[
https://issues.apache.org/jira/browse/SOLR-9555?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15830086#comment-15830086
]
Mark Miller commented on SOLR-9555:
-----------------------------------
bq. Whether the state of the replica is already RECOVERING or not makes no
difference. If the leader is not able to send an update then that guarantee
cannot be made and therefore it is safest to restart the recovery process.
And this is correct too. Everytime an update fails from leader to replica we
have to ensure a new recovery starts. This is why recoveries used to be able to
stack up and had to unwind, though we now at least short circuit stack ups.
> Recovery can hang if a node is put into LIR as it is starting up
> ----------------------------------------------------------------
>
> Key: SOLR-9555
> URL: https://issues.apache.org/jira/browse/SOLR-9555
> Project: Solr
> Issue Type: Bug
> Security Level: Public(Default Security Level. Issues are Public)
> Reporter: Alan Woodward
>
> See
> https://jenkins.thetaphi.de/job/Lucene-Solr-master-Linux/17888/consoleFull
> for an example
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
