[
https://issues.apache.org/jira/browse/SOLR-5576?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Christine Poerschke updated SOLR-5576:
--------------------------------------
Attachment: SOLR-5576.patch
Attaching patch to separate publish-as-down and waitForLeaderToSeeDownState
into separate for loops. Also replacing return with continue when
waitForLeaderToSeeDownState call can be skipped.
> ZkController.java registerAllCoresAsDown multiple cores logic
> -------------------------------------------------------------
>
> Key: SOLR-5576
> URL: https://issues.apache.org/jira/browse/SOLR-5576
> Project: Solr
> Issue Type: Bug
> Reporter: Christine Poerschke
> Attachments: SOLR-5576.patch
>
>
> The behaviour we saw was that considerable time elapsed between
> different cores within the same solr instance publishing themselves as down.
> Separately it appears from the code that some cores would not be published as
> down if another core returns from the function early because it will be its
> shard leader (return vs. continue in for loop).
--
This message was sent by Atlassian JIRA
(v6.1.5#6160)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
