[
https://issues.apache.org/jira/browse/RATIS-2499?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Tsz-wo Sze resolved RATIS-2499.
-------------------------------
Fix Version/s: 3.3.0
Resolution: Fixed
The pull request is now merged. Thanks, [~XiChen]!
> Allow the LogAppender restart when LogAppenderDaemon exception
> --------------------------------------------------------------
>
> Key: RATIS-2499
> URL: https://issues.apache.org/jira/browse/RATIS-2499
> Project: Ratis
> Issue Type: Bug
> Components: Leader
> Reporter: ChenXi
> Assignee: ChenXi
> Priority: Major
> Fix For: 3.3.0
>
> Time Spent: 40m
> Remaining Estimate: 0h
>
> RATIS-2427 let the LogAppender do not restart when the LogAppenderDaemon in
> CLOSING or CLOSED or EXCEPTION LifeCycle.States . However, if only
> LogAppenderDaemon is EXCEPTION LifeCycle.States, we should still restart
> LogAppender.
> LeaderStateImpl#restart will create a new GrpcLogAppender (see
> addAndStartSenders -> addSenders -> LeaderStateImpl#newLogAppender) and a new
> LogAppenderDaemon. The old LogAppenderDaemon will be shutdown
> Therefore, if the old LogAppenderDaemon enters the EXCEPTION lifecycle for
> some reason, we should attempt to restart LogAppender
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
