xxq250
/
homework-jianmu
1
0
Fork 0
Code Issues Pull Requests 1 Actions Packages Projects Releases Wiki Activity

Update replica-ch.md

This commit is contained in:
Xiaxin Li 2020-07-31 19:07:43 +08:00 committed by GitHub
parent 70d9a83209
commit f9448e3d0c
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 2 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
documentation20/webdocs/markdowndocs/replica-ch.md
View File

@ -105,7 +105,7 @@ TDengine采取的是Master-Slave模式进行同步，与流行的RAFT一致性
如果vnode A是master, vnode B是slave, vnode A能接受客户端的写请求，而vnode B不能。当vnode A收到写的请求后，遵循下面的流程：
<center> <img src="../assets/replica-forward.jpg"> </center>
<center> <img src="../assets/replica-forward.png"> </center>
1. 应用对写请求做基本的合法性检查，通过，则给改请求包打上一个版本号(version, 单调递增）
2. 应用将打上版本号的写请求封装一个WAL Head, 写入WAL(Write Ahead Log)
@ -140,7 +140,7 @@ TDengine采取的是Master-Slave模式进行同步，与流行的RAFT一致性
整个数据恢复流程分为两大步骤，第一步，先恢复archived data(file), 然后恢复wal。具体流程如下：
<center> <img src="../assets/replica-restore.jpg"> </center>
<center> <img src="../assets/replica-restore.png"> </center>
1. 通过已经建立的TCP链接，发送sync req给master节点
2. master收到sync req后，以client的身份，向vnode B主动建立一新的专用于同步的TCP链接（syncFd)