[GitHub] [rocketmq] makabakaboom opened a new issue #3015: message may be lost because of slave can report greater than local commitLog offset in sync_mater mode

[GitBox <gi...@apache.org>]

makabakaboom opened a new issue #3015:
URL: https://github.com/apache/rocketmq/issues/3015


   **BUG REPORT**
   
   1. Please describe the issue you observed:
   
   - What did you do (The steps to reproduce)?
   
   1、创建一个SYNC_MASTER模式的集群
   2、创建一个连接去连接同步端口，并往里面随机写一些数据
   3、master节点会当做是slave上报的offset，并更新push2SlaveMaxOffset，导致后续写入消息可能未等到slave同步就已经返回producer
   4、此时master宕机可能就会造成部分数据写master成功，但是还未由pageCache写入到磁盘，且未同步到slave而丢失
   
   - What did you expect to see?
   
   1、判断为不合适的offset，并断掉连接
   - What did you see instead?
   
   master节点会当做是slave上报的offset，并更新push2SlaveMaxOffset，导致后续写入消息可能未等到slave同步就已经返回producer
   
   2. Please tell us about your environment:
   
   最新环境
   3. Other information (e.g. detailed explanation, logs, related issues, suggestions how to fix, etc):
   
   


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org

[GitHub] [rocketmq] areyouok commented on issue #3015: message may be lost because of slave can report greater than local commitLog offset in sync_mater mode

[GitBox <gi...@apache.org>]

areyouok commented on issue #3015:
URL: https://github.com/apache/rocketmq/issues/3015#issuecomment-913631887


   这个问题我们也遇到，我们的解决办法是给主从同步的tcp连接增加一个握手，我认为这才是根本的解决办法。
   
   不过这样一改就有点不兼容了，比如新旧版本的broker之间无法建立连接，所以这个修改我一直没提交。


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: dev-unsubscribe@rocketmq.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org