verRange.minVer) && - (!hasBeenDropped(pDelList, &pIter->index, &key))) { + (!hasBeenDropped(pDelList, &pIter->index, &key, pReader->order))) { return pRow; } @@ -2556,7 +2356,7 @@ TSDBROW* getValidRow(SIterInfo* pIter, const SArray* pDelList, STsdbReader* pRea } if (key.version <= pReader->verRange.maxVer && key.version >= pReader->verRange.minVer && - (!hasBeenDropped(pDelList, &pIter->index, &key))) { + (!hasBeenDropped(pDelList, &pIter->index, &key, pReader->order))) { return pRow; } } @@ -2908,162 +2708,6 @@ int32_t tsdbGetStbIdList(SMeta* pMeta, int64_t suid, SArray* list) { return TSDB_CODE_SUCCESS; } -// static void destroyHelper(void* param) { -// if (param == NULL) { -// return; -// } - -// // tQueryInfo* pInfo = (tQueryInfo*)param; -// // if (pInfo->optr != TSDB_RELATION_IN) { -// // taosMemoryFreeClear(pInfo->q); -// // } else { -// // taosHashCleanup((SHashObj *)(pInfo->q)); -// // } - -// taosMemoryFree(param); -// } - -// #define TSDB_PREV_ROW 0x1 -// #define TSDB_NEXT_ROW 0x2 - -// static bool loadBlockOfActiveTable(STsdbReader* pTsdbReadHandle) { -// if (pTsdbReadHandle->checkFiles) { -// // check if the query range overlaps with the file data block -// bool exists = true; - -// int32_t code = buildBlockFromFiles(pTsdbReadHandle, &exists); -// if (code != TSDB_CODE_SUCCESS) { -// pTsdbReadHandle->checkFiles = false; -// return false; -// } - -// if (exists) { -// tsdbRetrieveDataBlock((STsdbReader**)pTsdbReadHandle, NULL); -// if (pTsdbReadHandle->currentLoadExternalRows && pTsdbReadHandle->window.skey == pTsdbReadHandle->window.ekey) { -// SColumnInfoData* pColInfo = taosArrayGet(pTsdbReadHandle->pColumns, 0); -// assert(*(int64_t*)pColInfo->pData == pTsdbReadHandle->window.skey); -// } - -// pTsdbReadHandle->currentLoadExternalRows = false; // clear the flag, since the exact matched row is found. -// return exists; -// } - -// pTsdbReadHandle->checkFiles = false; -// } - -// if (hasMoreDataInCache(pTsdbReadHandle)) { -// pTsdbReadHandle->currentLoadExternalRows = false; -// return true; -// } - -// // current result is empty -// if (pTsdbReadHandle->currentLoadExternalRows && pTsdbReadHandle->window.skey == pTsdbReadHandle->window.ekey && -// pTsdbReadHandle->cur.rows == 0) { -// // SMemTable* pMemRef = pTsdbReadHandle->pMemTable; - -// // doGetExternalRow(pTsdbReadHandle, TSDB_PREV_ROW, pMemRef); -// // doGetExternalRow(pTsdbReadHandle, TSDB_NEXT_ROW, pMemRef); - -// bool result = tsdbGetExternalRow(pTsdbReadHandle); - -// // pTsdbReadHandle->prev = doFreeColumnInfoData(pTsdbReadHandle->prev); -// // pTsdbReadHandle->next = doFreeColumnInfoData(pTsdbReadHandle->next); -// pTsdbReadHandle->currentLoadExternalRows = false; - -// return result; -// } - -// return false; -// } - -// static bool loadDataBlockFromTableSeq(STsdbReader* pTsdbReadHandle) { -// size_t numOfTables = taosArrayGetSize(pTsdbReadHandle->pTableCheckInfo); -// assert(numOfTables > 0); - -// int64_t stime = taosGetTimestampUs(); - -// while (pTsdbReadHandle->activeIndex < numOfTables) { -// if (loadBlockOfActiveTable(pTsdbReadHandle)) { -// return true; -// } - -// STableBlockScanInfo* pCheckInfo = taosArrayGet(pTsdbReadHandle->pTableCheckInfo, pTsdbReadHandle->activeIndex); -// pCheckInfo->numOfBlocks = 0; - -// pTsdbReadHandle->activeIndex += 1; -// pTsdbReadHandle->locateStart = false; -// pTsdbReadHandle->checkFiles = true; -// pTsdbReadHandle->cur.rows = 0; -// pTsdbReadHandle->currentLoadExternalRows = pTsdbReadHandle->loadExternalRow; - -// terrno = TSDB_CODE_SUCCESS; - -// int64_t elapsedTime = taosGetTimestampUs() - stime; -// pTsdbReadHandle->cost.checkForNextTime += elapsedTime; -// } - -// return false; -// } - -// bool tsdbGetExternalRow(STsdbReader* pHandle) { -// STsdbReader* pTsdbReadHandle = (STsdbReader*)pHandle; -// SQueryFilePos* cur = &pTsdbReadHandle->cur; - -// cur->fid = INT32_MIN; -// cur->mixBlock = true; -// if (pTsdbReadHandle->prev == NULL || pTsdbReadHandle->next == NULL) { -// cur->rows = 0; -// return false; -// } - -// int32_t numOfCols = (int32_t)QH_GET_NUM_OF_COLS(pTsdbReadHandle); -// for (int32_t i = 0; i < numOfCols; ++i) { -// SColumnInfoData* pColInfoData = taosArrayGet(pTsdbReadHandle->pColumns, i); -// SColumnInfoData* first = taosArrayGet(pTsdbReadHandle->prev, i); - -// memcpy(pColInfoData->pData, first->pData, pColInfoData->info.bytes); - -// SColumnInfoData* sec = taosArrayGet(pTsdbReadHandle->next, i); -// memcpy(((char*)pColInfoData->pData) + pColInfoData->info.bytes, sec->pData, pColInfoData->info.bytes); - -// if (i == 0 && pColInfoData->info.type == TSDB_DATA_TYPE_TIMESTAMP) { -// cur->win.skey = *(TSKEY*)pColInfoData->pData; -// cur->win.ekey = *(TSKEY*)(((char*)pColInfoData->pData) + TSDB_KEYSIZE); -// } -// } - -// cur->rows = 2; -// return true; -// } - -// static void* doFreeColumnInfoData(SArray* pColumnInfoData) { -// if (pColumnInfoData == NULL) { -// return NULL; -// } - -// size_t cols = taosArrayGetSize(pColumnInfoData); -// for (int32_t i = 0; i < cols; ++i) { -// SColumnInfoData* pColInfo = taosArrayGet(pColumnInfoData, i); -// colDataDestroy(pColInfo); -// } - -// taosArrayDestroy(pColumnInfoData); -// return NULL; -// } - -// static void* destroyTableCheckInfo(SArray* pTableCheckInfo) { -// size_t size = taosArrayGetSize(pTableCheckInfo); -// for (int32_t i = 0; i < size; ++i) { -// STableBlockScanInfo* p = taosArrayGet(pTableCheckInfo, i); -// destroyTableMemIterator(p); - -// taosMemoryFreeClear(p->pCompInfo); -// } - -// taosArrayDestroy(pTableCheckInfo); -// return NULL; -// } - // ====================================== EXPOSED APIs ====================================== int32_t tsdbReaderOpen(SVnode* pVnode, SQueryTableDataCond* pCond, SArray* pTableList, STsdbReader** ppReader, const char* idstr) { @@ -3126,16 +2770,11 @@ void tsdbReaderClose(STsdbReader* pReader) { } blockDataDestroy(pReader->pResBlock); - - taosMemoryFreeClear(pReader->suppInfo.pstatis); taosMemoryFreeClear(pReader->suppInfo.plist); + + taosArrayDestroy(pReader->suppInfo.pColAgg); taosMemoryFree(pReader->suppInfo.slotIds); - if (!isEmptyQueryTimeWindow(&pReader->window)) { - // tsdbMayUnTakeMemSnapshot(pTsdbReadHandle); - } else { - ASSERT(pReader->status.pTableMap == NULL); - } #if 0 // if (pReader->status.pTableScanInfo != NULL) { // pReader->status.pTableScanInfo = destroyTableCheckInfo(pReader->status.pTableScanInfo); @@ -3154,7 +2793,7 @@ void tsdbReaderClose(STsdbReader* pReader) { tsdbDebug("%p :io-cost summary: head-file read cnt:%" PRIu64 ", head-file time:%" PRIu64 " us, statis-info:%" PRId64 " us, datablock:%" PRId64 " us, check data:%" PRId64 " us, %s", - pReader, pCost->headFileLoad, pCost->headFileLoadTime, pCost->statisInfoLoadTime, pCost->blockLoadTime, + pReader, pCost->headFileLoad, pCost->headFileLoadTime, pCost->smaLoadTime, pCost->blockLoadTime, pCost->checkForNextTime, pReader->idStr); taosMemoryFree(pReader->idStr); @@ -3207,77 +2846,76 @@ void tsdbRetrieveDataBlockInfo(STsdbReader* pReader, SDataBlockInfo* pDataBlockI pDataBlockInfo->window = pReader->pResBlock->info.window; } -int32_t tsdbRetrieveDataBlockStatisInfo(STsdbReader* pReader, SColumnDataAgg*** pBlockStatis, bool* allHave) { +int32_t tsdbRetrieveDatablockSMA(STsdbReader* pReader, SColumnDataAgg*** pBlockStatis, bool* allHave) { int32_t code = 0; *allHave = false; + // there is no statistics data for composed block if (pReader->status.composedDataBlock) { *pBlockStatis = NULL; return TSDB_CODE_SUCCESS; } - // SFileBlockInfo* pBlockInfo = &pReader->pDataBlockInfo[c->slot]; - // assert((c->slot >= 0 && c->slot < pReader->numOfBlocks) || ((c->slot == pReader->numOfBlocks) && (c->slot == 0))); + SFileDataBlockInfo* pFBlock = getCurrentBlockInfo(&pReader->status.blockIter); + STableBlockScanInfo* pBlockScanInfo = taosHashGet(pReader->status.pTableMap, &pFBlock->uid, sizeof(pFBlock->uid)); + SBlock* pBlock = taosArrayGet(pBlockScanInfo->pBlockList, pFBlock->tbBlockIdx); - // // file block with sub-blocks has no statistics data - // if (pBlockInfo->compBlock->numOfSubBlocks > 1) { - // *pBlockStatis = NULL; - // return TSDB_CODE_SUCCESS; - // } + int64_t stime = taosGetTimestampUs(); - // int64_t stime = taosGetTimestampUs(); - // int statisStatus = tsdbLoadBlockStatis(&pReader->rhelper, pBlockInfo->compBlock); - // if (statisStatus < TSDB_STATIS_OK) { - // return terrno; - // } else if (statisStatus > TSDB_STATIS_OK) { - // *pBlockStatis = NULL; - // return TSDB_CODE_SUCCESS; - // } + SBlockLoadSuppInfo* pSup = &pReader->suppInfo; - // tsdbDebug("vgId:%d, succeed to load block statis part for uid %" PRIu64, REPO_ID(pReader->pTsdb), - // TSDB_READ_TABLE_UID(&pReader->rhelper)); + if (tBlockHasSma(pBlock)) { + code = tsdbReadBlockSma(pReader->pFileReader, pBlock, pSup->pColAgg, NULL); + if (code != TSDB_CODE_SUCCESS) { + tsdbDebug("vgId:%d, failed to load block SMA for uid %" PRIu64 ", code:%s, %s", 0, pFBlock->uid, tstrerror(code), + pReader->idStr); + return code; + } + } else { + *pBlockStatis = NULL; + return TSDB_CODE_SUCCESS; + } - // int16_t* colIds = pReader->suppInfo.defaultLoadColumn->pData; + *allHave = true; - // size_t numOfCols = QH_GET_NUM_OF_COLS(pReader); - // memset(pReader->suppInfo.plist, 0, numOfCols * POINTER_BYTES); - // memset(pReader->suppInfo.pstatis, 0, numOfCols * sizeof(SColumnDataAgg)); + // always load the first primary timestamp column data + SColumnDataAgg* pTsAgg = &pSup->tsColAgg; - // for (int32_t i = 0; i < numOfCols; ++i) { - // pReader->suppInfo.pstatis[i].colId = colIds[i]; - // } + pTsAgg->numOfNull = 0; + pTsAgg->colId = PRIMARYKEY_TIMESTAMP_COL_ID; + pTsAgg->min = pReader->pResBlock->info.window.skey; + pTsAgg->max = pReader->pResBlock->info.window.ekey; + pSup->plist[0] = pTsAgg; - // *allHave = true; - // tsdbGetBlockStatis(&pReader->rhelper, pReader->suppInfo.pstatis, (int)numOfCols, pBlockInfo->compBlock); + // update the number of NULL data rows + size_t numOfCols = blockDataGetNumOfCols(pReader->pResBlock); - // // always load the first primary timestamp column data - // SColumnDataAgg* pPrimaryColStatis = &pReader->suppInfo.pstatis[0]; - // assert(pPrimaryColStatis->colId == PRIMARYKEY_TIMESTAMP_COL_ID); + int32_t i = 0, j = 0; + while (j < numOfCols && i < taosArrayGetSize(pSup->pColAgg)) { + SColumnDataAgg* pAgg = taosArrayGet(pSup->pColAgg, i); + if (pAgg->colId == pSup->colIds[j]) { + if (IS_BSMA_ON(&(pReader->pSchema->columns[i]))) { + pSup->plist[j] = pAgg; + } else { + *allHave = false; + } + i += 1; + j += 1; + } else if (pAgg->colId < pSup->colIds[j]) { + i += 1; + } else if (pSup->colIds[j] < pAgg->colId) { + j += 1; + } + } - // pPrimaryColStatis->numOfNull = 0; - // pPrimaryColStatis->min = pBlockInfo->compBlock->minKey.ts; - // pPrimaryColStatis->max = pBlockInfo->compBlock->maxKey.ts; - // pReader->suppInfo.plist[0] = &pReader->suppInfo.pstatis[0]; + int64_t elapsed = taosGetTimestampUs() - stime; + pReader->cost.smaLoadTime += elapsed; - // // update the number of NULL data rows - // int32_t* slotIds = pReader->suppInfo.slotIds; - // for (int32_t i = 1; i < numOfCols; ++i) { - // ASSERT(colIds[i] == pReader->pSchema->columns[slotIds[i]].colId); - // if (IS_BSMA_ON(&(pReader->pSchema->columns[slotIds[i]]))) { - // if (pReader->suppInfo.pstatis[i].numOfNull == -1) { // set the column data are all NULL - // pReader->suppInfo.pstatis[i].numOfNull = pBlockInfo->compBlock->numOfRows; - // } + *pBlockStatis = pSup->plist; - // pReader->suppInfo.plist[i] = &pReader->suppInfo.pstatis[i]; - // } else { - // *allHave = false; - // } - // } + tsdbDebug("vgId:%d, succeed to load block SMA for uid %" PRIu64 ", elapsed time:%" PRId64 "us, %s", 0, pFBlock->uid, + elapsed, pReader->idStr); - // int64_t elapsed = taosGetTimestampUs() - stime; - // pReader->cost.statisInfoLoadTime += elapsed; - - // *pBlockStatis = pReader->suppInfo.plist; return code; } @@ -3312,19 +2950,19 @@ int32_t tsdbReaderReset(STsdbReader* pReader, SQueryTableDataCond* pCond, int32_ return TSDB_CODE_SUCCESS; } - setQueryTimewindow(pReader, pCond, tWinIdx); - - pReader->order = pCond->order; - pReader->type = BLOCK_LOAD_OFFSET_ORDER; + pReader->order = pCond->order; + pReader->type = BLOCK_LOAD_OFFSET_ORDER; pReader->status.loadFromFile = true; pReader->status.pTableIter = NULL; pReader->window = updateQueryTimeWindow(pReader->pTsdb, &pCond->twindows[tWinIdx]); // allocate buffer in order to load data blocks from file - memset(pReader->suppInfo.pstatis, 0, sizeof(SColumnDataAgg)); + memset(&pReader->suppInfo.tsColAgg, 0, sizeof(SColumnDataAgg)); memset(pReader->suppInfo.plist, 0, POINTER_BYTES); + pReader->suppInfo.tsColAgg.colId = PRIMARYKEY_TIMESTAMP_COL_ID; + // todo set the correct numOfTables int32_t numOfTables = 1; SDataBlockIter* pBlockIter = &pReader->status.blockIter; @@ -3350,114 +2988,80 @@ int32_t tsdbReaderReset(STsdbReader* pReader, SQueryTableDataCond* pCond, int32_ return code; } +static int32_t getBucketIndex(int32_t startRow, int32_t bucketRange, int32_t numOfRows) { + return (numOfRows - startRow) / bucketRange; +} + int32_t tsdbGetFileBlocksDistInfo(STsdbReader* pReader, STableBlockDistInfo* pTableBlockInfo) { - int32_t code = 0; - // pTableBlockInfo->totalSize = 0; - // pTableBlockInfo->totalRows = 0; + int32_t code = TSDB_CODE_SUCCESS; + pTableBlockInfo->totalSize = 0; + pTableBlockInfo->totalRows = 0; - // STsdbFS* pFileHandle = REPO_FS(pReader->pTsdb); + // find the start data block in file + SReaderStatus* pStatus = &pReader->status; - // // find the start data block in file - // pReader->locateStart = true; - // STsdbKeepCfg* pCfg = REPO_KEEP_CFG(pReader->pTsdb); - // int32_t fid = getFileIdFromKey(pReader->window.skey, pCfg->days, pCfg->precision); + STsdbCfg* pc = &pReader->pTsdb->pVnode->config.tsdbCfg; + pTableBlockInfo->defMinRows = pc->minRows; + pTableBlockInfo->defMaxRows = pc->maxRows; - // tsdbRLockFS(pFileHandle); - // tsdbFSIterInit(&pReader->fileIter, pFileHandle, pReader->order); - // tsdbFSIterSeek(&pReader->fileIter, fid); - // tsdbUnLockFS(pFileHandle); + int32_t bucketRange = ceil((pc->maxRows - pc->minRows) / 20.0); - // STsdbCfg* pc = REPO_CFG(pReader->pTsdb); - // pTableBlockInfo->defMinRows = pc->minRows; - // pTableBlockInfo->defMaxRows = pc->maxRows; + pTableBlockInfo->numOfFiles += 1; - // int32_t bucketRange = ceil((pc->maxRows - pc->minRows) / 20.0); + int32_t numOfTables = (int32_t)taosHashGetSize(pStatus->pTableMap); + int defaultRows = 4096; - // pTableBlockInfo->numOfFiles += 1; + SDataBlockIter* pBlockIter = &pStatus->blockIter; + pTableBlockInfo->numOfFiles += pStatus->fileIter.numOfFiles; + pTableBlockInfo->numOfBlocks += pBlockIter->numOfBlocks; - // int32_t code = TSDB_CODE_SUCCESS; - // int32_t numOfBlocks = 0; - // int32_t numOfTables = (int32_t)taosArrayGetSize(pReader->pTableCheckInfo); - // int defaultRows = 4096; - // STimeWindow win = TSWINDOW_INITIALIZER; + pTableBlockInfo->numOfTables = numOfTables; + bool hasNext = true; - // while (true) { - // numOfBlocks = 0; - // tsdbRLockFS(REPO_FS(pReader->pTsdb)); + while (true) { + if (hasNext) { + SFileDataBlockInfo* pFBlock = getCurrentBlockInfo(pBlockIter); + STableBlockScanInfo* pScanInfo = taosHashGet(pStatus->pTableMap, &pFBlock->uid, sizeof(pFBlock->uid)); + SBlock* pBlock = taosArrayGet(pScanInfo->pBlockList, pFBlock->tbBlockIdx); - // if ((pReader->pFileGroup = tsdbFSIterNext(&pReader->fileIter)) == NULL) { - // tsdbUnLockFS(REPO_FS(pReader->pTsdb)); - // break; - // } + int32_t numOfRows = pBlock->nRow; + pTableBlockInfo->totalRows += numOfRows; - // tsdbGetFidKeyRange(pCfg->days, pCfg->precision, pReader->pFileGroup->fid, &win.skey, &win.ekey); + if (numOfRows > pTableBlockInfo->maxRows) { + pTableBlockInfo->maxRows = numOfRows; + } - // // current file are not overlapped with query time window, ignore remain files - // if ((win.skey > pReader->window.ekey) /* || (!ascTraverse && win.ekey < pTsdbReadHandle->window.ekey)*/) { - // tsdbUnLockFS(REPO_FS(pReader->pTsdb)); - // tsdbDebug("%p remain files are not qualified for qrange:%" PRId64 "-%" PRId64 ", ignore, %s", pReader, - // pReader->window.skey, pReader->window.ekey, pReader->idStr); - // pReader->pFileGroup = NULL; - // break; - // } + if (numOfRows < pTableBlockInfo->minRows) { + pTableBlockInfo->minRows = numOfRows; + } - // pTableBlockInfo->numOfFiles += 1; - // if (tsdbSetAndOpenReadFSet(&pReader->rhelper, pReader->pFileGroup) < 0) { - // tsdbUnLockFS(REPO_FS(pReader->pTsdb)); - // code = terrno; - // break; - // } + if (numOfRows < defaultRows) { + pTableBlockInfo->numOfSmallBlocks += 1; + } - // tsdbUnLockFS(REPO_FS(pReader->pTsdb)); + int32_t bucketIndex = getBucketIndex(pTableBlockInfo->defMinRows, bucketRange, numOfRows); + pTableBlockInfo->blockRowsHisto[bucketIndex]++; - // if (tsdbLoadBlockIdx(&pReader->rhelper) < 0) { - // code = terrno; - // break; - // } + hasNext = blockIteratorNext(&pStatus->blockIter); - // if ((code = getFileCompInfo(pReader, &numOfBlocks)) != TSDB_CODE_SUCCESS) { - // break; - // } + } else { + code = initForFirstBlockInFile(pReader, pBlockIter); + if ((code != TSDB_CODE_SUCCESS) || (pReader->status.loadFromFile == false)) { + break; + } - // tsdbDebug("%p %d blocks found in file for %d table(s), fid:%d, %s", pReader, numOfBlocks, numOfTables, - // pReader->pFileGroup->fid, pReader->idStr); + pTableBlockInfo->numOfBlocks += pBlockIter->numOfBlocks; + } - // if (numOfBlocks == 0) { - // continue; - // } +/* + hasNext = blockIteratorNext(&pStatus->blockIter); +*/ - // pTableBlockInfo->numOfBlocks += numOfBlocks; - // for (int32_t i = 0; i < numOfTables; ++i) { - // STableBlockScanInfo* pCheckInfo = taosArrayGet(pReader->pTableCheckInfo, i); +// tsdbDebug("%p %d blocks found in file for %d table(s), fid:%d, %s", pReader, numOfBlocks, numOfTables, +// pReader->pFileGroup->fid, pReader->idStr); + } - // SBlock* pBlock = pCheckInfo->pCompInfo->blocks; - - // for (int32_t j = 0; j < pCheckInfo->numOfBlocks; ++j) { - // pTableBlockInfo->totalSize += pBlock[j].len; - - // int32_t numOfRows = pBlock[j].numOfRows; - // pTableBlockInfo->totalRows += numOfRows; - - // if (numOfRows > pTableBlockInfo->maxRows) { - // pTableBlockInfo->maxRows = numOfRows; - // } - - // if (numOfRows < pTableBlockInfo->minRows) { - // pTableBlockInfo->minRows = numOfRows; - // } - - // if (numOfRows < defaultRows) { - // pTableBlockInfo->numOfSmallBlocks += 1; - // } - - // int32_t bucketIndex = getBucketIndex(pTableBlockInfo->defMinRows, bucketRange, numOfRows); - // pTableBlockInfo->blockRowsHisto[bucketIndex]++; - // } - // } - // } - - // pTableBlockInfo->numOfTables = numOfTables; return code; } diff --git a/source/dnode/vnode/src/tsdb/tsdbReadImpl.c b/source/dnode/vnode/src/tsdb/tsdbReadImpl.c deleted file mode 100644 index fe0d3a1b6f..0000000000 --- a/source/dnode/vnode/src/tsdb/tsdbReadImpl.c +++ /dev/null @@ -1,16 +0,0 @@ -/* - * 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 . - */ - -#include "tsdb.h" diff --git a/source/dnode/vnode/src/tsdb/tsdbReaderWriter.c b/source/dnode/vnode/src/tsdb/tsdbReaderWriter.c index 2bd62348c9..a331f88e61 100644 --- a/source/dnode/vnode/src/tsdb/tsdbReaderWriter.c +++ b/source/dnode/vnode/src/tsdb/tsdbReaderWriter.c @@ -969,6 +969,8 @@ int32_t tsdbReadColData(SDataFReader *pReader, SBlockIdx *pBlockIdx, SBlock *pBl SBlockData *pBlockData1 = &(SBlockData){0}; SBlockData *pBlockData2 = &(SBlockData){0}; + tBlockDataInit(pBlockData1); + tBlockDataInit(pBlockData2); for (int32_t iSubBlock = 1; iSubBlock < pBlock->nSubBlock; iSubBlock++) { code = tsdbReadSubColData(pReader, pBlockIdx, pBlock, iSubBlock, aColId, nCol, pBlockData1, ppBuf1, ppBuf2); if (code) goto _err; @@ -1106,6 +1108,8 @@ int32_t tsdbReadBlockData(SDataFReader *pReader, SBlockIdx *pBlockIdx, SBlock *p SBlockData *pBlockData1 = &(SBlockData){0}; SBlockData *pBlockData2 = &(SBlockData){0}; + tBlockDataInit(pBlockData1); + tBlockDataInit(pBlockData2); for (iSubBlock = 1; iSubBlock < pBlock->nSubBlock; iSubBlock++) { code = tsdbReadSubBlockData(pReader, pBlockIdx, pBlock, iSubBlock, pBlockData1, ppBuf1, ppBuf2); if (code) { @@ -1152,7 +1156,7 @@ _err: int32_t tsdbReadBlockSma(SDataFReader *pReader, SBlock *pBlock, SArray *aColumnDataAgg, uint8_t **ppBuf) { int32_t code = 0; TdFilePtr pFD = pReader->pSmaFD; - int64_t offset = pBlock->aSubBlock[0].offset; + int64_t offset = pBlock->aSubBlock[0].sOffset; int64_t size = pBlock->aSubBlock[0].nSma * sizeof(SColumnDataAgg) + sizeof(TSCKSUM); uint8_t *pBuf = NULL; int64_t n; @@ -1175,10 +1179,13 @@ int32_t tsdbReadBlockSma(SDataFReader *pReader, SBlock *pBlock, SArray *aColumnD if (n < 0) { code = TAOS_SYSTEM_ERROR(errno); goto _err; + } else if (n < size) { + code = TSDB_CODE_FILE_CORRUPTED; + goto _err; } // check - if (!taosCheckChecksumWhole(NULL, size)) { + if (!taosCheckChecksumWhole(*ppBuf, size)) { code = TSDB_CODE_FILE_CORRUPTED; goto _err; } diff --git a/source/dnode/vnode/src/tsdb/tsdbUtil.c b/source/dnode/vnode/src/tsdb/tsdbUtil.c index f906ef1b54..ed8bdc17a5 100644 --- a/source/dnode/vnode/src/tsdb/tsdbUtil.c +++ b/source/dnode/vnode/src/tsdb/tsdbUtil.c @@ -821,16 +821,20 @@ int32_t tColDataCopy(SColData *pColDataSrc, SColData *pColDataDest) { int32_t code = 0; int32_t size; + ASSERT(pColDataSrc->nVal > 0); + pColDataDest->cid = pColDataSrc->cid; pColDataDest->type = pColDataSrc->type; pColDataDest->smaOn = pColDataSrc->smaOn; pColDataDest->nVal = pColDataSrc->nVal; pColDataDest->flag = pColDataSrc->flag; - size = BIT2_SIZE(pColDataSrc->nVal); - code = tRealloc(&pColDataDest->pBitMap, size); - if (code) goto _exit; - memcpy(pColDataDest->pBitMap, pColDataSrc->pBitMap, size); + if (pColDataSrc->flag != HAS_NONE && pColDataSrc->flag != HAS_NULL && pColDataSrc->flag != HAS_VALUE) { + size = BIT2_SIZE(pColDataSrc->nVal); + code = tRealloc(&pColDataDest->pBitMap, size); + if (code) goto _exit; + memcpy(pColDataDest->pBitMap, pColDataSrc->pBitMap, size); + } if (IS_VAR_DATA_TYPE(pColDataDest->type)) { size = sizeof(int32_t) * pColDataSrc->nVal; @@ -1230,10 +1234,26 @@ void tsdbCalcColDataSMA(SColData *pColData, SColumnDataAgg *pColAgg) { break; case TSDB_DATA_TYPE_SMALLINT: break; - case TSDB_DATA_TYPE_INT: + case TSDB_DATA_TYPE_INT: { + pColAgg->sum += colVal.value.i32; + if (pColAgg->min > colVal.value.i32) { + pColAgg->min = colVal.value.i32; + } + if (pColAgg->max < colVal.value.i32) { + pColAgg->max = colVal.value.i32; + } break; - case TSDB_DATA_TYPE_BIGINT: + } + case TSDB_DATA_TYPE_BIGINT: { + pColAgg->sum += colVal.value.i64; + if (pColAgg->min > colVal.value.i64) { + pColAgg->min = colVal.value.i64; + } + if (pColAgg->max < colVal.value.i64) { + pColAgg->max = colVal.value.i64; + } break; + } case TSDB_DATA_TYPE_FLOAT: break; case TSDB_DATA_TYPE_DOUBLE: diff --git a/source/dnode/vnode/src/vnd/vnodeStateMgr.c b/source/dnode/vnode/src/vnd/vnodeStateMgr.c deleted file mode 100644 index 6dea4a4e57..0000000000 --- a/source/dnode/vnode/src/vnd/vnodeStateMgr.c +++ /dev/null @@ -1,14 +0,0 @@ -/* - * 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 . - */ \ No newline at end of file diff --git a/source/libs/executor/inc/executorimpl.h b/source/libs/executor/inc/executorimpl.h index 105bf1476d..3488069925 100644 --- a/source/libs/executor/inc/executorimpl.h +++ b/source/libs/executor/inc/executorimpl.h @@ -365,7 +365,7 @@ typedef struct SStreamScanInfo { int32_t blockType; // current block type int32_t validBlockIndex; // Is current data has returned? uint64_t numOfExec; // execution times - void* streamReader;// stream block reader handle + STqReader* tqReader; int32_t tsArrayIndex; SArray* tsArray; @@ -383,6 +383,11 @@ typedef struct SStreamScanInfo { SSDataBlock* pPullDataRes; // pull data SSDataBlock SSDataBlock* pDeleteDataRes; // delete data SSDataBlock int32_t deleteDataIndex; + + // status for tmq + //SSchemaWrapper schema; + STqOffset offset; + } SStreamScanInfo; typedef struct SSysTableScanInfo { diff --git a/source/libs/executor/src/executor.c b/source/libs/executor/src/executor.c index 0155cdd416..a83565cbe0 100644 --- a/source/libs/executor/src/executor.c +++ b/source/libs/executor/src/executor.c @@ -45,7 +45,7 @@ static int32_t doSetStreamBlock(SOperatorInfo* pOperator, void* input, size_t nu pInfo->blockType = type; if (type == STREAM_INPUT__DATA_SUBMIT) { - if (tqReadHandleSetMsg(pInfo->streamReader, input, 0) < 0) { + if (tqReaderSetDataMsg(pInfo->tqReader, input, 0) < 0) { qError("submit msg messed up when initing stream block, %s" PRIx64, id); return TSDB_CODE_QRY_APP_ERROR; } @@ -105,7 +105,7 @@ int32_t qSetMultiStreamInput(qTaskInfo_t tinfo, const void* pBlocks, size_t numO return code; } -qTaskInfo_t qCreateStreamExecTaskInfo(void* msg, void* streamReadHandle) { +qTaskInfo_t qCreateStreamExecTaskInfo(void* msg, SReadHandle* readers) { if (msg == NULL) { return NULL; } @@ -120,7 +120,7 @@ qTaskInfo_t qCreateStreamExecTaskInfo(void* msg, void* streamReadHandle) { } qTaskInfo_t pTaskInfo = NULL; - code = qCreateExecTask(streamReadHandle, 0, 0, plan, &pTaskInfo, NULL, NULL, OPTR_EXEC_MODEL_STREAM); + code = qCreateExecTask(readers, 0, 0, plan, &pTaskInfo, NULL, NULL, OPTR_EXEC_MODEL_STREAM); if (code != TSDB_CODE_SUCCESS) { // TODO: destroy SSubplan & pTaskInfo terrno = code; @@ -174,11 +174,11 @@ int32_t qUpdateQualifiedTableId(qTaskInfo_t tinfo, const SArray* tableIdList, bo SArray* qa = filterQualifiedChildTables(pScanInfo, tableIdList); qDebug(" %d qualified child tables added into stream scanner", (int32_t)taosArrayGetSize(qa)); - code = tqReadHandleAddTbUidList(pScanInfo->streamReader, qa); + code = tqReaderAddTbUidList(pScanInfo->tqReader, qa); taosArrayDestroy(qa); } else { // remove the table id in current list qDebug(" %d remove child tables from the stream scanner", (int32_t)taosArrayGetSize(tableIdList)); - code = tqReadHandleRemoveTbUidList(pScanInfo->streamReader, tableIdList); + code = tqReaderRemoveTbUidList(pScanInfo->tqReader, tableIdList); } return code; diff --git a/source/libs/executor/src/executorMain.c b/source/libs/executor/src/executorMain.c index 7f84b4c174..d5e546f466 100644 --- a/source/libs/executor/src/executorMain.c +++ b/source/libs/executor/src/executorMain.c @@ -236,6 +236,37 @@ int32_t qDeserializeTaskStatus(qTaskInfo_t tinfo, const char* pInput, int32_t le return decodeOperator(pTaskInfo->pRoot, pInput, len); } +int32_t qExtractStreamScanner(qTaskInfo_t tinfo, void** scanner) { + SExecTaskInfo* pTaskInfo = (SExecTaskInfo*)tinfo; + SOperatorInfo* pOperator = pTaskInfo->pRoot; + + while (1) { + uint8_t type = pOperator->operatorType; + if (type == QUERY_NODE_PHYSICAL_PLAN_STREAM_SCAN) { + *scanner = pOperator->info; + return 0; + } else { + ASSERT(pOperator->numOfDownstream == 1); + pOperator = pOperator->pDownstream[0]; + } + } +} + +void* qExtractReaderFromStreamScanner(void* scanner) { + SStreamScanInfo* pInfo = scanner; + return (void*)pInfo->tqReader; +} + +const SSchemaWrapper* qExtractSchemaFromStreamScanner(void* scanner) { + SStreamScanInfo* pInfo = scanner; + return pInfo->tqReader->pSchemaWrapper; +} + +const STqOffset* qExtractStatusFromStreamScanner(void* scanner) { + SStreamScanInfo* pInfo = scanner; + return &pInfo->offset; +} + int32_t qStreamPrepareScan(qTaskInfo_t tinfo, uint64_t uid, int64_t ts) { SExecTaskInfo* pTaskInfo = (SExecTaskInfo*)tinfo; diff --git a/source/libs/executor/src/executorimpl.c b/source/libs/executor/src/executorimpl.c index 4c7b2c9102..581c3f179b 100644 --- a/source/libs/executor/src/executorimpl.c +++ b/source/libs/executor/src/executorimpl.c @@ -741,10 +741,10 @@ static int32_t doCreateConstantValColumnAggInfo(SInputColumnInfoData* pInput, SF if (type == TSDB_DATA_TYPE_BIGINT) { int64_t v = pFuncParam->param.i; - *da = (SColumnDataAgg){.numOfNull = 0, .min = v, .max = v, .maxIndex = 0, .minIndex = 0, .sum = v * numOfRows}; + *da = (SColumnDataAgg){.numOfNull = 0, .min = v, .max = v, .sum = v * numOfRows}; } else if (type == TSDB_DATA_TYPE_DOUBLE) { double v = pFuncParam->param.d; - *da = (SColumnDataAgg){.numOfNull = 0, .maxIndex = 0, .minIndex = 0}; + *da = (SColumnDataAgg){.numOfNull = 0}; *(double*)&da->min = v; *(double*)&da->max = v; @@ -752,7 +752,7 @@ static int32_t doCreateConstantValColumnAggInfo(SInputColumnInfoData* pInput, SF } else if (type == TSDB_DATA_TYPE_BOOL) { // todo validate this data type bool v = pFuncParam->param.i; - *da = (SColumnDataAgg){.numOfNull = 0, .maxIndex = 0, .minIndex = 0}; + *da = (SColumnDataAgg){.numOfNull = 0}; *(bool*)&da->min = 0; *(bool*)&da->max = v; *(bool*)&da->sum = v * numOfRows; @@ -1130,7 +1130,7 @@ int32_t loadDataBlockOnDemand(SExecTaskInfo* pTaskInfo, STableScanInfo* pTableSc } else if ((*status) == BLK_DATA_SMA_LOAD) { // this function never returns error? pCost->loadBlockStatis += 1; -// tsdbRetrieveDataBlockStatisInfo(pTableScanInfo->pTsdbReadHandle, &pBlock->pBlockAgg); +// tsdbRetrieveDatablockSMA(pTableScanInfo->pTsdbReadHandle, &pBlock->pBlockAgg); if (pBlock->pBlockAgg == NULL) { // data block statistics does not exist, load data block // pBlock->pDataBlock = tsdbRetrieveDataBlock(pTableScanInfo->pTsdbReadHandle, NULL); @@ -1141,7 +1141,7 @@ int32_t loadDataBlockOnDemand(SExecTaskInfo* pTaskInfo, STableScanInfo* pTableSc // load the data block statistics to perform further filter pCost->loadBlockStatis += 1; -// tsdbRetrieveDataBlockStatisInfo(pTableScanInfo->pTsdbReadHandle, &pBlock->pBlockAgg); +// tsdbRetrieveDatablockSMA(pTableScanInfo->pTsdbReadHandle, &pBlock->pBlockAgg); if (pQueryAttr->topBotQuery && pBlock->pBlockAgg != NULL) { { // set previous window @@ -2844,7 +2844,7 @@ int32_t getTableScanInfo(SOperatorInfo* pOperator, int32_t* order, int32_t* scan } int32_t doPrepareScan(SOperatorInfo* pOperator, uint64_t uid, int64_t ts) { - int32_t type = pOperator->operatorType; + uint8_t type = pOperator->operatorType; pOperator->status = OP_OPENED; @@ -4346,11 +4346,11 @@ SOperatorInfo* createOperatorTree(SPhysiNode* pPhyNode, SExecTaskInfo* pTaskInfo } else if (QUERY_NODE_PHYSICAL_PLAN_LAST_ROW_SCAN == type) { SLastRowScanPhysiNode* pScanNode = (SLastRowScanPhysiNode*)pPhyNode; -// int32_t code = createScanTableListInfo(pTableScanNode, pHandle, pTableListInfo, queryId, taskId); -// if (code) { -// pTaskInfo->code = code; -// return NULL; -// } + // int32_t code = createScanTableListInfo(pTableScanNode, pHandle, pTableListInfo, queryId, taskId); + // if (code) { + // pTaskInfo->code = code; + // return NULL; + // } int32_t code = extractTableSchemaVersion(pHandle, pScanNode->uid, pTaskInfo); if (code != TSDB_CODE_SUCCESS) { @@ -4407,8 +4407,8 @@ SOperatorInfo* createOperatorTree(SPhysiNode* pPhyNode, SExecTaskInfo* pTaskInfo pOptr = createGroupOperatorInfo(ops[0], pExprInfo, num, pResBlock, pColList, pAggNode->node.pConditions, pScalarExprInfo, numOfScalarExpr, pTaskInfo); } else { - pOptr = - createAggregateOperatorInfo(ops[0], pExprInfo, num, pResBlock, pAggNode->node.pConditions, pScalarExprInfo, numOfScalarExpr, pTaskInfo); + pOptr = createAggregateOperatorInfo(ops[0], pExprInfo, num, pResBlock, pAggNode->node.pConditions, + pScalarExprInfo, numOfScalarExpr, pTaskInfo); } } else if (QUERY_NODE_PHYSICAL_PLAN_HASH_INTERVAL == type || QUERY_NODE_PHYSICAL_PLAN_STREAM_INTERVAL == type) { SIntervalPhysiNode* pIntervalPhyNode = (SIntervalPhysiNode*)pPhyNode; @@ -4575,7 +4575,8 @@ SArray* extractColumnInfo(SNodeList* pNodeList) { return pList; } -STsdbReader* doCreateDataReader(STableScanPhysiNode* pTableScanNode, SReadHandle* pHandle, STableListInfo* pTableListInfo, const char* idstr) { +STsdbReader* doCreateDataReader(STableScanPhysiNode* pTableScanNode, SReadHandle* pHandle, + STableListInfo* pTableListInfo, const char* idstr) { int32_t code = getTableList(pHandle->meta, pHandle->vnode, &pTableScanNode->scan, pTableListInfo); if (code != TSDB_CODE_SUCCESS) { goto _error; @@ -4819,7 +4820,6 @@ int32_t createExecTaskInfoImpl(SSubplan* pPlan, SExecTaskInfo** pTaskInfo, SRead (*pTaskInfo)->pRoot = createOperatorTree(pPlan->pNode, *pTaskInfo, pHandle, queryId, taskId, &(*pTaskInfo)->tableqinfoList, pPlan->user); - if (NULL == (*pTaskInfo)->pRoot) { code = (*pTaskInfo)->code; goto _complete; diff --git a/source/libs/executor/src/scanoperator.c b/source/libs/executor/src/scanoperator.c index aae85bbbc1..e016344f2b 100644 --- a/source/libs/executor/src/scanoperator.c +++ b/source/libs/executor/src/scanoperator.c @@ -210,7 +210,10 @@ static int32_t loadDataBlock(SOperatorInfo* pOperator, STableScanInfo* pTableSca bool allColumnsHaveAgg = true; SColumnDataAgg** pColAgg = NULL; - tsdbRetrieveDataBlockStatisInfo(pTableScanInfo->dataReader, &pColAgg, &allColumnsHaveAgg); + int32_t code = tsdbRetrieveDatablockSMA(pTableScanInfo->dataReader, &pColAgg, &allColumnsHaveAgg); + if (code != TSDB_CODE_SUCCESS) { + longjmp(pTaskInfo->env, code); + } if (allColumnsHaveAgg == true) { int32_t numOfCols = taosArrayGetSize(pBlock->pDataBlock); @@ -1232,38 +1235,35 @@ static SSDataBlock* doStreamScan(SOperatorInfo* pOperator) { SDataBlockInfo* pBlockInfo = &pInfo->pRes->info; blockDataCleanup(pInfo->pRes); - while (tqNextDataBlock(pInfo->streamReader)) { + while (tqNextDataBlock(pInfo->tqReader)) { SSDataBlock block = {0}; // todo refactor - int32_t code = tqRetrieveDataBlock(&block, pInfo->streamReader); + int32_t code = tqRetrieveDataBlock(&block, pInfo->tqReader); - uint64_t groupId = block.info.groupId; - uint64_t uid = block.info.uid; - int32_t numOfRows = block.info.rows; - - if (code != TSDB_CODE_SUCCESS || numOfRows == 0) { + if (code != TSDB_CODE_SUCCESS || block.info.rows == 0) { pTaskInfo->code = code; return NULL; } - pInfo->pRes->info.groupId = groupId; - pInfo->pRes->info.rows = numOfRows; - pInfo->pRes->info.uid = uid; + pInfo->pRes->info.rows = block.info.rows; + pInfo->pRes->info.uid = block.info.uid; pInfo->pRes->info.type = STREAM_NORMAL; - pInfo->pRes->info.capacity = numOfRows; + pInfo->pRes->info.capacity = block.info.rows; + + + + uint64_t* groupIdPre = taosHashGet(pOperator->pTaskInfo->tableqinfoList.map, &block.info.uid, sizeof(int64_t)); + if (groupIdPre) { + pInfo->pRes->info.groupId = *groupIdPre; + } else { + pInfo->pRes->info.groupId = 0; + } // for generating rollup SMA result, each time is an independent time serie. // TODO temporarily used, when the statement of "partition by tbname" is ready, remove this if (pInfo->assignBlockUid) { - pInfo->pRes->info.groupId = uid; - } else { - pInfo->pRes->info.groupId = groupId; - } - - uint64_t* groupIdPre = taosHashGet(pOperator->pTaskInfo->tableqinfoList.map, &uid, sizeof(int64_t)); - if (groupIdPre) { - pInfo->pRes->info.groupId = *groupIdPre; + pInfo->pRes->info.groupId = block.info.uid; } // todo extract method @@ -1413,13 +1413,13 @@ SOperatorInfo* createStreamScanOperatorInfo(SReadHandle* pHandle, STableScanPhys } } - if (pHandle->initStreamReader) { - ASSERT(pHandle->streamReader == NULL); - pInfo->streamReader = tqInitSubmitMsgScanner(pHandle->meta); - ASSERT(pInfo->streamReader); + if (pHandle->initTqReader) { + ASSERT(pHandle->tqReader == NULL); + pInfo->tqReader = tqOpenReader(pHandle->vnode); + ASSERT(pInfo->tqReader); } else { - ASSERT(pHandle->streamReader); - pInfo->streamReader = pHandle->streamReader; + ASSERT(pHandle->tqReader); + pInfo->tqReader = pHandle->tqReader; } if (pSTInfo->interval.interval > 0) { @@ -1435,9 +1435,9 @@ SOperatorInfo* createStreamScanOperatorInfo(SReadHandle* pHandle, STableScanPhys pInfo->tableUid = pScanPhyNode->uid; // set the extract column id to streamHandle - tqReadHandleSetColIdList(pInfo->streamReader, pColIds); + tqReaderSetColIdList(pInfo->tqReader, pColIds); SArray* tableIdList = extractTableIdList(&pTaskInfo->tableqinfoList); - int32_t code = tqReadHandleSetTbUidList(pInfo->streamReader, tableIdList); + int32_t code = tqReaderSetTbUidList(pInfo->tqReader, tableIdList); if (code != 0) { taosArrayDestroy(tableIdList); goto _error; @@ -2352,7 +2352,7 @@ static int32_t loadDataBlockFromOneTable(SOperatorInfo* pOperator, STableMergeSc bool allColumnsHaveAgg = true; SColumnDataAgg** pColAgg = NULL; STsdbReader* reader = taosArrayGetP(pTableScanInfo->dataReaders, readerIdx); - tsdbRetrieveDataBlockStatisInfo(reader, &pColAgg, &allColumnsHaveAgg); + tsdbRetrieveDatablockSMA(reader, &pColAgg, &allColumnsHaveAgg); if (allColumnsHaveAgg == true) { int32_t numOfCols = taosArrayGetSize(pBlock->pDataBlock); diff --git a/source/libs/executor/src/timewindowoperator.c b/source/libs/executor/src/timewindowoperator.c index a89e166e8a..f2a9a13c52 100644 --- a/source/libs/executor/src/timewindowoperator.c +++ b/source/libs/executor/src/timewindowoperator.c @@ -1316,7 +1316,7 @@ bool doDeleteIntervalWindow(SAggSupporter* pAggSup, TSKEY ts, uint64_t groupId) // window has been closed return false; } - SFilePage* bufPage = getBufPage(pAggSup->pResultBuf, p1->pageId); + // SFilePage* bufPage = getBufPage(pAggSup->pResultBuf, p1->pageId); // dBufSetBufPageRecycled(pAggSup->pResultBuf, bufPage); taosHashRemove(pAggSup->pResultRowHashTable, pAggSup->keyBuf, GET_RES_WINDOW_KEY_LEN(bytes)); return true; @@ -1419,7 +1419,7 @@ static int32_t closeIntervalWindow(SHashObj* pHashMap, STimeWindowAggSupp* pSup, ASSERT(pRecyPages != NULL); taosArrayPush(pRecyPages, &pPos->pageId); } else { - SFilePage* bufPage = getBufPage(pDiscBuf, pPos->pageId); + // SFilePage* bufPage = getBufPage(pDiscBuf, pPos->pageId); // dBufSetBufPageRecycled(pDiscBuf, bufPage); } char keyBuf[GET_RES_WINDOW_KEY_LEN(sizeof(TSKEY))]; @@ -1446,7 +1446,7 @@ static void freeAllPages(SArray* pageIds, SDiskbasedBuf* pDiskBuf) { int32_t size = taosArrayGetSize(pageIds); for (int32_t i = 0; i < size; i++) { int32_t pageId = *(int32_t*)taosArrayGet(pageIds, i); - SFilePage* bufPage = getBufPage(pDiskBuf, pageId); + // SFilePage* bufPage = getBufPage(pDiskBuf, pageId); // dBufSetBufPageRecycled(pDiskBuf, bufPage); } taosArrayClear(pageIds); @@ -3189,7 +3189,7 @@ SResultWindowInfo* getCurSessionWindow(SStreamAggSupporter* pAggSup, TSKEY start *pIndex = index + 1; return pWin; } else if (endTs != INT64_MIN && isInWindow(pWin, endTs, gap)) { - *pIndex = index; + *pIndex = index + 1; return pWin; } } @@ -3442,7 +3442,7 @@ void deleteWindow(SArray* pWinInfos, int32_t index) { taosArrayRemove(pWinInfos, index); } -static void doDeleteSessionWindows(SStreamAggSupporter* pAggSup, SSDataBlock* pBlock, int64_t gap, SArray* result) { +static void doDeleteTimeWindows(SStreamAggSupporter* pAggSup, SSDataBlock* pBlock, int64_t gap, SArray* result) { SColumnInfoData* pStartTsCol = taosArrayGet(pBlock->pDataBlock, START_TS_COLUMN_INDEX); TSKEY* startDatas = (TSKEY*)pStartTsCol->pData; SColumnInfoData* pEndTsCol = taosArrayGet(pBlock->pDataBlock, END_TS_COLUMN_INDEX); @@ -3700,13 +3700,13 @@ static SSDataBlock* doStreamSessionAgg(SOperatorInfo* pOperator) { } else if (pBlock->info.type == STREAM_DELETE_DATA || pBlock->info.type == STREAM_DELETE_RESULT) { SArray* pWins = taosArrayInit(16, sizeof(SResultWindowInfo)); // gap must be 0 - doDeleteSessionWindows(&pInfo->streamAggSup, pBlock, 0, pWins); + doDeleteTimeWindows(&pInfo->streamAggSup, pBlock, 0, pWins); if (IS_FINAL_OP(pInfo)) { int32_t childIndex = getChildIndex(pBlock); SOperatorInfo* pChildOp = taosArrayGetP(pInfo->pChildren, childIndex); SStreamSessionAggOperatorInfo* pChildInfo = pChildOp->info; // gap must be 0 - doDeleteSessionWindows(&pChildInfo->streamAggSup, pBlock, 0, NULL); + doDeleteTimeWindows(&pChildInfo->streamAggSup, pBlock, 0, NULL); rebuildTimeWindow(pInfo, pWins, pBlock->info.groupId, pOperator->exprSupp.numOfExprs, pOperator); } copyDeleteWindowInfo(pWins, pInfo->pStDeleted); @@ -3840,7 +3840,7 @@ static SSDataBlock* doStreamSessionSemiAgg(SOperatorInfo* pOperator) { break; } else if (pBlock->info.type == STREAM_DELETE_DATA || pBlock->info.type == STREAM_DELETE_RESULT) { // gap must be 0 - doDeleteSessionWindows(&pInfo->streamAggSup, pBlock, 0, NULL); + doDeleteTimeWindows(&pInfo->streamAggSup, pBlock, 0, NULL); copyDataBlock(pInfo->pDelRes, pBlock); pInfo->pDelRes->info.type = STREAM_DELETE_RESULT; break; @@ -4232,6 +4232,12 @@ static SSDataBlock* doStreamStateAgg(SOperatorInfo* pOperator) { doClearStateWindows(&pInfo->streamAggSup, pBlock, pInfo->primaryTsIndex, &pInfo->stateCol, pInfo->stateCol.slotId, pSeUpdated, pInfo->pSeDeleted); continue; + } else if (pBlock->info.type == STREAM_DELETE_DATA) { + SArray* pWins = taosArrayInit(16, sizeof(SResultWindowInfo)); + doDeleteTimeWindows(&pInfo->streamAggSup, pBlock, 0, pWins); + copyDeleteWindowInfo(pWins, pInfo->pSeDeleted); + taosArrayDestroy(pWins); + continue; } else if (pBlock->info.type == STREAM_GET_ALL) { getAllSessionWindow(pInfo->streamAggSup.pResultRows, pUpdated, getResWinForState); continue; diff --git a/source/libs/function/src/builtinsimpl.c b/source/libs/function/src/builtinsimpl.c index f94522f0d8..d8fcca30c0 100644 --- a/source/libs/function/src/builtinsimpl.c +++ b/source/libs/function/src/builtinsimpl.c @@ -1080,6 +1080,19 @@ bool getMinmaxFuncEnv(SFunctionNode* UNUSED_PARAM(pFunc), SFuncExecEnv* pEnv) { static void saveTupleData(SqlFunctionCtx* pCtx, int32_t rowIndex, const SSDataBlock* pSrcBlock, STuplePos* pPos); static void copyTupleData(SqlFunctionCtx* pCtx, int32_t rowIndex, const SSDataBlock* pSrcBlock, STuplePos* pPos); +static int32_t findRowIndex(int32_t start, int32_t num, SColumnInfoData* pCol, const char* tval) { + // the data is loaded, not only the block SMA value + for(int32_t i = start; i < num + start; ++i) { + char* p = colDataGetData(pCol, i); + if (memcpy((void*)tval, p, pCol->info.bytes) == 0) { + return i; + } + } + + ASSERT(0); +} + + int32_t doMinMaxHelper(SqlFunctionCtx* pCtx, int32_t isMinFunc) { int32_t numOfElems = 0; @@ -1111,15 +1124,14 @@ int32_t doMinMaxHelper(SqlFunctionCtx* pCtx, int32_t isMinFunc) { if (isMinFunc) { tval = &pInput->pColumnDataAgg[0]->min; - index = pInput->pColumnDataAgg[0]->minIndex; } else { tval = &pInput->pColumnDataAgg[0]->max; - index = pInput->pColumnDataAgg[0]->maxIndex; } if (!pBuf->assign) { pBuf->v = *(int64_t*)tval; if (pCtx->subsidiaries.num > 0) { + index = findRowIndex(pInput->startRowIndex, pInput->numOfRows, pCol, tval); saveTupleData(pCtx, index, pCtx->pSrcBlock, &pBuf->tuplePos); } } else { @@ -1131,6 +1143,7 @@ int32_t doMinMaxHelper(SqlFunctionCtx* pCtx, int32_t isMinFunc) { if ((prev < val) ^ isMinFunc) { pBuf->v = val; if (pCtx->subsidiaries.num > 0) { + index = findRowIndex(pInput->startRowIndex, pInput->numOfRows, pCol, tval); saveTupleData(pCtx, index, pCtx->pSrcBlock, &pBuf->tuplePos); } } @@ -1143,6 +1156,7 @@ int32_t doMinMaxHelper(SqlFunctionCtx* pCtx, int32_t isMinFunc) { if ((prev < val) ^ isMinFunc) { pBuf->v = val; if (pCtx->subsidiaries.num > 0) { + index = findRowIndex(pInput->startRowIndex, pInput->numOfRows, pCol, tval); saveTupleData(pCtx, index, pCtx->pSrcBlock, &pBuf->tuplePos); } } @@ -1154,6 +1168,7 @@ int32_t doMinMaxHelper(SqlFunctionCtx* pCtx, int32_t isMinFunc) { if ((prev < val) ^ isMinFunc) { pBuf->v = val; if (pCtx->subsidiaries.num > 0) { + index = findRowIndex(pInput->startRowIndex, pInput->numOfRows, pCol, tval); saveTupleData(pCtx, index, pCtx->pSrcBlock, &pBuf->tuplePos); } } @@ -1167,6 +1182,7 @@ int32_t doMinMaxHelper(SqlFunctionCtx* pCtx, int32_t isMinFunc) { } if (pCtx->subsidiaries.num > 0) { + index = findRowIndex(pInput->startRowIndex, pInput->numOfRows, pCol, tval); saveTupleData(pCtx, index, pCtx->pSrcBlock, &pBuf->tuplePos); } } @@ -5547,30 +5563,18 @@ int32_t blockDistFinalize(SqlFunctionCtx* pCtx, SSDataBlock* pBlock) { } } - int32_t delta = maxVal - minVal; - int32_t step = delta / 50; - if (step == 0) { - step = 1; - } + // maximum number of step is 80 + double factor = pData->numOfBlocks / 80.0; int32_t numOfBuckets = sizeof(pData->blockRowsHisto) / sizeof(pData->blockRowsHisto[0]); - int32_t bucketRange = (pData->maxRows - pData->minRows) / numOfBuckets; - - bool singleModel = false; - if (bucketRange == 0) { - singleModel = true; - step = 20; - bucketRange = (pData->defMaxRows - pData->defMinRows) / numOfBuckets; - } + int32_t bucketRange = (pData->defMaxRows - pData->defMinRows) / numOfBuckets; for (int32_t i = 0; i < tListLen(pData->blockRowsHisto); ++i) { - len = sprintf(st + VARSTR_HEADER_SIZE, "%04d |", pData->defMinRows + bucketRange * (i + 1)); + len = sprintf(st + VARSTR_HEADER_SIZE, "%04d |", pData->defMinRows + bucketRange * i); int32_t num = 0; - if (singleModel && pData->blockRowsHisto[i] > 0) { - num = 20; - } else { - num = (pData->blockRowsHisto[i] + step - 1) / step; + if (pData->blockRowsHisto[i] > 0) { + num = (pData->blockRowsHisto[i]) / factor; } for (int32_t j = 0; j < num; ++j) { @@ -5578,9 +5582,10 @@ int32_t blockDistFinalize(SqlFunctionCtx* pCtx, SSDataBlock* pBlock) { len += x; } - double v = pData->blockRowsHisto[i] * 100.0 / pData->numOfBlocks; - len += sprintf(st + VARSTR_HEADER_SIZE + len, " %d (%.2f%c)", pData->blockRowsHisto[i], v, '%'); - printf("%s\n", st); + if (num > 0) { + double v = pData->blockRowsHisto[i] * 100.0 / pData->numOfBlocks; + len += sprintf(st + VARSTR_HEADER_SIZE + len, " %d (%.2f%c)", pData->blockRowsHisto[i], v, '%'); + } varDataSetLen(st, len); colDataAppend(pColInfo, row++, st, false); diff --git a/source/libs/sync/inc/syncInt.h b/source/libs/sync/inc/syncInt.h index 8936cd6ed9..9742f93824 100644 --- a/source/libs/sync/inc/syncInt.h +++ b/source/libs/sync/inc/syncInt.h @@ -67,7 +67,6 @@ typedef struct SSyncNode { char path[TSDB_FILENAME_LEN]; char raftStorePath[TSDB_FILENAME_LEN * 2]; char configPath[TSDB_FILENAME_LEN * 2]; - int32_t batchSize; // sync io SWal* pWal; diff --git a/source/libs/sync/inc/syncOnMessage.h b/source/libs/sync/inc/syncOnMessage.h deleted file mode 100644 index 2f8856e652..0000000000 --- a/source/libs/sync/inc/syncOnMessage.h +++ /dev/null @@ -1,72 +0,0 @@ -/* - * 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 . - */ - -#ifndef _TD_LIBS_SYNC_ON_MESSAGE_H -#define _TD_LIBS_SYNC_ON_MESSAGE_H - -#ifdef __cplusplus -extern "C" { -#endif - -#include -#include -#include -#include "taosdef.h" - -// TLA+ Spec -// Receive(m) == -// LET i == m.mdest -// j == m.msource -// IN \* Any RPC with a newer term causes the recipient to advance -// \* its term first. Responses with stale terms are ignored. -// \/ UpdateTerm(i, j, m) -// \/ /\ m.mtype = RequestVoteRequest -// /\ HandleRequestVoteRequest(i, j, m) -// \/ /\ m.mtype = RequestVoteResponse -// /\ \/ DropStaleResponse(i, j, m) -// \/ HandleRequestVoteResponse(i, j, m) -// \/ /\ m.mtype = AppendEntriesRequest -// /\ HandleAppendEntriesRequest(i, j, m) -// \/ /\ m.mtype = AppendEntriesResponse -// /\ \/ DropStaleResponse(i, j, m) -// \/ HandleAppendEntriesResponse(i, j, m) - -// DuplicateMessage(m) == -// /\ Send(m) -// /\ UNCHANGED <> - -// DropMessage(m) == -// /\ Discard(m) -// /\ UNCHANGED <> - -// Next == /\ \/ \E i \in Server : Restart(i) -// \/ \E i \in Server : Timeout(i) -// \/ \E i,j \in Server : RequestVote(i, j) -// \/ \E i \in Server : BecomeLeader(i) -// \/ \E i \in Server, v \in Value : ClientRequest(i, v) -// \/ \E i \in Server : AdvanceCommitIndex(i) -// \/ \E i,j \in Server : AppendEntries(i, j) -// \/ \E m \in DOMAIN messages : Receive(m) -// \/ \E m \in DOMAIN messages : DuplicateMessage(m) -// \/ \E m \in DOMAIN messages : DropMessage(m) -// \* History variable that tracks every log ever: -// /\ allLogs' = allLogs \cup {log[i] : i \in Server} -// - -#ifdef __cplusplus -} -#endif - -#endif /*_TD_LIBS_SYNC_ON_MESSAGE_H*/ diff --git a/source/libs/sync/src/syncAppendEntries.c b/source/libs/sync/src/syncAppendEntries.c index 885ab1acae..e43c1e4b33 100644 --- a/source/libs/sync/src/syncAppendEntries.c +++ b/source/libs/sync/src/syncAppendEntries.c @@ -162,6 +162,17 @@ int32_t syncNodeOnAppendEntriesCb(SSyncNode* ths, SyncAppendEntries* pMsg) { pReply->success = false; pReply->matchIndex = SYNC_INDEX_INVALID; + // msg event log + do { + char host[128]; + uint16_t port; + syncUtilU642Addr(pReply->destId.addr, host, sizeof(host), &port); + sDebug( + "vgId:%d, send sync-append-entries-reply to %s:%d, {term:%lu, pterm:%lu, success:%d, " + "match-index:%ld}", + ths->vgId, host, port, pReply->term, pReply->privateTerm, pReply->success, pReply->matchIndex); + } while (0); + SRpcMsg rpcMsg; syncAppendEntriesReply2RpcMsg(pReply, &rpcMsg); syncNodeSendMsgById(&pReply->destId, ths, &rpcMsg); @@ -334,270 +345,16 @@ int32_t syncNodeOnAppendEntriesCb(SSyncNode* ths, SyncAppendEntries* pMsg) { pReply->matchIndex = pMsg->prevLogIndex; } - SRpcMsg rpcMsg; - syncAppendEntriesReply2RpcMsg(pReply, &rpcMsg); - syncNodeSendMsgById(&pReply->destId, ths, &rpcMsg); - syncAppendEntriesReplyDestroy(pReply); - - // maybe update commit index from leader - if (pMsg->commitIndex > ths->commitIndex) { - // has commit entry in local - if (pMsg->commitIndex <= ths->pLogStore->getLastIndex(ths->pLogStore)) { - SyncIndex beginIndex = ths->commitIndex + 1; - SyncIndex endIndex = pMsg->commitIndex; - - // update commit index - ths->commitIndex = pMsg->commitIndex; - - // call back Wal - ths->pLogStore->updateCommitIndex(ths->pLogStore, ths->commitIndex); - - int32_t code = syncNodeCommit(ths, beginIndex, endIndex, ths->state); - ASSERT(code == 0); - } - } - } - - return ret; -} - -#if 0 - -int32_t syncNodeOnAppendEntriesCb(SSyncNode* ths, SyncAppendEntries* pMsg) { - int32_t ret = 0; - - char logBuf[128] = {0}; - snprintf(logBuf, sizeof(logBuf), "==syncNodeOnAppendEntriesCb== term:%lu", ths->pRaftStore->currentTerm); - syncAppendEntriesLog2(logBuf, pMsg); - - if (pMsg->term > ths->pRaftStore->currentTerm) { - syncNodeUpdateTerm(ths, pMsg->term); - } - ASSERT(pMsg->term <= ths->pRaftStore->currentTerm); - - // reset elect timer - if (pMsg->term == ths->pRaftStore->currentTerm) { - ths->leaderCache = pMsg->srcId; - syncNodeResetElectTimer(ths); - } - ASSERT(pMsg->dataLen >= 0); - - SyncTerm localPreLogTerm = 0; - if (pMsg->prevLogIndex >= SYNC_INDEX_BEGIN && pMsg->prevLogIndex <= ths->pLogStore->getLastIndex(ths->pLogStore)) { - SSyncRaftEntry* pEntry = ths->pLogStore->getEntry(ths->pLogStore, pMsg->prevLogIndex); - if (pEntry == NULL) { - char logBuf[128]; - snprintf(logBuf, sizeof(logBuf), "getEntry error, index:%ld, since %s", pMsg->prevLogIndex, terrstr()); - syncNodeErrorLog(ths, logBuf); - return -1; - } - - localPreLogTerm = pEntry->term; - syncEntryDestory(pEntry); - } - - bool logOK = - (pMsg->prevLogIndex == SYNC_INDEX_INVALID) || - ((pMsg->prevLogIndex >= SYNC_INDEX_BEGIN) && - (pMsg->prevLogIndex <= ths->pLogStore->getLastIndex(ths->pLogStore)) && (pMsg->prevLogTerm == localPreLogTerm)); - - // reject request - if ((pMsg->term < ths->pRaftStore->currentTerm) || - ((pMsg->term == ths->pRaftStore->currentTerm) && (ths->state == TAOS_SYNC_STATE_FOLLOWER) && !logOK)) { - sTrace( - "syncNodeOnAppendEntriesCb --> reject, pMsg->term:%lu, ths->pRaftStore->currentTerm:%lu, ths->state:%d, " - "logOK:%d", - pMsg->term, ths->pRaftStore->currentTerm, ths->state, logOK); - - SyncAppendEntriesReply* pReply = syncAppendEntriesReplyBuild(ths->vgId); - pReply->srcId = ths->myRaftId; - pReply->destId = pMsg->srcId; - pReply->term = ths->pRaftStore->currentTerm; - pReply->success = false; - pReply->matchIndex = SYNC_INDEX_INVALID; - - SRpcMsg rpcMsg; - syncAppendEntriesReply2RpcMsg(pReply, &rpcMsg); - syncNodeSendMsgById(&pReply->destId, ths, &rpcMsg); - syncAppendEntriesReplyDestroy(pReply); - - return ret; - } - - // return to follower state - if (pMsg->term == ths->pRaftStore->currentTerm && ths->state == TAOS_SYNC_STATE_CANDIDATE) { - sTrace( - "syncNodeOnAppendEntriesCb --> return to follower, pMsg->term:%lu, ths->pRaftStore->currentTerm:%lu, " - "ths->state:%d, logOK:%d", - pMsg->term, ths->pRaftStore->currentTerm, ths->state, logOK); - - syncNodeBecomeFollower(ths, "from candidate by append entries"); - - // ret or reply? - return ret; - } - - // accept request - if (pMsg->term == ths->pRaftStore->currentTerm && ths->state == TAOS_SYNC_STATE_FOLLOWER && logOK) { - // preIndex = -1, or has preIndex entry in local log - ASSERT(pMsg->prevLogIndex <= ths->pLogStore->getLastIndex(ths->pLogStore)); - - // has extra entries (> preIndex) in local log - bool hasExtraEntries = pMsg->prevLogIndex < ths->pLogStore->getLastIndex(ths->pLogStore); - - // has entries in SyncAppendEntries msg - bool hasAppendEntries = pMsg->dataLen > 0; - - sTrace( - "syncNodeOnAppendEntriesCb --> accept, pMsg->term:%lu, ths->pRaftStore->currentTerm:%lu, ths->state:%d, " - "logOK:%d, hasExtraEntries:%d, hasAppendEntries:%d", - pMsg->term, ths->pRaftStore->currentTerm, ths->state, logOK, hasExtraEntries, hasAppendEntries); - - if (hasExtraEntries && hasAppendEntries) { - // not conflict by default - bool conflict = false; - - SyncIndex extraIndex = pMsg->prevLogIndex + 1; - SSyncRaftEntry* pExtraEntry = ths->pLogStore->getEntry(ths->pLogStore, extraIndex); - if (pExtraEntry == NULL) { - char logBuf[128]; - snprintf(logBuf, sizeof(logBuf), "getEntry error2, index:%ld, since %s", extraIndex, terrstr()); - syncNodeErrorLog(ths, logBuf); - return -1; - } - - SSyncRaftEntry* pAppendEntry = syncEntryDeserialize(pMsg->data, pMsg->dataLen); - if (pAppendEntry == NULL) { - syncNodeErrorLog(ths, "syncEntryDeserialize pAppendEntry error"); - return -1; - } - - // log not match, conflict - ASSERT(extraIndex == pAppendEntry->index); - if (pExtraEntry->term != pAppendEntry->term) { - conflict = true; - } - - if (conflict) { - // roll back - SyncIndex delBegin = ths->pLogStore->getLastIndex(ths->pLogStore); - SyncIndex delEnd = extraIndex; - - sTrace("syncNodeOnAppendEntriesCb --> conflict:%d, delBegin:%ld, delEnd:%ld", conflict, delBegin, delEnd); - - // notice! reverse roll back! - for (SyncIndex index = delEnd; index >= delBegin; --index) { - if (ths->pFsm->FpRollBackCb != NULL) { - SSyncRaftEntry* pRollBackEntry = ths->pLogStore->getEntry(ths->pLogStore, index); - if (pRollBackEntry == NULL) { - char logBuf[128]; - snprintf(logBuf, sizeof(logBuf), "getEntry error3, index:%ld, since %s", index, terrstr()); - syncNodeErrorLog(ths, logBuf); - return -1; - } - - // if (pRollBackEntry->msgType != TDMT_SYNC_NOOP) { - if (syncUtilUserRollback(pRollBackEntry->msgType)) { - SRpcMsg rpcMsg; - syncEntry2OriginalRpc(pRollBackEntry, &rpcMsg); - - SFsmCbMeta cbMeta = {0}; - cbMeta.index = pRollBackEntry->index; - cbMeta.lastConfigIndex = syncNodeGetSnapshotConfigIndex(ths, cbMeta.index); - cbMeta.isWeak = pRollBackEntry->isWeak; - cbMeta.code = 0; - cbMeta.state = ths->state; - cbMeta.seqNum = pRollBackEntry->seqNum; - ths->pFsm->FpRollBackCb(ths->pFsm, &rpcMsg, cbMeta); - rpcFreeCont(rpcMsg.pCont); - } - - syncEntryDestory(pRollBackEntry); - } - } - - // delete confict entries - ths->pLogStore->truncate(ths->pLogStore, extraIndex); - - // append new entries - ths->pLogStore->appendEntry(ths->pLogStore, pAppendEntry); - - // pre commit - SRpcMsg rpcMsg; - syncEntry2OriginalRpc(pAppendEntry, &rpcMsg); - if (ths->pFsm != NULL) { - // if (ths->pFsm->FpPreCommitCb != NULL && pAppendEntry->originalRpcType != TDMT_SYNC_NOOP) { - if (ths->pFsm->FpPreCommitCb != NULL && syncUtilUserPreCommit(pAppendEntry->originalRpcType)) { - SFsmCbMeta cbMeta = {0}; - cbMeta.index = pAppendEntry->index; - cbMeta.lastConfigIndex = syncNodeGetSnapshotConfigIndex(ths, cbMeta.index); - cbMeta.isWeak = pAppendEntry->isWeak; - cbMeta.code = 2; - cbMeta.state = ths->state; - cbMeta.seqNum = pAppendEntry->seqNum; - ths->pFsm->FpPreCommitCb(ths->pFsm, &rpcMsg, cbMeta); - } - } - rpcFreeCont(rpcMsg.pCont); - } - - // free memory - syncEntryDestory(pExtraEntry); - syncEntryDestory(pAppendEntry); - - } else if (hasExtraEntries && !hasAppendEntries) { - // do nothing - - } else if (!hasExtraEntries && hasAppendEntries) { - SSyncRaftEntry* pAppendEntry = syncEntryDeserialize(pMsg->data, pMsg->dataLen); - if (pAppendEntry == NULL) { - syncNodeErrorLog(ths, "syncEntryDeserialize pAppendEntry2 error"); - return -1; - } - - // append new entries - ths->pLogStore->appendEntry(ths->pLogStore, pAppendEntry); - - // pre commit - SRpcMsg rpcMsg; - syncEntry2OriginalRpc(pAppendEntry, &rpcMsg); - if (ths->pFsm != NULL) { - // if (ths->pFsm->FpPreCommitCb != NULL && pAppendEntry->originalRpcType != TDMT_SYNC_NOOP) { - if (ths->pFsm->FpPreCommitCb != NULL && syncUtilUserPreCommit(pAppendEntry->originalRpcType)) { - SFsmCbMeta cbMeta = {0}; - cbMeta.index = pAppendEntry->index; - cbMeta.lastConfigIndex = syncNodeGetSnapshotConfigIndex(ths, cbMeta.index); - cbMeta.isWeak = pAppendEntry->isWeak; - cbMeta.code = 3; - cbMeta.state = ths->state; - cbMeta.seqNum = pAppendEntry->seqNum; - ths->pFsm->FpPreCommitCb(ths->pFsm, &rpcMsg, cbMeta); - } - } - rpcFreeCont(rpcMsg.pCont); - - // free memory - syncEntryDestory(pAppendEntry); - - } else if (!hasExtraEntries && !hasAppendEntries) { - // do nothing - - } else { - syncNodeLog3("", ths); - ASSERT(0); - } - - SyncAppendEntriesReply* pReply = syncAppendEntriesReplyBuild(ths->vgId); - pReply->srcId = ths->myRaftId; - pReply->destId = pMsg->srcId; - pReply->term = ths->pRaftStore->currentTerm; - pReply->success = true; - - if (hasAppendEntries) { - pReply->matchIndex = pMsg->prevLogIndex + 1; - } else { - pReply->matchIndex = pMsg->prevLogIndex; - } + // msg event log + do { + char host[128]; + uint16_t port; + syncUtilU642Addr(pReply->destId.addr, host, sizeof(host), &port); + sDebug( + "vgId:%d, send sync-append-entries-reply to %s:%d, {term:%lu, pterm:%lu, success:%d, " + "match-index:%ld}", + ths->vgId, host, port, pReply->term, pReply->privateTerm, pReply->success, pReply->matchIndex); + } while (0); SRpcMsg rpcMsg; syncAppendEntriesReply2RpcMsg(pReply, &rpcMsg); @@ -626,8 +383,6 @@ int32_t syncNodeOnAppendEntriesCb(SSyncNode* ths, SyncAppendEntries* pMsg) { return ret; } -#endif - static int32_t syncNodeMakeLogSame(SSyncNode* ths, SyncAppendEntries* pMsg) { int32_t code; @@ -842,8 +597,8 @@ int32_t syncNodeOnAppendEntriesSnapshot2Cb(SSyncNode* ths, SyncAppendEntriesBatc do { char logBuf[128]; snprintf(logBuf, sizeof(logBuf), - "recv sync-append-entries-batch, fake match2, pre-index:%ld, pre-term:%lu, datalen:%d", - pMsg->prevLogIndex, pMsg->prevLogTerm, pMsg->dataLen); + "recv sync-append-entries-batch, fake match2, {pre-index:%ld, pre-term:%lu, datalen:%d, datacount:%d}", + pMsg->prevLogIndex, pMsg->prevLogTerm, pMsg->dataLen, pMsg->dataCount); syncNodeEventLog(ths, logBuf); } while (0); @@ -876,7 +631,7 @@ int32_t syncNodeOnAppendEntriesSnapshot2Cb(SSyncNode* ths, SyncAppendEntriesBatc code = syncNodePreCommit(ths, pAppendEntry); ASSERT(code == 0); - syncEntryDestory(pAppendEntry); + // syncEntryDestory(pAppendEntry); } // fsync once @@ -897,6 +652,17 @@ int32_t syncNodeOnAppendEntriesSnapshot2Cb(SSyncNode* ths, SyncAppendEntriesBatc pReply->success = true; pReply->matchIndex = matchIndex; + // msg event log + do { + char host[128]; + uint16_t port; + syncUtilU642Addr(pReply->destId.addr, host, sizeof(host), &port); + sDebug( + "vgId:%d, send sync-append-entries-reply to %s:%d, {term:%lu, pterm:%lu, success:%d, " + "match-index:%ld}", + ths->vgId, host, port, pReply->term, pReply->privateTerm, pReply->success, pReply->matchIndex); + } while (0); + // send response SRpcMsg rpcMsg; syncAppendEntriesReply2RpcMsg(pReply, &rpcMsg); @@ -931,8 +697,8 @@ int32_t syncNodeOnAppendEntriesSnapshot2Cb(SSyncNode* ths, SyncAppendEntriesBatc do { char logBuf[128]; snprintf(logBuf, sizeof(logBuf), - "recv sync-append-entries-batch, not match, pre-index:%ld, pre-term:%lu, datalen:%d", - pMsg->prevLogIndex, pMsg->prevLogTerm, pMsg->dataLen); + "recv sync-append-entries-batch, not match, {pre-index:%ld, pre-term:%lu, datalen:%d, datacount:%d}", + pMsg->prevLogIndex, pMsg->prevLogTerm, pMsg->dataLen, pMsg->dataCount); syncNodeEventLog(ths, logBuf); } while (0); @@ -945,6 +711,17 @@ int32_t syncNodeOnAppendEntriesSnapshot2Cb(SSyncNode* ths, SyncAppendEntriesBatc pReply->success = false; pReply->matchIndex = SYNC_INDEX_INVALID; + // msg event log + do { + char host[128]; + uint16_t port; + syncUtilU642Addr(pReply->destId.addr, host, sizeof(host), &port); + sDebug( + "vgId:%d, send sync-append-entries-reply to %s:%d, {term:%lu, pterm:%lu, success:%d, " + "match-index:%ld}", + ths->vgId, host, port, pReply->term, pReply->privateTerm, pReply->success, pReply->matchIndex); + } while (0); + // send response SRpcMsg rpcMsg; syncAppendEntriesReply2RpcMsg(pReply, &rpcMsg); @@ -976,8 +753,9 @@ int32_t syncNodeOnAppendEntriesSnapshot2Cb(SSyncNode* ths, SyncAppendEntriesBatc do { char logBuf[128]; - snprintf(logBuf, sizeof(logBuf), "recv sync-append-entries, match, pre-index:%ld, pre-term:%lu, datalen:%d", - pMsg->prevLogIndex, pMsg->prevLogTerm, pMsg->dataLen); + snprintf(logBuf, sizeof(logBuf), + "recv sync-append-entries-batch, match, {pre-index:%ld, pre-term:%lu, datalen:%d, datacount:%d}", + pMsg->prevLogIndex, pMsg->prevLogTerm, pMsg->dataLen, pMsg->dataCount); syncNodeEventLog(ths, logBuf); } while (0); @@ -999,7 +777,7 @@ int32_t syncNodeOnAppendEntriesSnapshot2Cb(SSyncNode* ths, SyncAppendEntriesBatc code = syncNodePreCommit(ths, pAppendEntry); ASSERT(code == 0); - syncEntryDestory(pAppendEntry); + // syncEntryDestory(pAppendEntry); } // fsync once @@ -1017,6 +795,17 @@ int32_t syncNodeOnAppendEntriesSnapshot2Cb(SSyncNode* ths, SyncAppendEntriesBatc pReply->success = true; pReply->matchIndex = hasAppendEntries ? pMsg->prevLogIndex + pMsg->dataCount : pMsg->prevLogIndex; + // msg event log + do { + char host[128]; + uint16_t port; + syncUtilU642Addr(pReply->destId.addr, host, sizeof(host), &port); + sDebug( + "vgId:%d, send sync-append-entries-reply to %s:%d, {term:%lu, pterm:%lu, success:%d, " + "match-index:%ld}", + ths->vgId, host, port, pReply->term, pReply->privateTerm, pReply->success, pReply->matchIndex); + } while (0); + // send response SRpcMsg rpcMsg; syncAppendEntriesReply2RpcMsg(pReply, &rpcMsg); @@ -1226,6 +1015,17 @@ int32_t syncNodeOnAppendEntriesSnapshotCb(SSyncNode* ths, SyncAppendEntries* pMs pReply->success = true; pReply->matchIndex = matchIndex; + // msg event log + do { + char host[128]; + uint16_t port; + syncUtilU642Addr(pReply->destId.addr, host, sizeof(host), &port); + sDebug( + "vgId:%d, send sync-append-entries-reply to %s:%d, {term:%lu, pterm:%lu, success:%d, " + "match-index:%ld}", + ths->vgId, host, port, pReply->term, pReply->privateTerm, pReply->success, pReply->matchIndex); + } while (0); + // send response SRpcMsg rpcMsg; syncAppendEntriesReply2RpcMsg(pReply, &rpcMsg); @@ -1271,6 +1071,17 @@ int32_t syncNodeOnAppendEntriesSnapshotCb(SSyncNode* ths, SyncAppendEntries* pMs pReply->success = false; pReply->matchIndex = SYNC_INDEX_INVALID; + // msg event log + do { + char host[128]; + uint16_t port; + syncUtilU642Addr(pReply->destId.addr, host, sizeof(host), &port); + sDebug( + "vgId:%d, send sync-append-entries-reply to %s:%d, {term:%lu, pterm:%lu, success:%d, " + "match-index:%ld}", + ths->vgId, host, port, pReply->term, pReply->privateTerm, pReply->success, pReply->matchIndex); + } while (0); + // send response SRpcMsg rpcMsg; syncAppendEntriesReply2RpcMsg(pReply, &rpcMsg); @@ -1336,6 +1147,17 @@ int32_t syncNodeOnAppendEntriesSnapshotCb(SSyncNode* ths, SyncAppendEntries* pMs pReply->success = true; pReply->matchIndex = hasAppendEntries ? pMsg->prevLogIndex + 1 : pMsg->prevLogIndex; + // msg event log + do { + char host[128]; + uint16_t port; + syncUtilU642Addr(pReply->destId.addr, host, sizeof(host), &port); + sDebug( + "vgId:%d, send sync-append-entries-reply to %s:%d, {term:%lu, pterm:%lu, success:%d, " + "match-index:%ld}", + ths->vgId, host, port, pReply->term, pReply->privateTerm, pReply->success, pReply->matchIndex); + } while (0); + // send response SRpcMsg rpcMsg; syncAppendEntriesReply2RpcMsg(pReply, &rpcMsg); diff --git a/source/libs/sync/src/syncAppendEntriesReply.c b/source/libs/sync/src/syncAppendEntriesReply.c index e4724fc999..3f19b6759e 100644 --- a/source/libs/sync/src/syncAppendEntriesReply.c +++ b/source/libs/sync/src/syncAppendEntriesReply.c @@ -174,8 +174,12 @@ int32_t syncNodeOnAppendEntriesReplySnapshot2Cb(SSyncNode* ths, SyncAppendEntrie SyncIndex newNextIndex = pMsg->matchIndex + 1; SyncIndex newMatchIndex = pMsg->matchIndex; - if (ths->pLogStore->syncLogExist(ths->pLogStore, newNextIndex) && - ths->pLogStore->syncLogExist(ths->pLogStore, newNextIndex - 1)) { + bool needStartSnapshot = false; + if (newMatchIndex >= SYNC_INDEX_BEGIN && !ths->pLogStore->syncLogExist(ths->pLogStore, newMatchIndex)) { + needStartSnapshot = true; + } + + if (!needStartSnapshot) { // update next-index, match-index syncIndexMgrSetIndex(ths->pNextIndex, &(pMsg->srcId), newNextIndex); syncIndexMgrSetIndex(ths->pMatchIndex, &(pMsg->srcId), newMatchIndex); @@ -197,15 +201,36 @@ int32_t syncNodeOnAppendEntriesReplySnapshot2Cb(SSyncNode* ths, SyncAppendEntrie syncIndexMgrSetIndex(ths->pMatchIndex, &(pMsg->srcId), newMatchIndex); } + // event log, update next-index + do { + char host[64]; + int16_t port; + syncUtilU642Addr(pMsg->srcId.addr, host, sizeof(host), &port); + + char logBuf[256]; + snprintf(logBuf, sizeof(logBuf), "reset next-index:%ld, match-index:%ld for %s:%d", newNextIndex, newMatchIndex, + host, port); + syncNodeEventLog(ths, logBuf); + + } while (0); + } else { SyncIndex nextIndex = syncIndexMgrGetIndex(ths->pNextIndex, &(pMsg->srcId)); if (nextIndex > SYNC_INDEX_BEGIN) { --nextIndex; - if (ths->pLogStore->syncLogExist(ths->pLogStore, nextIndex) && - ths->pLogStore->syncLogExist(ths->pLogStore, nextIndex - 1)) { + bool needStartSnapshot = false; + if (nextIndex >= SYNC_INDEX_BEGIN && !ths->pLogStore->syncLogExist(ths->pLogStore, nextIndex)) { + needStartSnapshot = true; + } + if (nextIndex - 1 >= SYNC_INDEX_BEGIN && !ths->pLogStore->syncLogExist(ths->pLogStore, nextIndex - 1)) { + needStartSnapshot = true; + } + + if (!needStartSnapshot) { // do nothing + } else { SSyncRaftEntry* pEntry; int32_t code = ths->pLogStore->syncLogGetEntry(ths->pLogStore, nextIndex, &pEntry); @@ -227,6 +252,21 @@ int32_t syncNodeOnAppendEntriesReplySnapshot2Cb(SSyncNode* ths, SyncAppendEntrie nextIndex = SYNC_INDEX_BEGIN; } syncIndexMgrSetIndex(ths->pNextIndex, &(pMsg->srcId), nextIndex); + + // event log, update next-index + do { + char host[64]; + int16_t port; + syncUtilU642Addr(pMsg->srcId.addr, host, sizeof(host), &port); + + SyncIndex newNextIndex = nextIndex; + SyncIndex newMatchIndex = syncIndexMgrGetIndex(ths->pMatchIndex, &(pMsg->srcId)); + char logBuf[256]; + snprintf(logBuf, sizeof(logBuf), "reset2 next-index:%ld, match-index:%ld for %s:%d", newNextIndex, newMatchIndex, + host, port); + syncNodeEventLog(ths, logBuf); + + } while (0); } return 0; diff --git a/source/libs/sync/src/syncMain.c b/source/libs/sync/src/syncMain.c index 56fc6594ab..7fa13c9dc8 100644 --- a/source/libs/sync/src/syncMain.c +++ b/source/libs/sync/src/syncMain.c @@ -1311,8 +1311,10 @@ int32_t syncNodeSendMsgById(const SRaftId* destRaftId, SSyncNode* pSyncNode, SRp pMsg->info.noResp = 1; pSyncNode->FpSendMsg(&epSet, pMsg); } else { - sTrace("syncNodeSendMsgById pSyncNode->FpSendMsg is NULL"); + sError("vgId:%d, sync send msg by id error, fp-send-msg is null", pSyncNode->vgId); + return -1; } + return 0; } @@ -1326,7 +1328,7 @@ int32_t syncNodeSendMsgByInfo(const SNodeInfo* nodeInfo, SSyncNode* pSyncNode, S pMsg->info.noResp = 1; pSyncNode->FpSendMsg(&epSet, pMsg); } else { - sTrace("syncNodeSendMsgByInfo pSyncNode->FpSendMsg is NULL"); + sError("vgId:%d, sync send msg by info error, fp-send-msg is null", pSyncNode->vgId); } return 0; } @@ -1526,12 +1528,14 @@ void syncNodeEventLog(const SSyncNode* pSyncNode, char* str) { if (pSyncNode != NULL && pSyncNode->pRaftCfg != NULL && pSyncNode->pRaftStore != NULL) { snprintf(logBuf, sizeof(logBuf), "vgId:%d, sync %s %s, term:%lu, commit:%ld, beginlog:%ld, lastlog:%ld, lastsnapshot:%ld, standby:%d, " + "strategy:%d, batch:%d, " "replica-num:%d, " "lconfig:%ld, changing:%d, restore:%d, %s", pSyncNode->vgId, syncUtilState2String(pSyncNode->state), str, pSyncNode->pRaftStore->currentTerm, pSyncNode->commitIndex, logBeginIndex, logLastIndex, snapshot.lastApplyIndex, - pSyncNode->pRaftCfg->isStandBy, pSyncNode->replicaNum, pSyncNode->pRaftCfg->lastConfigIndex, - pSyncNode->changing, pSyncNode->restoreFinish, printStr); + pSyncNode->pRaftCfg->isStandBy, pSyncNode->pRaftCfg->snapshotStrategy, pSyncNode->pRaftCfg->batchSize, + pSyncNode->replicaNum, pSyncNode->pRaftCfg->lastConfigIndex, pSyncNode->changing, + pSyncNode->restoreFinish, printStr); } else { snprintf(logBuf, sizeof(logBuf), "%s", str); } @@ -1543,12 +1547,14 @@ void syncNodeEventLog(const SSyncNode* pSyncNode, char* str) { if (pSyncNode != NULL && pSyncNode->pRaftCfg != NULL && pSyncNode->pRaftStore != NULL) { snprintf(s, len, "vgId:%d, sync %s %s, term:%lu, commit:%ld, beginlog:%ld, lastlog:%ld, lastsnapshot:%ld, standby:%d, " + "strategy:%d, batch:%d, " "replica-num:%d, " "lconfig:%ld, changing:%d, restore:%d, %s", pSyncNode->vgId, syncUtilState2String(pSyncNode->state), str, pSyncNode->pRaftStore->currentTerm, pSyncNode->commitIndex, logBeginIndex, logLastIndex, snapshot.lastApplyIndex, - pSyncNode->pRaftCfg->isStandBy, pSyncNode->replicaNum, pSyncNode->pRaftCfg->lastConfigIndex, - pSyncNode->changing, pSyncNode->restoreFinish, printStr); + pSyncNode->pRaftCfg->isStandBy, pSyncNode->pRaftCfg->snapshotStrategy, pSyncNode->pRaftCfg->batchSize, + pSyncNode->replicaNum, pSyncNode->pRaftCfg->lastConfigIndex, pSyncNode->changing, + pSyncNode->restoreFinish, printStr); } else { snprintf(s, len, "%s", str); } diff --git a/source/libs/sync/src/syncMessage.c b/source/libs/sync/src/syncMessage.c index e53b9d5e36..697c25f304 100644 --- a/source/libs/sync/src/syncMessage.c +++ b/source/libs/sync/src/syncMessage.c @@ -1605,7 +1605,7 @@ void syncAppendEntriesLog2(char* s, const SyncAppendEntries* pMsg) { SyncAppendEntriesBatch* syncAppendEntriesBatchBuild(SSyncRaftEntry** entryPArr, int32_t arrSize, int32_t vgId) { ASSERT(entryPArr != NULL); - ASSERT(arrSize > 0); + ASSERT(arrSize >= 0); int32_t dataLen = 0; int32_t metaArrayLen = sizeof(SOffsetAndContLen) * arrSize; // diff --git a/source/libs/sync/src/syncOnMessage.c b/source/libs/sync/src/syncOnMessage.c deleted file mode 100644 index ce8bed9cd3..0000000000 --- a/source/libs/sync/src/syncOnMessage.c +++ /dev/null @@ -1,56 +0,0 @@ -/* - * 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 . - */ - -#include "syncOnMessage.h" - -// TLA+ Spec -// Receive(m) == -// LET i == m.mdest -// j == m.msource -// IN \* Any RPC with a newer term causes the recipient to advance -// \* its term first. Responses with stale terms are ignored. -// \/ UpdateTerm(i, j, m) -// \/ /\ m.mtype = RequestVoteRequest -// /\ HandleRequestVoteRequest(i, j, m) -// \/ /\ m.mtype = RequestVoteResponse -// /\ \/ DropStaleResponse(i, j, m) -// \/ HandleRequestVoteResponse(i, j, m) -// \/ /\ m.mtype = AppendEntriesRequest -// /\ HandleAppendEntriesRequest(i, j, m) -// \/ /\ m.mtype = AppendEntriesResponse -// /\ \/ DropStaleResponse(i, j, m) -// \/ HandleAppendEntriesResponse(i, j, m) - -// DuplicateMessage(m) == -// /\ Send(m) -// /\ UNCHANGED <> - -// DropMessage(m) == -// /\ Discard(m) -// /\ UNCHANGED <> - -// Next == /\ \/ \E i \in Server : Restart(i) -// \/ \E i \in Server : Timeout(i) -// \/ \E i,j \in Server : RequestVote(i, j) -// \/ \E i \in Server : BecomeLeader(i) -// \/ \E i \in Server, v \in Value : ClientRequest(i, v) -// \/ \E i \in Server : AdvanceCommitIndex(i) -// \/ \E i,j \in Server : AppendEntries(i, j) -// \/ \E m \in DOMAIN messages : Receive(m) -// \/ \E m \in DOMAIN messages : DuplicateMessage(m) -// \/ \E m \in DOMAIN messages : DropMessage(m) -// \* History variable that tracks every log ever: -// /\ allLogs' = allLogs \cup {log[i] : i \in Server} -// \ No newline at end of file diff --git a/source/libs/sync/src/syncRaftCfg.c b/source/libs/sync/src/syncRaftCfg.c index ead1168632..6c381f6e7d 100644 --- a/source/libs/sync/src/syncRaftCfg.c +++ b/source/libs/sync/src/syncRaftCfg.c @@ -101,7 +101,7 @@ cJSON *syncCfg2Json(SSyncCfg *pSyncCfg) { char *syncCfg2Str(SSyncCfg *pSyncCfg) { cJSON *pJson = syncCfg2Json(pSyncCfg); - char * serialized = cJSON_Print(pJson); + char *serialized = cJSON_Print(pJson); cJSON_Delete(pJson); return serialized; } @@ -109,7 +109,7 @@ char *syncCfg2Str(SSyncCfg *pSyncCfg) { char *syncCfg2SimpleStr(SSyncCfg *pSyncCfg) { if (pSyncCfg != NULL) { int32_t len = 512; - char * s = taosMemoryMalloc(len); + char *s = taosMemoryMalloc(len); memset(s, 0, len); snprintf(s, len, "{replica-num:%d, my-index:%d, ", pSyncCfg->replicaNum, pSyncCfg->myIndex); @@ -206,7 +206,7 @@ cJSON *raftCfg2Json(SRaftCfg *pRaftCfg) { char *raftCfg2Str(SRaftCfg *pRaftCfg) { cJSON *pJson = raftCfg2Json(pRaftCfg); - char * serialized = cJSON_Print(pJson); + char *serialized = cJSON_Print(pJson); cJSON_Delete(pJson); return serialized; } @@ -285,7 +285,7 @@ int32_t raftCfgFromJson(const cJSON *pRoot, SRaftCfg *pRaftCfg) { (pRaftCfg->configIndexArr)[i] = atoll(pIndex->valuestring); } - cJSON * pJsonSyncCfg = cJSON_GetObjectItem(pJson, "SSyncCfg"); + cJSON *pJsonSyncCfg = cJSON_GetObjectItem(pJson, "SSyncCfg"); int32_t code = syncCfgFromJson(pJsonSyncCfg, &(pRaftCfg->cfg)); ASSERT(code == 0); diff --git a/source/libs/sync/src/syncRaftLog.c b/source/libs/sync/src/syncRaftLog.c index c5d339b08f..3a44933eea 100644 --- a/source/libs/sync/src/syncRaftLog.c +++ b/source/libs/sync/src/syncRaftLog.c @@ -326,6 +326,14 @@ static int32_t raftLogTruncate(struct SSyncLogStore* pLogStore, SyncIndex fromIn ASSERT(0); } + + // event log + do { + char logBuf[128]; + snprintf(logBuf, sizeof(logBuf), "wal truncate, from-index:%ld", fromIndex); + syncNodeEventLog(pData->pSyncNode, logBuf); + } while (0); + return code; } @@ -463,6 +471,14 @@ int32_t logStoreTruncate(SSyncLogStore* pLogStore, SyncIndex fromIndex) { ASSERT(0); } + + // event log + do { + char logBuf[128]; + snprintf(logBuf, sizeof(logBuf), "wal truncate, from-index:%ld", fromIndex); + syncNodeEventLog(pData->pSyncNode, logBuf); + } while (0); + return 0; } diff --git a/source/libs/sync/src/syncReplication.c b/source/libs/sync/src/syncReplication.c index da31e9c4c4..f1c093fe37 100644 --- a/source/libs/sync/src/syncReplication.c +++ b/source/libs/sync/src/syncReplication.c @@ -148,7 +148,7 @@ int32_t syncNodeAppendEntriesPeersSnapshot2(SSyncNode* pSyncNode) { // get entry batch int32_t getCount = 0; SyncIndex getEntryIndex = nextIndex; - for (int32_t i = 0; i < pSyncNode->batchSize; ++i) { + for (int32_t i = 0; i < pSyncNode->pRaftCfg->batchSize; ++i) { SSyncRaftEntry* pEntry = NULL; int32_t code = pSyncNode->pLogStore->syncLogGetEntry(pSyncNode->pLogStore, getEntryIndex, &pEntry); if (code == 0) { @@ -162,12 +162,22 @@ int32_t syncNodeAppendEntriesPeersSnapshot2(SSyncNode* pSyncNode) { } } + // event log + do { + char logBuf[128]; + char host[64]; + uint16_t port; + syncUtilU642Addr(pDestId->addr, host, sizeof(host), &port); + snprintf(logBuf, sizeof(logBuf), "build batch:%d for %s:%d", getCount, host, port); + syncNodeEventLog(pSyncNode, logBuf); + } while (0); + // build msg SyncAppendEntriesBatch* pMsg = syncAppendEntriesBatchBuild(entryPArr, getCount, pSyncNode->vgId); ASSERT(pMsg != NULL); // free entries - for (int32_t i = 0; i < pSyncNode->batchSize; ++i) { + for (int32_t i = 0; i < pSyncNode->pRaftCfg->batchSize; ++i) { SSyncRaftEntry* pEntry = entryPArr[i]; if (pEntry != NULL) { syncEntryDestory(pEntry); diff --git a/source/libs/sync/test/CMakeLists.txt b/source/libs/sync/test/CMakeLists.txt index 37d9707cfd..e1f3a2b2fc 100644 --- a/source/libs/sync/test/CMakeLists.txt +++ b/source/libs/sync/test/CMakeLists.txt @@ -54,6 +54,7 @@ add_executable(syncRaftLogTest2 "") add_executable(syncRaftLogTest3 "") add_executable(syncLeaderTransferTest "") add_executable(syncReconfigFinishTest "") +add_executable(syncRestoreFromSnapshot "") target_sources(syncTest @@ -280,6 +281,10 @@ target_sources(syncReconfigFinishTest PRIVATE "syncReconfigFinishTest.cpp" ) +target_sources(syncRestoreFromSnapshot + PRIVATE + "syncRestoreFromSnapshot.cpp" +) target_include_directories(syncTest @@ -562,6 +567,11 @@ target_include_directories(syncReconfigFinishTest "${TD_SOURCE_DIR}/include/libs/sync" "${CMAKE_CURRENT_SOURCE_DIR}/../inc" ) +target_include_directories(syncRestoreFromSnapshot + PUBLIC + "${TD_SOURCE_DIR}/include/libs/sync" + "${CMAKE_CURRENT_SOURCE_DIR}/../inc" +) target_link_libraries(syncTest @@ -788,6 +798,10 @@ target_link_libraries(syncReconfigFinishTest sync gtest_main ) +target_link_libraries(syncRestoreFromSnapshot + sync + gtest_main +) enable_testing() diff --git a/source/libs/sync/test/syncLogStoreTest.cpp b/source/libs/sync/test/syncLogStoreTest.cpp index 27e1009335..9cb8194aa7 100644 --- a/source/libs/sync/test/syncLogStoreTest.cpp +++ b/source/libs/sync/test/syncLogStoreTest.cpp @@ -85,6 +85,7 @@ void logStoreTest() { } int main(int argc, char** argv) { + gRaftDetailLog = true; tsAsyncLog = 0; sDebugFlag = DEBUG_TRACE + DEBUG_SCREEN + DEBUG_FILE; diff --git a/source/libs/sync/test/syncRestoreFromSnapshot.cpp b/source/libs/sync/test/syncRestoreFromSnapshot.cpp new file mode 100644 index 0000000000..470dd678b0 --- /dev/null +++ b/source/libs/sync/test/syncRestoreFromSnapshot.cpp @@ -0,0 +1,78 @@ +#include +#include +#include "syncEnv.h" +#include "syncIO.h" +#include "syncInt.h" +#include "syncRaftLog.h" +#include "syncRaftStore.h" +#include "syncUtil.h" +#include "wal.h" + +void logTest() { + sTrace("--- sync log test: trace"); + sDebug("--- sync log test: debug"); + sInfo("--- sync log test: info"); + sWarn("--- sync log test: warn"); + sError("--- sync log test: error"); + sFatal("--- sync log test: fatal"); +} + +void init() { + int code = walInit(); + assert(code == 0); +} + +void cleanup() { walCleanUp(); } + +SWal* createWal(char* path, int32_t vgId) { + SWalCfg walCfg; + memset(&walCfg, 0, sizeof(SWalCfg)); + walCfg.vgId = vgId; + walCfg.fsyncPeriod = 1000; + walCfg.retentionPeriod = 1000; + walCfg.rollPeriod = 1000; + walCfg.retentionSize = 1000; + walCfg.segSize = 1000; + walCfg.level = TAOS_WAL_FSYNC; + SWal* pWal = walOpen(path, &walCfg); + assert(pWal != NULL); + return pWal; +} + +SSyncNode* createSyncNode(SWal* pWal) { + SSyncNode* pSyncNode = (SSyncNode*)taosMemoryMalloc(sizeof(SSyncNode)); + memset(pSyncNode, 0, sizeof(SSyncNode)); + pSyncNode->pWal = pWal; + return pSyncNode; +} + +void usage(char* exe) { printf("usage: %s path vgId snapshotIndex \n", exe); } + +int main(int argc, char** argv) { + if (argc != 4) { + usage(argv[0]); + exit(-1); + } + char* path = argv[1]; + int32_t vgId = atoi(argv[2]); + int64_t snapshotIndex = atoll(argv[3]); + + init(); + SWal* pWal = createWal(path, vgId); + assert(pWal != NULL); + SSyncNode* pSyncNode = createSyncNode(pWal); + assert(pSyncNode != NULL); + + SSyncLogStore* pLog = logStoreCreate(pSyncNode); + assert(pLog != NULL); + + int32_t code = pLog->syncLogRestoreFromSnapshot(pLog, snapshotIndex); + assert(code == 0); + + walClose(pWal); + logStoreDestory(pLog); + taosMemoryFree(pSyncNode); + + cleanup(); + return 0; +} diff --git a/source/libs/transport/src/transCli.c b/source/libs/transport/src/transCli.c index 94163fb369..6fd3199990 100644 --- a/source/libs/transport/src/transCli.c +++ b/source/libs/transport/src/transCli.c @@ -573,8 +573,8 @@ static void cliRecvCb(uv_stream_t* handle, ssize_t nread, const uv_buf_t* buf) { return; } if (nread < 0) { - tError("%s conn %p read error: %s, ref: %d", CONN_GET_INST_LABEL(conn), conn, uv_err_name(nread), - T_REF_VAL_GET(conn)); + tWarn("%s conn %p read error: %s, ref: %d", CONN_GET_INST_LABEL(conn), conn, uv_err_name(nread), + T_REF_VAL_GET(conn)); conn->broken = true; cliHandleExcept(conn); } diff --git a/source/libs/transport/src/transSvr.c b/source/libs/transport/src/transSvr.c index 120649f839..2624c252a2 100644 --- a/source/libs/transport/src/transSvr.c +++ b/source/libs/transport/src/transSvr.c @@ -305,7 +305,7 @@ void uvOnRecvCb(uv_stream_t* cli, ssize_t nread, const uv_buf_t* buf) { return; } - tError("%s conn %p read error:%s", transLabel(pTransInst), conn, uv_err_name(nread)); + tWarn("%s conn %p read error:%s", transLabel(pTransInst), conn, uv_err_name(nread)); if (nread < 0) { conn->broken = true; if (conn->status == ConnAcquire) { diff --git a/source/libs/transport/test/CMakeLists.txt b/source/libs/transport/test/CMakeLists.txt index 98a252e008..5645f49284 100644 --- a/source/libs/transport/test/CMakeLists.txt +++ b/source/libs/transport/test/CMakeLists.txt @@ -1,30 +1,16 @@ add_executable(transportTest "") -add_executable(client "") -add_executable(server "") add_executable(transUT "") -add_executable(syncClient "") add_executable(pushServer "") target_sources(transUT PRIVATE "transUT.cpp" -) +) + target_sources(transportTest PRIVATE "transportTests.cpp" -) -target_sources (client - PRIVATE - "rclient.c" -) -target_sources (server - PRIVATE - "rserver.c" -) -target_sources (syncClient - PRIVATE - "syncClient.c" -) +) target_sources(pushServer PRIVATE @@ -35,7 +21,7 @@ target_include_directories(transportTest PUBLIC "${TD_SOURCE_DIR}/include/libs/transport" "${CMAKE_CURRENT_SOURCE_DIR}/../inc" -) +) target_link_libraries (transportTest os @@ -44,6 +30,7 @@ target_link_libraries (transportTest gtest_main transport ) + target_link_libraries (transUT os util @@ -52,56 +39,18 @@ target_link_libraries (transUT transport ) -target_include_directories(client - PUBLIC - "${TD_SOURCE_DIR}/include/libs/transport" - "${CMAKE_CURRENT_SOURCE_DIR}/../inc" -) - -target_link_libraries (client - os - util - common - gtest_main - transport -) -target_include_directories(server - PUBLIC - "${TD_SOURCE_DIR}/include/libs/transport" - "${CMAKE_CURRENT_SOURCE_DIR}/../inc" -) - target_include_directories(transUT PUBLIC "${TD_SOURCE_DIR}/include/libs/transport" "${CMAKE_CURRENT_SOURCE_DIR}/../inc" -) - -target_link_libraries (server - os - util - common - gtest_main - transport -) -target_include_directories(syncClient - PUBLIC - "${TD_SOURCE_DIR}/include/libs/transport" - "${CMAKE_CURRENT_SOURCE_DIR}/../inc" -) -target_link_libraries (syncClient - os - util - common - gtest_main - transport ) target_include_directories(pushServer PUBLIC "${TD_SOURCE_DIR}/include/libs/transport" "${CMAKE_CURRENT_SOURCE_DIR}/../inc" -) +) + target_link_libraries (pushServer os util @@ -110,7 +59,6 @@ target_link_libraries (pushServer transport ) - add_test( NAME transUT COMMAND transUT diff --git a/source/libs/transport/test/rclient.c b/source/libs/transport/test/rclient.c deleted file mode 100644 index 55e6dd000a..0000000000 --- a/source/libs/transport/test/rclient.c +++ /dev/null @@ -1,216 +0,0 @@ -/* - * 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 . - */ -#include -#include "os.h" -#include "taoserror.h" -#include "tglobal.h" -#include "transLog.h" -#include "trpc.h" -#include "tutil.h" - -typedef struct { - int index; - SEpSet epSet; - int num; - int numOfReqs; - int msgSize; - tsem_t rspSem; - tsem_t * pOverSem; - TdThread thread; - void * pRpc; -} SInfo; -static void processResponse(void *pParent, SRpcMsg *pMsg, SEpSet *pEpSet) { - SInfo *pInfo = (SInfo *)pMsg->info.ahandle; - // tError("thread:%d, response is received, type:%d contLen:%d code:0x%x", pInfo->index, pMsg->msgType, pMsg->contLen, - // pMsg->code); - - if (pEpSet) pInfo->epSet = *pEpSet; - - rpcFreeCont(pMsg->pCont); - // tsem_post(&pInfo->rspSem); - tsem_post(&pInfo->rspSem); -} - -static int tcount = 0; - -static void *sendRequest(void *param) { - SInfo * pInfo = (SInfo *)param; - SRpcMsg rpcMsg = {0}; - - tError("thread:%d, start to send request", pInfo->index); - - tError("thread:%d, reqs: %d", pInfo->index, pInfo->numOfReqs); - int u100 = 0; - int u500 = 0; - int u1000 = 0; - int u10000 = 0; - - while (pInfo->numOfReqs == 0 || pInfo->num < pInfo->numOfReqs) { - pInfo->num++; - rpcMsg.pCont = rpcMallocCont(pInfo->msgSize); - rpcMsg.contLen = pInfo->msgSize; - rpcMsg.info.ahandle = pInfo; - rpcMsg.msgType = 1; - // tDebug("thread:%d, send request, contLen:%d num:%d", pInfo->index, pInfo->msgSize, pInfo->num); - int64_t start = taosGetTimestampUs(); - rpcSendRequest(pInfo->pRpc, &pInfo->epSet, &rpcMsg, NULL); - if (pInfo->num % 20000 == 0) tError("thread:%d, %d requests have been sent", pInfo->index, pInfo->num); - // tsem_wait(&pInfo->rspSem); - tsem_wait(&pInfo->rspSem); - int64_t end = taosGetTimestampUs() - start; - if (end <= 100) { - u100++; - } else if (end > 100 && end <= 500) { - u500++; - } else if (end > 500 && end < 1000) { - u1000++; - } else { - u10000++; - } - - tDebug("recv response succefully"); - - // taosSsleep(100); - } - - tError("send and recv sum: %d, %d, %d, %d", u100, u500, u1000, u10000); - tError("thread:%d, it is over", pInfo->index); - tcount++; - - return NULL; -} - -int main(int argc, char *argv[]) { - SRpcInit rpcInit; - SEpSet epSet = {0}; - int msgSize = 128; - int numOfReqs = 0; - int appThreads = 1; - char serverIp[40] = "127.0.0.1"; - char secret[20] = "mypassword"; - struct timeval systemTime; - int64_t startTime, endTime; - TdThreadAttr thattr; - - // server info - epSet.inUse = 0; - addEpIntoEpSet(&epSet, serverIp, 7000); - addEpIntoEpSet(&epSet, "192.168.0.1", 7000); - - // client info - memset(&rpcInit, 0, sizeof(rpcInit)); - rpcInit.localPort = 0; - rpcInit.label = "APP"; - rpcInit.numOfThreads = 1; - rpcInit.cfp = processResponse; - rpcInit.sessions = 100; - rpcInit.idleTime = 100; - rpcInit.user = "michael"; - rpcInit.connType = TAOS_CONN_CLIENT; - rpcDebugFlag = 131; - - for (int i = 1; i < argc; ++i) { - if (strcmp(argv[i], "-p") == 0 && i < argc - 1) { - epSet.eps[0].port = atoi(argv[++i]); - } else if (strcmp(argv[i], "-i") == 0 && i < argc - 1) { - tstrncpy(epSet.eps[0].fqdn, argv[++i], sizeof(epSet.eps[0].fqdn)); - } else if (strcmp(argv[i], "-t") == 0 && i < argc - 1) { - rpcInit.numOfThreads = atoi(argv[++i]); - } else if (strcmp(argv[i], "-m") == 0 && i < argc - 1) { - msgSize = atoi(argv[++i]); - } else if (strcmp(argv[i], "-s") == 0 && i < argc - 1) { - rpcInit.sessions = atoi(argv[++i]); - } else if (strcmp(argv[i], "-n") == 0 && i < argc - 1) { - numOfReqs = atoi(argv[++i]); - } else if (strcmp(argv[i], "-a") == 0 && i < argc - 1) { - appThreads = atoi(argv[++i]); - } else if (strcmp(argv[i], "-o") == 0 && i < argc - 1) { - tsCompressMsgSize = atoi(argv[++i]); - } else if (strcmp(argv[i], "-u") == 0 && i < argc - 1) { - rpcInit.user = argv[++i]; - } else if (strcmp(argv[i], "-k") == 0 && i < argc - 1) { - } else if (strcmp(argv[i], "-spi") == 0 && i < argc - 1) { - } else if (strcmp(argv[i], "-d") == 0 && i < argc - 1) { - rpcDebugFlag = atoi(argv[++i]); - } else { - printf("\nusage: %s [options] \n", argv[0]); - printf(" [-i ip]: first server IP address, default is:%s\n", serverIp); - printf(" [-p port]: server port number, default is:%d\n", epSet.eps[0].port); - printf(" [-t threads]: number of rpc threads, default is:%d\n", rpcInit.numOfThreads); - printf(" [-s sessions]: number of rpc sessions, default is:%d\n", rpcInit.sessions); - printf(" [-m msgSize]: message body size, default is:%d\n", msgSize); - printf(" [-a threads]: number of app threads, default is:%d\n", appThreads); - printf(" [-n requests]: number of requests per thread, default is:%d\n", numOfReqs); - printf(" [-o compSize]: compression message size, default is:%d\n", tsCompressMsgSize); - printf(" [-u user]: user name for the connection, default is:%s\n", rpcInit.user); - printf(" [-d debugFlag]: debug flag, default:%d\n", rpcDebugFlag); - printf(" [-h help]: print out this help\n\n"); - exit(0); - } - } - - const char *path = TD_TMP_DIR_PATH "transport/client"; - taosRemoveDir(path); - taosMkDir(path); - tstrncpy(tsLogDir, path, PATH_MAX); - taosInitLog("client.log", 10); - - void *pRpc = rpcOpen(&rpcInit); - if (pRpc == NULL) { - tError("failed to initialize RPC"); - return -1; - } - - tError("client is initialized"); - tError("threads:%d msgSize:%d requests:%d", appThreads, msgSize, numOfReqs); - - taosGetTimeOfDay(&systemTime); - startTime = systemTime.tv_sec * 1000000 + systemTime.tv_usec; - - SInfo *pInfo = (SInfo *)taosMemoryCalloc(1, sizeof(SInfo) * appThreads); - - taosThreadAttrInit(&thattr); - taosThreadAttrSetDetachState(&thattr, PTHREAD_CREATE_JOINABLE); - - for (int i = 0; i < appThreads; ++i) { - pInfo->index = i; - pInfo->epSet = epSet; - pInfo->numOfReqs = numOfReqs; - pInfo->msgSize = msgSize; - tsem_init(&pInfo->rspSem, 0, 0); - pInfo->pRpc = pRpc; - taosThreadCreate(&pInfo->thread, &thattr, sendRequest, pInfo); - pInfo++; - } - - do { - taosUsleep(1); - } while (tcount < appThreads); - - taosGetTimeOfDay(&systemTime); - endTime = systemTime.tv_sec * 1000000 + systemTime.tv_usec; - float usedTime = (endTime - startTime) / 1000.0f; // mseconds - - tError("it takes %.3f mseconds to send %d requests to server", usedTime, numOfReqs * appThreads); - tError("Performance: %.3f requests per second, msgSize:%d bytes", 1000.0 * numOfReqs * appThreads / usedTime, - msgSize); - - int ch = getchar(); - UNUSED(ch); - - taosCloseLog(); - - return 0; -} diff --git a/source/libs/transport/test/rsclient.c b/source/libs/transport/test/rsclient.c deleted file mode 100644 index e7fc7b45e1..0000000000 --- a/source/libs/transport/test/rsclient.c +++ /dev/null @@ -1,196 +0,0 @@ -/* - * 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 . - */ - - -#include "os.h" -#include "tutil.h" -#include "tglobal.h" -#include "rpcLog.h" -#include "trpc.h" -#include "taoserror.h" - -typedef struct { - int index; - SRpcEpSet epSet; - int num; - int numOfReqs; - int msgSize; - tsem_t rspSem; - tsem_t *pOverSem; - TdThread thread; - void *pRpc; -} SInfo; - - -static int tcount = 0; -static int terror = 0; - -static void *sendRequest(void *param) { - SInfo *pInfo = (SInfo *)param; - SRpcMsg rpcMsg, rspMsg; - - tDebug("thread:%d, start to send request", pInfo->index); - - while ( pInfo->numOfReqs == 0 || pInfo->num < pInfo->numOfReqs) { - pInfo->num++; - rpcMsg.pCont = rpcMallocCont(pInfo->msgSize); - rpcMsg.contLen = pInfo->msgSize; - rpcMsg.handle = pInfo; - rpcMsg.msgType = 1; - tDebug("thread:%d, send request, contLen:%d num:%d", pInfo->index, pInfo->msgSize, pInfo->num); - - rpcSendRecv(pInfo->pRpc, &pInfo->epSet, &rpcMsg, &rspMsg); - - // handle response - if (rspMsg.code != 0) terror++; - - tDebug("thread:%d, rspLen:%d code:%d", pInfo->index, rspMsg.contLen, rspMsg.code); - - rpcFreeCont(rspMsg.pCont); - - if ( pInfo->num % 20000 == 0 ) - tInfo("thread:%d, %d requests have been sent", pInfo->index, pInfo->num); - } - - tDebug("thread:%d, it is over", pInfo->index); - tcount++; - - return NULL; -} - -int main(int argc, char *argv[]) { - SRpcInit rpcInit; - SRpcEpSet epSet; - int msgSize = 128; - int numOfReqs = 0; - int appThreads = 1; - char serverIp[40] = "127.0.0.1"; - char secret[TSDB_KEY_LEN] = "mypassword"; - struct timeval systemTime; - int64_t startTime, endTime; - TdThreadAttr thattr; - - // server info - epSet.numOfEps = 1; - epSet.inUse = 0; - epSet.port[0] = 7000; - epSet.port[1] = 7000; - strcpy(epSet.fqdn[0], serverIp); - strcpy(epSet.fqdn[1], "192.168.0.1"); - - // client info - memset(&rpcInit, 0, sizeof(rpcInit)); - //rpcInit.localIp = "0.0.0.0"; - rpcInit.localPort = 0; - rpcInit.label = "APP"; - rpcInit.numOfThreads = 1; - rpcInit.sessions = 100; - rpcInit.idleTime = 3000; //tsShellActivityTimer*1000; - rpcInit.user = "michael"; - rpcInit.secret = secret; - rpcInit.ckey = "key"; - rpcInit.spi = 1; - rpcInit.connType = TAOS_CONN_CLIENT; - - for (int i=1; iindex = i; - pInfo->epSet = epSet; - pInfo->numOfReqs = numOfReqs; - pInfo->msgSize = msgSize; - tsem_init(&pInfo->rspSem, 0, 0); - pInfo->pRpc = pRpc; - taosThreadCreate(&pInfo->thread, &thattr, sendRequest, pInfo); - pInfo++; - } - - do { - taosUsleep(1); - } while ( tcount < appThreads); - - taosGetTimeOfDay(&systemTime); - endTime = systemTime.tv_sec*1000000 + systemTime.tv_usec; - float usedTime = (endTime - startTime)/1000.0; // mseconds - - tInfo("it takes %.3f mseconds to send %d requests to server, error num:%d", usedTime, numOfReqs*appThreads, terror); - tInfo("Performance: %.3f requests per second, msgSize:%d bytes", 1000.0*numOfReqs*appThreads/usedTime, msgSize); - - taosCloseLog(); - - return 0; -} - - diff --git a/source/libs/transport/test/rserver.c b/source/libs/transport/test/rserver.c deleted file mode 100644 index 1fd78be77d..0000000000 --- a/source/libs/transport/test/rserver.c +++ /dev/null @@ -1,194 +0,0 @@ -/* - * 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 . - */ - -//#define _DEFAULT_SOURCE -#include "os.h" -#include "tglobal.h" -#include "tqueue.h" -#include "transLog.h" -#include "trpc.h" - -int msgSize = 128; -int commit = 0; -TdFilePtr pDataFile = NULL; -STaosQueue *qhandle = NULL; -STaosQset * qset = NULL; - -void processShellMsg() { - static int num = 0; - STaosQall *qall; - SRpcMsg * pRpcMsg, rpcMsg; - int type; - void * pvnode; - - qall = taosAllocateQall(); - - while (1) { - int numOfMsgs = taosReadAllQitemsFromQset(qset, qall, &pvnode, NULL); - tDebug("%d shell msgs are received", numOfMsgs); - if (numOfMsgs <= 0) break; - - for (int i = 0; i < numOfMsgs; ++i) { - taosGetQitem(qall, (void **)&pRpcMsg); - - if (pDataFile != NULL) { - if (taosWriteFile(pDataFile, pRpcMsg->pCont, pRpcMsg->contLen) < 0) { - tInfo("failed to write data file, reason:%s", strerror(errno)); - } - } - } - - if (commit >= 2) { - num += numOfMsgs; - // if (taosFsync(pDataFile) < 0) { - // tInfo("failed to flush data to file, reason:%s", strerror(errno)); - //} - - if (num % 10000 == 0) { - tInfo("%d request have been written into disk", num); - } - } - - taosResetQitems(qall); - for (int i = 0; i < numOfMsgs; ++i) { - taosGetQitem(qall, (void **)&pRpcMsg); - rpcFreeCont(pRpcMsg->pCont); - - memset(&rpcMsg, 0, sizeof(rpcMsg)); - rpcMsg.pCont = rpcMallocCont(msgSize); - rpcMsg.contLen = msgSize; - rpcMsg.info = pRpcMsg->info; - rpcMsg.code = 0; - rpcSendResponse(&rpcMsg); - - taosFreeQitem(pRpcMsg); - } - } - - taosFreeQall(qall); -} - -int retrieveAuthInfo(void *parent, char *meterId, char *spi, char *encrypt, char *secret, char *ckey) { - // app shall retrieve the auth info based on meterID from DB or a data file - // demo code here only for simple demo - int ret = 0; - - if (strcmp(meterId, "michael") == 0) { - *spi = 1; - *encrypt = 0; - strcpy(secret, "mypassword"); - strcpy(ckey, "key"); - } else if (strcmp(meterId, "jeff") == 0) { - *spi = 0; - *encrypt = 0; - } else { - ret = -1; // user not there - } - - return ret; -} - -void processRequestMsg(void *pParent, SRpcMsg *pMsg, SEpSet *pEpSet) { - SRpcMsg *pTemp; - - pTemp = taosAllocateQitem(sizeof(SRpcMsg), DEF_QITEM); - memcpy(pTemp, pMsg, sizeof(SRpcMsg)); - - tDebug("request is received, type:%d, contLen:%d, item:%p", pMsg->msgType, pMsg->contLen, pTemp); - taosWriteQitem(qhandle, pTemp); -} - -int main(int argc, char *argv[]) { - SRpcInit rpcInit; - char dataName[20] = "server.data"; - - taosBlockSIGPIPE(); - - memset(&rpcInit, 0, sizeof(rpcInit)); - rpcInit.localPort = 7000; - rpcInit.label = "SER"; - rpcInit.numOfThreads = 1; - rpcInit.cfp = processRequestMsg; - rpcInit.sessions = 1000; - rpcInit.idleTime = 2 * 1500; - - rpcDebugFlag = 131; - - for (int i = 1; i < argc; ++i) { - if (strcmp(argv[i], "-p") == 0 && i < argc - 1) { - rpcInit.localPort = atoi(argv[++i]); - } else if (strcmp(argv[i], "-t") == 0 && i < argc - 1) { - rpcInit.numOfThreads = atoi(argv[++i]); - } else if (strcmp(argv[i], "-m") == 0 && i < argc - 1) { - msgSize = atoi(argv[++i]); - } else if (strcmp(argv[i], "-s") == 0 && i < argc - 1) { - rpcInit.sessions = atoi(argv[++i]); - } else if (strcmp(argv[i], "-o") == 0 && i < argc - 1) { - tsCompressMsgSize = atoi(argv[++i]); - } else if (strcmp(argv[i], "-w") == 0 && i < argc - 1) { - commit = atoi(argv[++i]); - } else if (strcmp(argv[i], "-d") == 0 && i < argc - 1) { - rpcDebugFlag = atoi(argv[++i]); - dDebugFlag = rpcDebugFlag; - uDebugFlag = rpcDebugFlag; - } else { - printf("\nusage: %s [options] \n", argv[0]); - printf(" [-p port]: server port number, default is:%d\n", rpcInit.localPort); - printf(" [-t threads]: number of rpc threads, default is:%d\n", rpcInit.numOfThreads); - printf(" [-s sessions]: number of sessions, default is:%d\n", rpcInit.sessions); - printf(" [-m msgSize]: message body size, default is:%d\n", msgSize); - printf(" [-o compSize]: compression message size, default is:%d\n", tsCompressMsgSize); - printf(" [-w write]: write received data to file(0, 1, 2), default is:%d\n", commit); - printf(" [-d debugFlag]: debug flag, default:%d\n", rpcDebugFlag); - printf(" [-h help]: print out this help\n\n"); - exit(0); - } - } - - tsAsyncLog = 0; - rpcInit.connType = TAOS_CONN_SERVER; - - const char *path = TD_TMP_DIR_PATH "transport/server"; - taosRemoveDir(path); - taosMkDir(path); - tstrncpy(tsLogDir, path, PATH_MAX); - taosInitLog("server.log", 10); - - void *pRpc = rpcOpen(&rpcInit); - if (pRpc == NULL) { - tError("failed to start RPC server"); - return -1; - } - // sleep(5); - - tInfo("RPC server is running, ctrl-c to exit"); - - if (commit) { - pDataFile = taosOpenFile(dataName, TD_FILE_CREATE | TD_FILE_WRITE | TD_FILE_APPEND); - if (pDataFile == NULL) tInfo("failed to open data file, reason:%s", strerror(errno)); - } - qhandle = taosOpenQueue(); - qset = taosOpenQset(); - taosAddIntoQset(qset, qhandle, NULL); - - processShellMsg(); - - if (pDataFile != NULL) { - taosCloseFile(&pDataFile); - taosRemoveFile(dataName); - } - - return 0; -} diff --git a/source/libs/transport/test/syncClient.c b/source/libs/transport/test/syncClient.c deleted file mode 100644 index bc6461eaed..0000000000 --- a/source/libs/transport/test/syncClient.c +++ /dev/null @@ -1,211 +0,0 @@ -/* - * 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 . - */ -#include -#include "os.h" -#include "taoserror.h" -#include "tglobal.h" -#include "transLog.h" -#include "trpc.h" -#include "tutil.h" - -typedef struct { - int index; - SEpSet epSet; - int num; - int numOfReqs; - int msgSize; - tsem_t rspSem; - tsem_t * pOverSem; - TdThread thread; - void * pRpc; -} SInfo; -static void processResponse(void *pParent, SRpcMsg *pMsg, SEpSet *pEpSet) { - SInfo *pInfo = (SInfo *)pMsg->info.ahandle; - tDebug("thread:%d, response is received, type:%d contLen:%d code:0x%x", pInfo->index, pMsg->msgType, pMsg->contLen, - pMsg->code); - - if (pEpSet) pInfo->epSet = *pEpSet; - - rpcFreeCont(pMsg->pCont); - // tsem_post(&pInfo->rspSem); - tsem_post(&pInfo->rspSem); -} - -static int tcount = 0; - -static void *sendRequest(void *param) { - SInfo * pInfo = (SInfo *)param; - SRpcMsg rpcMsg = {0}; - - tDebug("thread:%d, start to send request", pInfo->index); - - tDebug("thread:%d, reqs: %d", pInfo->index, pInfo->numOfReqs); - int u100 = 0; - int u500 = 0; - int u1000 = 0; - int u10000 = 0; - SRpcMsg respMsg = {0}; - while (pInfo->numOfReqs == 0 || pInfo->num < pInfo->numOfReqs) { - pInfo->num++; - rpcMsg.pCont = rpcMallocCont(pInfo->msgSize); - rpcMsg.contLen = pInfo->msgSize; - rpcMsg.info.ahandle = pInfo; - rpcMsg.msgType = 1; - // tDebug("thread:%d, send request, contLen:%d num:%d", pInfo->index, pInfo->msgSize, pInfo->num); - int64_t start = taosGetTimestampUs(); - rpcSendRecv(pInfo->pRpc, &pInfo->epSet, &rpcMsg, &respMsg); - // rpcSendRequest(pInfo->pRpc, &pInfo->epSet, &rpcMsg, NULL); - if (pInfo->num % 20000 == 0) tInfo("thread:%d, %d requests have been sent", pInfo->index, pInfo->num); - // tsem_wait(&pInfo->rspSem); - // wtsem_wait(&pInfo->rspSem); - int64_t end = taosGetTimestampUs() - start; - if (end <= 100) { - u100++; - } else if (end > 100 && end <= 500) { - u500++; - } else if (end > 500 && end < 1000) { - u1000++; - } else { - u10000++; - } - - tDebug("recv response succefully"); - - // taosSsleep(100); - } - - tError("send and recv sum: %d, %d, %d, %d", u100, u500, u1000, u10000); - tDebug("thread:%d, it is over", pInfo->index); - tcount++; - - return NULL; -} - -int main(int argc, char *argv[]) { - SRpcInit rpcInit; - SEpSet epSet = {0}; - int msgSize = 128; - int numOfReqs = 0; - int appThreads = 1; - char serverIp[40] = "127.0.0.1"; - char secret[20] = "mypassword"; - struct timeval systemTime; - int64_t startTime, endTime; - TdThreadAttr thattr; - - // server info - epSet.inUse = 0; - addEpIntoEpSet(&epSet, serverIp, 7000); - addEpIntoEpSet(&epSet, "192.168.0.1", 7000); - - // client info - memset(&rpcInit, 0, sizeof(rpcInit)); - rpcInit.localPort = 0; - rpcInit.label = "APP"; - rpcInit.numOfThreads = 1; - rpcInit.cfp = processResponse; - rpcInit.sessions = 100; - rpcInit.idleTime = 100; - rpcInit.user = "michael"; - rpcInit.connType = TAOS_CONN_CLIENT; - - for (int i = 1; i < argc; ++i) { - if (strcmp(argv[i], "-p") == 0 && i < argc - 1) { - epSet.eps[0].port = atoi(argv[++i]); - } else if (strcmp(argv[i], "-i") == 0 && i < argc - 1) { - tstrncpy(epSet.eps[0].fqdn, argv[++i], sizeof(epSet.eps[0].fqdn)); - } else if (strcmp(argv[i], "-t") == 0 && i < argc - 1) { - rpcInit.numOfThreads = atoi(argv[++i]); - } else if (strcmp(argv[i], "-m") == 0 && i < argc - 1) { - msgSize = atoi(argv[++i]); - } else if (strcmp(argv[i], "-s") == 0 && i < argc - 1) { - rpcInit.sessions = atoi(argv[++i]); - } else if (strcmp(argv[i], "-n") == 0 && i < argc - 1) { - numOfReqs = atoi(argv[++i]); - } else if (strcmp(argv[i], "-a") == 0 && i < argc - 1) { - appThreads = atoi(argv[++i]); - } else if (strcmp(argv[i], "-o") == 0 && i < argc - 1) { - tsCompressMsgSize = atoi(argv[++i]); - } else if (strcmp(argv[i], "-u") == 0 && i < argc - 1) { - rpcInit.user = argv[++i]; - } else if (strcmp(argv[i], "-k") == 0 && i < argc - 1) { - } else if (strcmp(argv[i], "-spi") == 0 && i < argc - 1) { - } else if (strcmp(argv[i], "-d") == 0 && i < argc - 1) { - rpcDebugFlag = atoi(argv[++i]); - } else { - printf("\nusage: %s [options] \n", argv[0]); - printf(" [-i ip]: first server IP address, default is:%s\n", serverIp); - printf(" [-p port]: server port number, default is:%d\n", epSet.eps[0].port); - printf(" [-t threads]: number of rpc threads, default is:%d\n", rpcInit.numOfThreads); - printf(" [-s sessions]: number of rpc sessions, default is:%d\n", rpcInit.sessions); - printf(" [-m msgSize]: message body size, default is:%d\n", msgSize); - printf(" [-a threads]: number of app threads, default is:%d\n", appThreads); - printf(" [-n requests]: number of requests per thread, default is:%d\n", numOfReqs); - printf(" [-o compSize]: compression message size, default is:%d\n", tsCompressMsgSize); - printf(" [-u user]: user name for the connection, default is:%s\n", rpcInit.user); - printf(" [-d debugFlag]: debug flag, default:%d\n", rpcDebugFlag); - printf(" [-h help]: print out this help\n\n"); - exit(0); - } - } - - taosInitLog("client.log", 10); - - void *pRpc = rpcOpen(&rpcInit); - if (pRpc == NULL) { - tError("failed to initialize RPC"); - return -1; - } - - tInfo("client is initialized"); - tInfo("threads:%d msgSize:%d requests:%d", appThreads, msgSize, numOfReqs); - - taosGetTimeOfDay(&systemTime); - startTime = systemTime.tv_sec * 1000000 + systemTime.tv_usec; - - SInfo *pInfo = (SInfo *)taosMemoryCalloc(1, sizeof(SInfo) * appThreads); - - taosThreadAttrInit(&thattr); - taosThreadAttrSetDetachState(&thattr, PTHREAD_CREATE_JOINABLE); - - for (int i = 0; i < appThreads; ++i) { - pInfo->index = i; - pInfo->epSet = epSet; - pInfo->numOfReqs = numOfReqs; - pInfo->msgSize = msgSize; - tsem_init(&pInfo->rspSem, 0, 0); - pInfo->pRpc = pRpc; - taosThreadCreate(&pInfo->thread, &thattr, sendRequest, pInfo); - pInfo++; - } - - do { - taosUsleep(1); - } while (tcount < appThreads); - - taosGetTimeOfDay(&systemTime); - endTime = systemTime.tv_sec * 1000000 + systemTime.tv_usec; - float usedTime = (endTime - startTime) / 1000.0f; // mseconds - - tInfo("it takes %.3f mseconds to send %d requests to server", usedTime, numOfReqs * appThreads); - tInfo("Performance: %.3f requests per second, msgSize:%d bytes", 1000.0 * numOfReqs * appThreads / usedTime, msgSize); - - int ch = getchar(); - UNUSED(ch); - - taosCloseLog(); - - return 0; -} diff --git a/source/os/src/osStrptime.c b/source/os/src/osStrptime.c deleted file mode 100644 index 8d878577ea..0000000000 --- a/source/os/src/osStrptime.c +++ /dev/null @@ -1,405 +0,0 @@ -/* $Id$ */ -/* $NetBSD: strptime.c,v 1.18 1999/04/29 02:58:30 tv Exp $ */ - -/*- -* Copyright (c) 1997, 1998 The NetBSD Foundation, Inc. -* All rights reserved. -* -* This code was contributed to The NetBSD Foundation by Klaus Klein. -* -* Redistribution and use in source and binary forms, with or without -* modification, are permitted provided that the following conditions -* are met: -* 1. Redistributions of source code must retain the above copyright -* notice, this list of conditions and the following disclaimer. -* 2. Redistributions in binary form must reproduce the above copyright -* notice, this list of conditions and the following disclaimer in the -* documentation and/or other materials provided with the distribution. -* 3. All advertising materials mentioning features or use of this software -* must display the following acknowledgement: -* This product includes software developed by the NetBSD -* Foundation, Inc. and its contributors. -* 4. Neither the name of The NetBSD Foundation nor the names of its -* contributors may be used to endorse or promote products derived -* from this software without specific prior written permission. -* -* THIS SOFTWARE IS PROVIDED BY THE NETBSD FOUNDATION, INC. AND CONTRIBUTORS -* ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED -* TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR -* PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE FOUNDATION OR CONTRIBUTORS -* BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR -* CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF -* SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS -* INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN -* CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) -* ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE -* POSSIBILITY OF SUCH DAMAGE. -*/ -// -//#include "lukemftp.h" - -// #ifdef WINDOWS - -// #include -// #include -// #include -// #include -// //#define TM_YEAR_BASE 1970 //origin -// #define TM_YEAR_BASE 1900 //slguan -// /* -// * We do not implement alternate representations. However, we always -// * check whether a given modifier is allowed for a certain conversion. -// */ -// #define ALT_E 0x01 -// #define ALT_O 0x02 -// #define LEGAL_ALT(x) { if (alt_format & ~(x)) return (0); } - - -// static int conv_num(const char **buf, int *dest, int llim, int ulim) -// { -// int result = 0; - -// /* The limit also determines the number of valid digits. */ -// int rulim = ulim; - -// if (**buf < '0' || **buf > '9') -// return (0); - -// do { -// result *= 10; -// result += *(*buf)++ - '0'; -// rulim /= 10; -// } while ((result * 10 <= ulim) && rulim && **buf >= '0' && **buf <= '9'); - -// if (result < llim || result > ulim) -// return (0); - -// *dest = result; -// return (1); -// } - -// static const char *day[7] = { -// "Sunday", "Monday", "Tuesday", "Wednesday", "Thursday", -// "Friday", "Saturday" -// }; -// static const char *abday[7] = { -// "Sun", "Mon", "Tue", "Wed", "Thu", "Fri", "Sat" -// }; -// static const char *mon[12] = { -// "January", "February", "March", "April", "May", "June", "July", -// "August", "September", "October", "November", "December" -// }; -// static const char *abmon[12] = { -// "Jan", "Feb", "Mar", "Apr", "May", "Jun", -// "Jul", "Aug", "Sep", "Oct", "Nov", "Dec" -// }; -// static const char *am_pm[2] = { -// "AM", "PM" -// }; - -// #endif - -// char *taosStrpTime(const char *buf, const char *fmt, struct tm *tm) { -// #ifdef WINDOWS -// char c; -// const char *bp; -// size_t len = 0; -// int alt_format, i, split_year = 0; - -// bp = buf; - -// while ((c = *fmt) != '\0') { -// /* Clear `alternate' modifier prior to new conversion. */ -// alt_format = 0; - -// /* Eat up white-space. */ -// if (isspace(c)) { -// while (isspace(*bp)) -// bp++; - -// fmt++; -// continue; -// } - -// if ((c = *fmt++) != '%') -// goto literal; - - -// again: switch (c = *fmt++) { -// case '%': /* "%%" is converted to "%". */ -// literal : -// if (c != *bp++) -// return (0); -// break; - -// /* -// * "Alternative" modifiers. Just set the appropriate flag -// * and start over again. -// */ -// case 'E': /* "%E?" alternative conversion modifier. */ -// LEGAL_ALT(0); -// alt_format |= ALT_E; -// goto again; - -// case 'O': /* "%O?" alternative conversion modifier. */ -// LEGAL_ALT(0); -// alt_format |= ALT_O; -// goto again; - -// /* -// * "Complex" conversion rules, implemented through recursion. -// */ -// case 'c': /* Date and time, using the locale's format. */ -// LEGAL_ALT(ALT_E); -// if (!(bp = taosStrpTime(bp, "%x %X", tm))) -// return (0); -// break; - -// case 'D': /* The date as "%m/%d/%y". */ -// LEGAL_ALT(0); -// if (!(bp = taosStrpTime(bp, "%m/%d/%y", tm))) -// return (0); -// break; - -// case 'R': /* The time as "%H:%M". */ -// LEGAL_ALT(0); -// if (!(bp = taosStrpTime(bp, "%H:%M", tm))) -// return (0); -// break; - -// case 'r': /* The time in 12-hour clock representation. */ -// LEGAL_ALT(0); -// if (!(bp = taosStrpTime(bp, "%I:%M:%S %p", tm))) -// return (0); -// break; - -// case 'T': /* The time as "%H:%M:%S". */ -// LEGAL_ALT(0); -// if (!(bp = taosStrpTime(bp, "%H:%M:%S", tm))) -// return (0); -// break; - -// case 'X': /* The time, using the locale's format. */ -// LEGAL_ALT(ALT_E); -// if (!(bp = taosStrpTime(bp, "%H:%M:%S", tm))) -// return (0); -// break; - -// case 'x': /* The date, using the locale's format. */ -// LEGAL_ALT(ALT_E); -// if (!(bp = taosStrpTime(bp, "%m/%d/%y", tm))) -// return (0); -// break; - -// /* -// * "Elementary" conversion rules. -// */ -// case 'A': /* The day of week, using the locale's form. */ -// case 'a': -// LEGAL_ALT(0); -// for (i = 0; i < 7; i++) { -// /* Full name. */ -// len = strlen(day[i]); -// if (strncmp(day[i], bp, len) == 0) -// break; - -// /* Abbreviated name. */ -// len = strlen(abday[i]); -// if (strncmp(abday[i], bp, len) == 0) -// break; -// } - -// /* Nothing matched. */ -// if (i == 7) -// return (0); - -// tm->tm_wday = i; -// bp += len; -// break; - -// case 'B': /* The month, using the locale's form. */ -// case 'b': -// case 'h': -// LEGAL_ALT(0); -// for (i = 0; i < 12; i++) { -// /* Full name. */ -// len = strlen(mon[i]); -// if (strncmp(mon[i], bp, len) == 0) -// break; - -// /* Abbreviated name. */ -// len = strlen(abmon[i]); -// if (strncmp(abmon[i], bp, len) == 0) -// break; -// } - -// /* Nothing matched. */ -// if (i == 12) -// return (0); - -// tm->tm_mon = i; -// bp += len; -// break; - -// case 'C': /* The century number. */ -// LEGAL_ALT(ALT_E); -// if (!(conv_num(&bp, &i, 0, 99))) -// return (0); - -// if (split_year) { -// tm->tm_year = (tm->tm_year % 100) + (i * 100); -// } -// else { -// tm->tm_year = i * 100; -// split_year = 1; -// } -// break; - -// case 'd': /* The day of month. */ -// case 'e': -// LEGAL_ALT(ALT_O); -// if (!(conv_num(&bp, &tm->tm_mday, 1, 31))) -// return (0); -// break; - -// case 'k': /* The hour (24-hour clock representation). */ -// LEGAL_ALT(0); -// /* FALLTHROUGH */ -// case 'H': -// LEGAL_ALT(ALT_O); -// if (!(conv_num(&bp, &tm->tm_hour, 0, 23))) -// return (0); -// break; - -// case 'l': /* The hour (12-hour clock representation). */ -// LEGAL_ALT(0); -// /* FALLTHROUGH */ -// case 'I': -// LEGAL_ALT(ALT_O); -// if (!(conv_num(&bp, &tm->tm_hour, 1, 12))) -// return (0); -// if (tm->tm_hour == 12) -// tm->tm_hour = 0; -// break; - -// case 'j': /* The day of year. */ -// LEGAL_ALT(0); -// if (!(conv_num(&bp, &i, 1, 366))) -// return (0); -// tm->tm_yday = i - 1; -// break; - -// case 'M': /* The minute. */ -// LEGAL_ALT(ALT_O); -// if (!(conv_num(&bp, &tm->tm_min, 0, 59))) -// return (0); -// break; - -// case 'm': /* The month. */ -// LEGAL_ALT(ALT_O); -// if (!(conv_num(&bp, &i, 1, 12))) -// return (0); -// tm->tm_mon = i - 1; -// break; - -// case 'p': /* The locale's equivalent of AM/PM. */ -// LEGAL_ALT(0); -// /* AM? */ -// if (strcmp(am_pm[0], bp) == 0) { -// if (tm->tm_hour > 11) -// return (0); - -// bp += strlen(am_pm[0]); -// break; -// } -// /* PM? */ -// else if (strcmp(am_pm[1], bp) == 0) { -// if (tm->tm_hour > 11) -// return (0); - -// tm->tm_hour += 12; -// bp += strlen(am_pm[1]); -// break; -// } - -// /* Nothing matched. */ -// return (0); - -// case 'S': /* The seconds. */ -// LEGAL_ALT(ALT_O); -// if (!(conv_num(&bp, &tm->tm_sec, 0, 61))) -// return (0); -// break; - -// case 'U': /* The week of year, beginning on sunday. */ -// case 'W': /* The week of year, beginning on monday. */ -// LEGAL_ALT(ALT_O); -// /* -// * XXX This is bogus, as we can not assume any valid -// * information present in the tm structure at this -// * point to calculate a real value, so just check the -// * range for now. -// */ -// if (!(conv_num(&bp, &i, 0, 53))) -// return (0); -// break; - -// case 'w': /* The day of week, beginning on sunday. */ -// LEGAL_ALT(ALT_O); -// if (!(conv_num(&bp, &tm->tm_wday, 0, 6))) -// return (0); -// break; - -// case 'Y': /* The year. */ -// LEGAL_ALT(ALT_E); -// if (!(conv_num(&bp, &i, 0, 9999))) -// return (0); - -// tm->tm_year = i - TM_YEAR_BASE; -// break; - -// case 'y': /* The year within 100 years of the epoch. */ -// LEGAL_ALT(ALT_E | ALT_O); -// if (!(conv_num(&bp, &i, 0, 99))) -// return (0); - -// if (split_year) { -// tm->tm_year = ((tm->tm_year / 100) * 100) + i; -// break; -// } -// split_year = 1; -// if (i <= 68) -// tm->tm_year = i + 2000 - TM_YEAR_BASE; -// else -// tm->tm_year = i + 1900 - TM_YEAR_BASE; -// break; - -// /* -// * Miscellaneous conversions. -// */ -// case 'n': /* Any kind of white-space. */ -// case 't': -// LEGAL_ALT(0); -// while (isspace(*bp)) -// bp++; -// break; - - -// default: /* Unknown/unsupported conversion. */ -// return (0); -// } - - -// } - -// /* LINTED functional specification */ -// return ((char *)bp); -// #elif defined(_TD_DARWIN_64) -// return strptime(buf, fmt, tm); -// #else -// return strptime(buf, fmt, tm); -// #endif -// } - - - diff --git a/source/util/src/tpagedbuf.c b/source/util/src/tpagedbuf.c index f50fd22ba6..0e608d0da2 100644 --- a/source/util/src/tpagedbuf.c +++ b/source/util/src/tpagedbuf.c @@ -518,8 +518,8 @@ void releaseBufPageInfo(SDiskbasedBuf* pBuf, SPageInfo* pi) { #ifdef BUF_PAGE_DEBUG uDebug("page_releaseBufPageInfo pageId:%d, used:%d, offset:%"PRId64, pi->pageId, pi->used, pi->offset); #endif - assert(pi->pData != NULL && pi->used == true); -// assert(pi->pData != NULL); + // assert(pi->pData != NULL && pi->used == true); + assert(pi->pData != NULL); pi->used = false; pBuf->statis.releasePages += 1; } diff --git a/source/util/src/ttrace.c b/source/util/src/ttrace.c deleted file mode 100644 index f183fd58fd..0000000000 --- a/source/util/src/ttrace.c +++ /dev/null @@ -1,79 +0,0 @@ -/* - * 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 . - */ -#include "ttrace.h" -#include "taos.h" -#include "thash.h" -#include "tuuid.h" - -// clang-format off -//static TdThreadOnce init = PTHREAD_ONCE_INIT; -//static void * ids = NULL; -//static TdThreadMutex mtx; -// -//void traceInit() { -// ids = taosHashInit(4096, taosGetDefaultHashFunction(TSDB_DATA_TYPE_INT), true, HASH_ENTRY_LOCK); -// taosThreadMutexInit(&mtx, NULL); -//} -//void traceCreateEnv() { -// taosThreadOnce(&init, traceInit); -//} -//void traceDestroyEnv() { -// taosThreadMutexDestroy(&mtx); -// taosHashCleanup(ids); -//} -// -//STraceId traceInitId(STraceSubId *h, STraceSubId *l) { -// STraceId id = *h; -// id = ((id << 32) & 0xFFFFFFFF) | ((*l) & 0xFFFFFFFF); -// return id; -//} -//void traceId2Str(STraceId *id, char *buf) { -// int32_t f = (*id >> 32) & 0xFFFFFFFF; -// int32_t s = (*id) & 0xFFFFFFFF; -// sprintf(buf, "%d:%d", f, s); -//} -// -//void traceSetSubId(STraceId *id, STraceSubId *subId) { -// int32_t parent = ((*id >> 32) & 0xFFFFFFFF); -// taosThreadMutexLock(&mtx); -// taosHashPut(ids, subId, sizeof(*subId), &parent, sizeof(parent)); -// taosThreadMutexUnlock(&mtx); -//} -// -//STraceSubId traceGetParentId(STraceId *id) { -// int32_t parent = ((*id >> 32) & 0xFFFFFFFF); -// taosThreadMutexLock(&mtx); -// STraceSubId *p = taosHashGet(ids, (void *)&parent, sizeof(parent)); -// parent = *p; -// taosThreadMutexUnlock(&mtx); -// -// return parent; -//} -// -//STraceSubId traceGenSubId() { -// return tGenIdPI32(); -//} -//void traceSetRootId(STraceId *traceid, int64_t rootId) { -// traceId->rootId = rootId; -//} -//int64_t traceGetRootId(STraceId *traceId); -// -//void traceSetMsgId(STraceId *traceid, int64_t msgId); -//int64_t traceGetMsgId(STraceId *traceid); -// -//char *trace2Str(STraceId *id); -// -//void traceSetSubId(STraceId *id, int32_t *subId); -// clang-format on diff --git a/tests/pytest/crash_gen.sh b/tests/pytest/crash_gen.sh index 127e13c5be..539314dea4 100755 --- a/tests/pytest/crash_gen.sh +++ b/tests/pytest/crash_gen.sh @@ -48,7 +48,12 @@ fi PYTHON_EXEC=python3.8 # First we need to set up a path for Python to find our own TAOS modules, so that "import" can work. -export PYTHONPATH=$(pwd)/../../src/connector/python:$(pwd) +# export PYTHONPATH=$(pwd)/../../src/connector/python:$(pwd) +# NOTE: we are now pointing outside the current github, per Wade on 7/7/2022, we'll be keeping connectors outside +# and there does not seem to be a module to reference that. +PROJECT_PARENT=$(pwd)/../../.. +TAOS_PYTHON_PROJECT_DIR=$PROJECT_PARENT/taos-connector-python +export PYTHONPATH=$TAOS_PYTHON_PROJECT_DIR:$(pwd) # Then let us set up the library path so that our compiled SO file can be loaded by Python export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:$LIB_DIR diff --git a/tests/pytest/crash_gen/crash_gen_main.py b/tests/pytest/crash_gen/crash_gen_main.py index b743eee2ef..08155f656b 100755 --- a/tests/pytest/crash_gen/crash_gen_main.py +++ b/tests/pytest/crash_gen/crash_gen_main.py @@ -466,6 +466,7 @@ class ThreadCoordinator: self._te = None # No more executor, time to end Logging.debug("Main thread tapping all threads one last time...") self.tapAllThreads() # Let the threads run one last time + #TODO: looks like we are not capturing the failures for the last step yet (i.e. calling registerFailure if neccessary) Logging.debug("\r\n\n--> Main thread ready to finish up...") Logging.debug("Main thread joining all threads") @@ -1290,6 +1291,7 @@ class Task(): def _isErrAcceptable(self, errno, msg): if errno in [ + # TDengine 2.x Error Codes: 0x05, # TSDB_CODE_RPC_NOT_READY 0x0B, # Unable to establish connection, more details in TD-1648 # 0x200, # invalid SQL, TODO: re-examine with TD-934 @@ -1310,6 +1312,18 @@ class Task(): 0x14, # db not ready, errno changed 0x600, # Invalid table ID, why? 0x218, # Table does not exist + + # TDengine 3.0 Error Codes: + 0x0333, # Object is creating # TODO: this really is NOT an acceptable error + 0x03A0, # STable already exists + 0x03A1, # STable [does] not exist + 0x03AA, # Tag already exists + 0x0603, # Table already exists + 0x2602, # Table does not exist + 0x260d, # Tags number not matched + + + 1000 # REST catch-all error ]: return True # These are the ALWAYS-ACCEPTABLE ones @@ -1749,6 +1763,8 @@ class TdSuperTable: tagType = tags[tagName] if tagType == 'BINARY': tagStrs.append("'Beijing-Shanghai-LosAngeles'") + elif tagType== 'VARCHAR': + tagStrs.append("'London-Paris-Berlin'") elif tagType == 'FLOAT': tagStrs.append('9.9') elif tagType == 'INT': diff --git a/tests/script/jenkins/basic.txt b/tests/script/jenkins/basic.txt index 4e009e702d..0783aa0fd1 100644 --- a/tests/script/jenkins/basic.txt +++ b/tests/script/jenkins/basic.txt @@ -163,8 +163,8 @@ # --- sma ./test.sh -f tsim/sma/drop_sma.sim ./test.sh -f tsim/sma/tsmaCreateInsertQuery.sim -#./test.sh -f tsim/sma/rsmaCreateInsertQuery.sim -#./test.sh -f tsim/sma/rsmaPersistenceRecovery.sim +./test.sh -f tsim/sma/rsmaCreateInsertQuery.sim +./test.sh -f tsim/sma/rsmaPersistenceRecovery.sim # --- valgrind ./test.sh -f tsim/valgrind/checkError1.sim diff --git a/tests/script/loop.sh b/tests/script/loop.sh index b435eef123..b3bc9728de 100755 --- a/tests/script/loop.sh +++ b/tests/script/loop.sh @@ -13,7 +13,7 @@ CMD_NAME= LOOP_TIMES=5 SLEEP_TIME=0 -while getopts "f:t:s:" arg +while getopts "hf:t:s:" arg do case $arg in f) @@ -25,6 +25,12 @@ do s) SLEEP_TIME=$OPTARG ;; + h) + echo "Usage: $(basename $0) -f [cmd name] " + echo " -t [loop times] " + echo " -s [sleep time] " + exit 0 + ;; ?) echo "unknow argument" ;; diff --git a/tests/script/tsim/sync/vnodesnapshot.sim b/tests/script/tsim/sync/vnodesnapshot.sim index 7ad3afe309..aa42047400 100644 --- a/tests/script/tsim/sync/vnodesnapshot.sim +++ b/tests/script/tsim/sync/vnodesnapshot.sim @@ -142,20 +142,33 @@ sql create table ct1 using stb tags(1000) system sh/exec.sh -n dnode4 -s stop -x SIGINT -sql insert into ct1 values(now+0s, 10, 2.0, 3.0) -sql insert into ct1 values(now+1s, 11, 2.1, 3.1)(now+2s, -12, -2.2, -3.2)(now+3s, -13, -2.3, -3.3) - -sql flush database db; - -system sh/exec.sh -n dnode4 -s start - -sql insert into ct1 values(now+1s, 81, 8.1, 8.1)(now+2s, -92, -9.2, -9.2)(now+3s, -73, -7.3, -7.3) +$N = 100 +$count = 0 +while $count < $N + $ms = 1591200000000 + $count + sql insert into ct1 values( $ms , $count , 2.1, 3.1) + $count = $count + 1 +endw -#system sh/exec.sh -n dnode1 -s stop -x SIGINT -#system sh/exec.sh -n dnode2 -s stop -x SIGINT -#system sh/exec.sh -n dnode3 -s stop -x SIGINT +#sql insert into ct1 values(now+0s, 10, 2.0, 3.0) +#sql insert into ct1 values(now+1s, 11, 2.1, 3.1)(now+2s, -12, -2.2, -3.2)(now+3s, -13, -2.3, -3.3) + +#sql flush database db; + +#system sh/exec.sh -n dnode4 -s start + +#sql insert into ct1 values(now+1s, 81, 8.1, 8.1)(now+2s, -92, -9.2, -9.2)(now+3s, -73, -7.3, -7.3) + + + +sleep 5000 + + +system sh/exec.sh -n dnode1 -s stop -x SIGINT +system sh/exec.sh -n dnode2 -s stop -x SIGINT +system sh/exec.sh -n dnode3 -s stop -x SIGINT #system sh/exec.sh -n dnode4 -s stop -x SIGINT diff --git a/tests/system-test/2-query/and_or_for_byte.py b/tests/system-test/2-query/and_or_for_byte.py index 3a0b6652ba..aab9cce040 100644 --- a/tests/system-test/2-query/and_or_for_byte.py +++ b/tests/system-test/2-query/and_or_for_byte.py @@ -230,7 +230,7 @@ class TDTestCase: work_sql += f"cast({arg} as bigint){opera}" if not agg: - work_sql+= f" from {tbname} order by ts" + work_sql+= f" from {tbname} order by tbname ,ts" else: work_sql+= f" from {tbname} " tdSql.query(work_sql) @@ -243,7 +243,7 @@ class TDTestCase: else: origin_sql += f"cast({arg} as bigint)," if not agg: - origin_sql+= f" from {tbname} order by ts" + origin_sql+= f" from {tbname} order by tbname ,ts" else: origin_sql+= f" from {tbname} " tdSql.query(origin_sql) diff --git a/tests/system-test/2-query/json_tag.py b/tests/system-test/2-query/json_tag.py index 79ce04d7d1..f66f0f67ab 100644 --- a/tests/system-test/2-query/json_tag.py +++ b/tests/system-test/2-query/json_tag.py @@ -323,12 +323,12 @@ class TDTestCase: # where json value is bool tdSql.query("select * from jsons1 where jtag->'tag1'=true") tdSql.checkRows(0) - #tdSql.query("select * from jsons1 where jtag->'tag1'=false") - #tdSql.checkRows(1) + tdSql.query("select * from jsons1 where jtag->'tag1'=false") + tdSql.checkRows(1) tdSql.query("select * from jsons1 where jtag->'tag1'!=false") tdSql.checkRows(0) - #tdSql.query("select * from jsons1 where jtag->'tag1'>false") - #tdSql.checkRows(0) + tdSql.query("select * from jsons1 where jtag->'tag1'>false") + tdSql.checkRows(0) # where json value is null tdSql.query("select * from jsons1 where jtag->'tag1'=null") diff --git a/tests/system-test/7-tmq/tmqConsFromTsdb-1ctb-funcNFilter.py b/tests/system-test/7-tmq/tmqConsFromTsdb-1ctb-funcNFilter.py index bd8531d7c8..d22d183e86 100644 --- a/tests/system-test/7-tmq/tmqConsFromTsdb-1ctb-funcNFilter.py +++ b/tests/system-test/7-tmq/tmqConsFromTsdb-1ctb-funcNFilter.py @@ -64,8 +64,9 @@ class TDTestCase: startTs=paraDict["startTs"],ctbStartIdx=paraDict['ctbStartIdx']) tdLog.info("restart taosd to ensure that the data falls into the disk") - tdDnodes.stop(1) - tdDnodes.start(1) + # tdDnodes.stop(1) + # tdDnodes.start(1) + tdSql.query("flush database %s"%(paraDict['dbName'])) return def tmqCase1(self): diff --git a/tests/system-test/7-tmq/tmqConsFromTsdb-1ctb.py b/tests/system-test/7-tmq/tmqConsFromTsdb-1ctb.py index 1d2aaccda5..951a747069 100644 --- a/tests/system-test/7-tmq/tmqConsFromTsdb-1ctb.py +++ b/tests/system-test/7-tmq/tmqConsFromTsdb-1ctb.py @@ -64,8 +64,9 @@ class TDTestCase: startTs=paraDict["startTs"],ctbStartIdx=paraDict['ctbStartIdx']) tdLog.info("restart taosd to ensure that the data falls into the disk") - tdDnodes.stop(1) - tdDnodes.start(1) + # tdDnodes.stop(1) + # tdDnodes.start(1) + tdSql.query("flush database %s"%(paraDict['dbName'])) return def tmqCase1(self): diff --git a/tests/system-test/7-tmq/tmqConsFromTsdb-mutilVg-mutilCtb-funcNFilter.py b/tests/system-test/7-tmq/tmqConsFromTsdb-mutilVg-mutilCtb-funcNFilter.py index 2720b7cdce..6ee089af4e 100644 --- a/tests/system-test/7-tmq/tmqConsFromTsdb-mutilVg-mutilCtb-funcNFilter.py +++ b/tests/system-test/7-tmq/tmqConsFromTsdb-mutilVg-mutilCtb-funcNFilter.py @@ -64,9 +64,9 @@ class TDTestCase: startTs=paraDict["startTs"],ctbStartIdx=paraDict['ctbStartIdx']) tdLog.info("restart taosd to ensure that the data falls into the disk") - tdDnodes.stop(1) - # tdDnodes.start(1) - tdDnodes.starttaosd(1) + # tdDnodes.stop(1) + # tdDnodes.starttaosd(1) + tdSql.query("flush database %s"%(paraDict['dbName'])) return def tmqCase1(self): diff --git a/tests/system-test/7-tmq/tmqConsFromTsdb-mutilVg-mutilCtb.py b/tests/system-test/7-tmq/tmqConsFromTsdb-mutilVg-mutilCtb.py index 0b2dddd24a..882989bfb6 100644 --- a/tests/system-test/7-tmq/tmqConsFromTsdb-mutilVg-mutilCtb.py +++ b/tests/system-test/7-tmq/tmqConsFromTsdb-mutilVg-mutilCtb.py @@ -64,9 +64,10 @@ class TDTestCase: startTs=paraDict["startTs"],ctbStartIdx=paraDict['ctbStartIdx']) tdLog.info("restart taosd to ensure that the data falls into the disk") - tdDnodes.stop(1) - # tdDnodes.start(1) - tdDnodes.starttaosd(1) + # tdDnodes.stop(1) + # # tdDnodes.start(1) + # tdDnodes.starttaosd(1) + tdSql.query("flush database %s"%(paraDict['dbName'])) return def tmqCase1(self): diff --git a/tests/system-test/7-tmq/tmqConsFromTsdb-mutilVg.py b/tests/system-test/7-tmq/tmqConsFromTsdb-mutilVg.py index a4d6648276..c334ff752b 100644 --- a/tests/system-test/7-tmq/tmqConsFromTsdb-mutilVg.py +++ b/tests/system-test/7-tmq/tmqConsFromTsdb-mutilVg.py @@ -64,9 +64,10 @@ class TDTestCase: startTs=paraDict["startTs"],ctbStartIdx=paraDict['ctbStartIdx']) tdLog.info("restart taosd to ensure that the data falls into the disk") - tdDnodes.stop(1) - # tdDnodes.start(1) - tdDnodes.starttaosd(1) + # tdDnodes.stop(1) + # # tdDnodes.start(1) + # tdDnodes.starttaosd(1) + tdSql.query("flush database %s"%(paraDict['dbName'])) return def tmqCase1(self): diff --git a/tests/system-test/7-tmq/tmqConsFromTsdb.py b/tests/system-test/7-tmq/tmqConsFromTsdb.py index c18474dcc3..a4a242365a 100644 --- a/tests/system-test/7-tmq/tmqConsFromTsdb.py +++ b/tests/system-test/7-tmq/tmqConsFromTsdb.py @@ -64,8 +64,9 @@ class TDTestCase: startTs=paraDict["startTs"],ctbStartIdx=paraDict['ctbStartIdx']) tdLog.info("restart taosd to ensure that the data falls into the disk") - tdDnodes.stop(1) - tdDnodes.start(1) + # tdDnodes.stop(1) + # tdDnodes.start(1) + tdSql.query("flush database %s"%(paraDict['dbName'])) return def tmqCase1(self): diff --git a/tests/system-test/7-tmq/tmqConsFromTsdb1-1ctb-funcNFilter.py b/tests/system-test/7-tmq/tmqConsFromTsdb1-1ctb-funcNFilter.py index 540c9dbbe1..ec11b3286c 100644 --- a/tests/system-test/7-tmq/tmqConsFromTsdb1-1ctb-funcNFilter.py +++ b/tests/system-test/7-tmq/tmqConsFromTsdb1-1ctb-funcNFilter.py @@ -64,8 +64,9 @@ class TDTestCase: startTs=paraDict["startTs"],ctbStartIdx=paraDict['ctbStartIdx']) tdLog.info("restart taosd to ensure that the data falls into the disk") - tdDnodes.stop(1) - tdDnodes.start(1) + # tdDnodes.stop(1) + # tdDnodes.start(1) + tdSql.query("flush database %s"%(paraDict['dbName'])) return def tmqCase3(self): diff --git a/tests/system-test/7-tmq/tmqConsFromTsdb1-1ctb.py b/tests/system-test/7-tmq/tmqConsFromTsdb1-1ctb.py index 5cb373092b..4878652593 100644 --- a/tests/system-test/7-tmq/tmqConsFromTsdb1-1ctb.py +++ b/tests/system-test/7-tmq/tmqConsFromTsdb1-1ctb.py @@ -64,8 +64,9 @@ class TDTestCase: startTs=paraDict["startTs"],ctbStartIdx=paraDict['ctbStartIdx']) tdLog.info("restart taosd to ensure that the data falls into the disk") - tdDnodes.stop(1) - tdDnodes.start(1) + # tdDnodes.stop(1) + # tdDnodes.start(1) + tdSql.query("flush database %s"%(paraDict['dbName'])) return def tmqCase3(self): diff --git a/tests/system-test/7-tmq/tmqConsFromTsdb1-mutilVg-mutilCtb-funcNFilter.py b/tests/system-test/7-tmq/tmqConsFromTsdb1-mutilVg-mutilCtb-funcNFilter.py index eaef134845..451dc43343 100644 --- a/tests/system-test/7-tmq/tmqConsFromTsdb1-mutilVg-mutilCtb-funcNFilter.py +++ b/tests/system-test/7-tmq/tmqConsFromTsdb1-mutilVg-mutilCtb-funcNFilter.py @@ -64,8 +64,9 @@ class TDTestCase: startTs=paraDict["startTs"],ctbStartIdx=paraDict['ctbStartIdx']) tdLog.info("restart taosd to ensure that the data falls into the disk") - tdDnodes.stop(1) - tdDnodes.start(1) + # tdDnodes.stop(1) + # tdDnodes.start(1) + tdSql.query("flush database %s"%(paraDict['dbName'])) return def tmqCase3(self): diff --git a/tests/system-test/7-tmq/tmqConsFromTsdb1-mutilVg-mutilCtb.py b/tests/system-test/7-tmq/tmqConsFromTsdb1-mutilVg-mutilCtb.py index c1b327e5f1..3b3467f9f9 100644 --- a/tests/system-test/7-tmq/tmqConsFromTsdb1-mutilVg-mutilCtb.py +++ b/tests/system-test/7-tmq/tmqConsFromTsdb1-mutilVg-mutilCtb.py @@ -64,8 +64,9 @@ class TDTestCase: startTs=paraDict["startTs"],ctbStartIdx=paraDict['ctbStartIdx']) tdLog.info("restart taosd to ensure that the data falls into the disk") - tdDnodes.stop(1) - tdDnodes.start(1) + # tdDnodes.stop(1) + # tdDnodes.start(1) + tdSql.query("flush database %s"%(paraDict['dbName'])) return def tmqCase3(self): diff --git a/tests/system-test/7-tmq/tmqConsFromTsdb1-mutilVg.py b/tests/system-test/7-tmq/tmqConsFromTsdb1-mutilVg.py index dd8a0ad33a..d1fe69f0b7 100644 --- a/tests/system-test/7-tmq/tmqConsFromTsdb1-mutilVg.py +++ b/tests/system-test/7-tmq/tmqConsFromTsdb1-mutilVg.py @@ -64,8 +64,9 @@ class TDTestCase: startTs=paraDict["startTs"],ctbStartIdx=paraDict['ctbStartIdx']) tdLog.info("restart taosd to ensure that the data falls into the disk") - tdDnodes.stop(1) - tdDnodes.start(1) + # tdDnodes.stop(1) + # tdDnodes.start(1) + tdSql.query("flush database %s"%(paraDict['dbName'])) return def tmqCase3(self): diff --git a/tests/system-test/7-tmq/tmqConsFromTsdb1.py b/tests/system-test/7-tmq/tmqConsFromTsdb1.py index a183eda1c7..597f7968ae 100644 --- a/tests/system-test/7-tmq/tmqConsFromTsdb1.py +++ b/tests/system-test/7-tmq/tmqConsFromTsdb1.py @@ -64,8 +64,9 @@ class TDTestCase: startTs=paraDict["startTs"],ctbStartIdx=paraDict['ctbStartIdx']) tdLog.info("restart taosd to ensure that the data falls into the disk") - tdDnodes.stop(1) - tdDnodes.start(1) + # tdDnodes.stop(1) + # tdDnodes.start(1) + tdSql.query("flush database %s"%(paraDict['dbName'])) return def tmqCase3(self): diff --git a/tests/system-test/fulltest.sh b/tests/system-test/fulltest.sh index 1e305f2518..409fb5e930 100755 --- a/tests/system-test/fulltest.sh +++ b/tests/system-test/fulltest.sh @@ -117,7 +117,7 @@ python3 ./test.py -f 2-query/distribute_agg_avg.py python3 ./test.py -f 2-query/distribute_agg_stddev.py python3 ./test.py -f 2-query/twa.py python3 ./test.py -f 2-query/irate.py -#python3 ./test.py -f 2-query/and_or_for_byte.py +python3 ./test.py -f 2-query/and_or_for_byte.py python3 ./test.py -f 2-query/function_null.py python3 ./test.py -f 2-query/queryQnode.py diff --git a/tests/system-test/loop.sh b/tests/system-test/loop.sh new file mode 100755 index 0000000000..b3bc9728de --- /dev/null +++ b/tests/system-test/loop.sh @@ -0,0 +1,55 @@ +#!/bin/bash + +################################################## +# +# Do simulation test +# +################################################## + +set -e +#set -x + +CMD_NAME= +LOOP_TIMES=5 +SLEEP_TIME=0 + +while getopts "hf:t:s:" arg +do + case $arg in + f) + CMD_NAME=$OPTARG + ;; + t) + LOOP_TIMES=$OPTARG + ;; + s) + SLEEP_TIME=$OPTARG + ;; + h) + echo "Usage: $(basename $0) -f [cmd name] " + echo " -t [loop times] " + echo " -s [sleep time] " + exit 0 + ;; + ?) + echo "unknow argument" + ;; + esac +done + +echo LOOP_TIMES ${LOOP_TIMES} +echo CMD_NAME ${CMD_NAME} +echo SLEEP_TIME ${SLEEP_TIME} + +GREEN='\033[1;32m' +GREEN_DARK='\033[0;32m' +GREEN_UNDERLINE='\033[4;32m' +NC='\033[0m' + +for ((i=0; i<$LOOP_TIMES; i++ )) +do + echo -e $GREEN loop $i $NC + echo -e $GREEN cmd $CMD_NAME $NC + $CMD_NAME + sleep ${SLEEP_TIME} +done diff --git a/tools/taosadapter b/tools/taosadapter index 389047db71..c885e967e4 160000 --- a/tools/taosadapter +++ b/tools/taosadapter @@ -1 +1 @@ -Subproject commit 389047db713a3dddfbce292c3260b0864b17d936 +Subproject commit c885e967e490105999b84d009a15168728dfafaf