[
https://issues.apache.org/jira/browse/HBASE-17033?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Enis Soztutar updated HBASE-17033:
----------------------------------
Resolution: Fixed
Hadoop Flags: Reviewed
Fix Version/s: 1.4.0
2.0.0
Status: Resolved (was: Patch Available)
Pushed this to branch-1+. Thanks for the review Anoop.
Test failures are not related.
> LogRoller makes a lot of allocations unnecessarily
> --------------------------------------------------
>
> Key: HBASE-17033
> URL: https://issues.apache.org/jira/browse/HBASE-17033
> Project: HBase
> Issue Type: Bug
> Reporter: Enis Soztutar
> Assignee: Enis Soztutar
> Fix For: 2.0.0, 1.4.0
>
> Attachments: Screen Shot 2016-11-04 at 6.39.00 PM.png,
> hbase-17033_v1.patch
>
>
> I was looking at the other allocations for HBASE-17017. Seems that log roller
> thread allocates 200MB for ~7% of the TLAB space. This is a lot of
> allocations.
> I think the reason is this:
> {code}
> while (true) {
> if (this.safePointAttainedLatch.await(1, TimeUnit.NANOSECONDS)) {
> break;
> }
> if (syncFuture.isThrowable()) {
> throw new
> FailedSyncBeforeLogCloseException(syncFuture.getThrowable());
> }
> }
> {code}
> This busy wait is causing a lot allocations because the thread is added to
> the waiting list.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
