[
https://issues.apache.org/jira/browse/SOLR-8368?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15049172#comment-15049172
]
Mike Drob commented on SOLR-8368:
---------------------------------
Remove the TODO about doing replay before leader election?
Could put ulog declaration in replayTransactionLogIfNeeded, cut down on a
method parameter and some scope that way.
bq. The only change beyond calling tlog replay before entering the election is
that we have to remove an || isLeader check against it since it happens before
we know if we will be the leader or not.
This means that a non-leader in shard construction state will no longer do a
replay. I don't know enough about how shard splits work to know if this is ok,
but I think the impact is that replicas will take longer to come online since
they will need to do an additional recovery once the sub-shard leader is
active. If that's the case then the rest of the patch LGTM.
I see a couple of minor places where we could clarify things, but it's not
worth it for this issue.
> A SolrCore needs to replay it's tlog before the leader election process.
> ------------------------------------------------------------------------
>
> Key: SOLR-8368
> URL: https://issues.apache.org/jira/browse/SOLR-8368
> Project: Solr
> Issue Type: Bug
> Reporter: Mark Miller
> Assignee: Mark Miller
> Attachments: SOLR-8368.patch
>
>
> If we do it after like now, the correct leader may not be able to become
> leader.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org
For additional commands, e-mail: dev-h...@lucene.apache.org
