[
https://issues.apache.org/jira/browse/HDDS-16185?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=18104660#comment-18104660
]
Ivan Andika commented on HDDS-16185:
------------------------------------
We need to come up with short term solution to prevent this, as I mentioned in
https://issues.apache.org/jira/browse/HDDS-15129 the current SCM Replicate
annotation is very easy to misuse and this can cause SCM crash. The long term
is the leader-side execution, but meanwhile we need to think about the
short-term solution.
> SCM ratis replay fails after HDDS-13846 on upgrade
> --------------------------------------------------
>
> Key: HDDS-16185
> URL: https://issues.apache.org/jira/browse/HDDS-16185
> Project: Apache Ozone
> Issue Type: Bug
> Affects Versions: 2.2.0
> Reporter: Saketa Chalamchala
> Priority: Blocker
>
> On Upgrade, SCM crashes on ratis reply if there are pending transactions
> which require `ContainerStateManagerImpl.updateContainerState(ContainerID,
> LifeCycleEvent)`.
> This method was removed in [HDDS-13846] Container CLOSED with sequence ID
> lower than a replica - ASF Jira
> Fix for this would require adding back
> `ContainerStateManagerImpl.updateContainerState(ContainerID, LifeCycleEvent)`
> as a deprecated method to be upgrade compatible.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
