[
https://issues.apache.org/jira/browse/FLINK-9755?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16538520#comment-16538520
]
ASF GitHub Bot commented on FLINK-9755:
---------------------------------------
Github user NicoK commented on the issue:
https://github.com/apache/flink/pull/6272
actually, I forgot to adapt the unit test which relied on the previous
order of the calls and assumed that the buffer got recycled. With the changes
from the fixup, the buffer is not recycled by the failure directly but rather
during error handling which will eventually release the channel's resources.
-> this is now fixed as well
> Exceptions in RemoteInputChannel#notifyBufferAvailable() are not propagated
> to the responsible thread
> -----------------------------------------------------------------------------------------------------
>
> Key: FLINK-9755
> URL: https://issues.apache.org/jira/browse/FLINK-9755
> Project: Flink
> Issue Type: Bug
> Components: Network
> Affects Versions: 1.5.0
> Reporter: Nico Kruber
> Assignee: Nico Kruber
> Priority: Critical
> Labels: pull-request-available
> Fix For: 1.5.2, 1.6.0
>
>
> The credit-based flow control implementation of
> RemoteInputChannel#notifyBufferAvailable() does not forward errors (like the
> {{IllegalStateException}}) to the thread that is being notified. The calling
> code at {{LocalBufferPool#recycle}}, however, relies on the callback
> forwarding errors and completely ignores any failures.
> Therefore, we could end up with a program waiting forever for the callback
> and not even a failure message in the logs.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
