[
https://issues.apache.org/jira/browse/SENTRY-1441?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15423206#comment-15423206
]
Rahul Sharma commented on SENTRY-1441:
--------------------------------------
So the rename table is called when the daemon tends to becomeActive. This issue
would also exist when the first leader creates the fence table.
Do we catch the exception and retry(max times) at the fencer level?
> Error during fencing table rename can disable master
> ----------------------------------------------------
>
> Key: SENTRY-1441
> URL: https://issues.apache.org/jira/browse/SENTRY-1441
> Project: Sentry
> Issue Type: Sub-task
> Components: Sentry
> Affects Versions: sentry-ha-redesign
> Reporter: Alexander Kolbasov
> Fix For: sentry-ha-redesign
>
>
> The code that handles fencing table renames handles errors by throwing
> exceptions. These exceptions are propagated all the way to the caller, so the
> leader will not re-try the table rename and will never be able to do any
> other DB operation. The only recovery is to switch back and forth between
> master and slave.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
