[
https://issues.apache.org/jira/browse/SOLR-17700?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Houston Putman resolved SOLR-17700.
-----------------------------------
Fix Version/s: 9.9
Assignee: Houston Putman
Resolution: Fixed
> CoreContainer does not obtain a lock when unloading cores
> ---------------------------------------------------------
>
> Key: SOLR-17700
> URL: https://issues.apache.org/jira/browse/SOLR-17700
> Project: Solr
> Issue Type: Bug
> Reporter: Houston Putman
> Assignee: Houston Putman
> Priority: Major
> Labels: pull-request-available
> Fix For: 9.9
>
> Time Spent: 20m
> Remaining Estimate: 0h
>
> CoreContainer uses {{solrCores.waitAddPendingCoreOps(name)}} when doing an
> operation on a core. And the documentation for
> {{SolrCores.waitAddPendingCoreOps(name)}} mentions that:
> bq. // Wait here until any pending operations (load, unload or reload) are
> completed on this core.
> However, CoreContainer does not use this locking when unloading a core. This
> is the reason why we are seeing race condition issues when deleting and
> reloading a collection at the same time.
> Loading a core can possibly require unloading said core, when
> "deleteUnknownCoresOnStartup" is used, thus we need a way of unloading
> without a lock. But other than this edge case, locking the unload operation
> should only improve stability in Solr.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
