[ 
https://issues.apache.org/jira/browse/HBASE-17792?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15928625#comment-15928625
 ] 

huaxiang sun commented on HBASE-17792:
--------------------------------------

I just checked, it seems that there is no recent change about the behavior.
I went back to the commit before HBASE-17309 and it is same behavior.

> Use a shared thread pool for AtomicityWriter, AtomicGetReader, 
> AtomicScanReader's connections in TestAcidGuarantees
> -------------------------------------------------------------------------------------------------------------------
>
>                 Key: HBASE-17792
>                 URL: https://issues.apache.org/jira/browse/HBASE-17792
>             Project: HBase
>          Issue Type: Bug
>    Affects Versions: 2.0.0
>            Reporter: huaxiang sun
>            Assignee: huaxiang sun
>            Priority: Minor
>         Attachments: HBASE-17792-master-001.patch
>
>
> AtomicityWriter, AtomicGetReader, AtomicScanReader creates connection inside, 
> since the thread pool is not shared, by default each connection will create 
> 256 threads. If there are 5 AtomicityWriters, 1 AtomicGetReader, and 1 
> AtomicScanReader, it will create lots of threads and causes max user 
> processes  to be reached, and OOME. Use a share thread pool to work around 
> this issue.



--
This message was sent by Atlassian JIRA
(v6.3.15#6346)

Reply via email to