[
https://issues.apache.org/jira/browse/SOLR-10983?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16074173#comment-16074173
]
Shalin Shekhar Mangar commented on SOLR-10983:
----------------------------------------------
Nice catch!
Your patch solves another problem -- today if an exception happens, we run
through items in the work-queue and the last item from state-update-queue (the
one during which the exception happened) so we run the same item twice.
Considering that DOWNNODE is the only command that enqueues multiple
ZkWriteCommands, I think we should add a method to ZkStateWriter which calls
enqueue only once for the entire batch. That and your patch solve all problems
nicely i.e.
# DOWNNODE creating multiple work queue items
# Exceptions not clearing work queue
# Overseer executing same item twice from work queue and state update queue on
an exception
> Fix DOWNNODE -> queue-work explosion
> ------------------------------------
>
> Key: SOLR-10983
> URL: https://issues.apache.org/jira/browse/SOLR-10983
> Project: Solr
> Issue Type: Bug
> Security Level: Public(Default Security Level. Issues are Public)
> Components: SolrCloud
> Reporter: Scott Blum
> Assignee: Scott Blum
> Attachments: SOLR-10983.patch
>
>
> Every DOWNNODE command enqueues N copies of itself into queue-work, where N
> is number of collections affected by the DOWNNODE.
> This rarely matters in practice, because queue-work gets immediately dumped--
> however, if anything throws an exception (such as ZK bad version), we don't
> clear queue-work. Then the next time through the loop we run the expensive
> DOWNNODE command potentially hundreds of times.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
