[
https://issues.apache.org/jira/browse/FLINK-18646?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17166346#comment-17166346
]
Andrey Zagrebin edited comment on FLINK-18646 at 7/28/20, 11:27 AM:
--------------------------------------------------------------------
merged into master by 3d056c8fea72ca40b663d12570913679be87c0a9
merged into 1.11 by bcc97082639280ab14f465463fb07b27167c37e3
[~TsReaper] I am closing the issue as the verification should not block the RPC
thread any more. Reopen it if you notice any problems with it. If there are
still problems with the normal memory allocation timeout (given there is no
real leak), we can discuss it in another issue.
was (Author: azagrebin):
merged into master by 3d056c8fea72ca40b663d12570913679be87c0a9
merged into 1.10 by bcc97082639280ab14f465463fb07b27167c37e3
[~TsReaper] I am closing the issue as the verification should not block the RPC
thread any more. Reopen it if you notice any problems with it. If there are
still problems with the normal memory allocation timeout (given there is no
real leak), we can discuss it in another issue.
> Managed memory released check can block RPC thread
> --------------------------------------------------
>
> Key: FLINK-18646
> URL: https://issues.apache.org/jira/browse/FLINK-18646
> Project: Flink
> Issue Type: Bug
> Components: Runtime / Task
> Affects Versions: 1.11.0
> Reporter: Andrey Zagrebin
> Assignee: Andrey Zagrebin
> Priority: Critical
> Labels: pull-request-available
> Fix For: 1.12.0, 1.11.2
>
> Attachments: log1.png, log2.png
>
>
> UnsafeMemoryBudget#verifyEmpty, called on slot freeing, needs time to wait on
> GC of all allocated/released managed memory. If there are a lot of segments
> to GC then it can take time to finish the check. If slot freeing happens in
> RPC thread, the GC waiting can block it and TM risks to miss its heartbeat.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
