You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@hbase.apache.org by "Bryan Beaudreault (Jira)" <ji...@apache.org> on 2022/12/09 14:01:00 UTC

[jira] [Resolved] (HBASE-27491) AsyncProcess should not clear meta cache for RejectedExecutionException

     [ https://issues.apache.org/jira/browse/HBASE-27491?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Bryan Beaudreault resolved HBASE-27491.
---------------------------------------
    Fix Version/s: 2.4.16
                   2.5.3
       Resolution: Fixed

Pushed to branch-2, 2.5, and 2.5. 

> AsyncProcess should not clear meta cache for RejectedExecutionException
> -----------------------------------------------------------------------
>
>                 Key: HBASE-27491
>                 URL: https://issues.apache.org/jira/browse/HBASE-27491
>             Project: HBase
>          Issue Type: Improvement
>            Reporter: Bryan Beaudreault
>            Assignee: Briana Augenreich
>            Priority: Major
>             Fix For: 2.4.16, 2.5.3
>
>
> Batch requests using AsyncProcess get submitted to a thread pool. If the pool is too small or backing regionservers are overloaded, the pool may reject requests. Currently this causes a cache clear, which only exacerbates issues in most cases.
> If there's a real reason to clear cache, the individual failing tasks that actually get accepted into the pool should be enough to appropriately clear the cache. So let's special case REE in AsyncRequestFutureImpl to not clear cache.



--
This message was sent by Atlassian Jira
(v8.20.10#820010)