[
https://issues.apache.org/jira/browse/FLINK-8816?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16381731#comment-16381731
]
ASF GitHub Bot commented on FLINK-8816:
---------------------------------------
Github user sihuazhou commented on the issue:
https://github.com/apache/flink/pull/5604
@tillrohrmann I just doubt whether the new `TaskExecutor` is a faker, If
it's a faker, it will evict the old valid one. I'd like to close this If you
think this can't be happen ;).
> Remove the oldWorker only after starting newWorker successfully in
> registerTaskExecutorInternal()
> -------------------------------------------------------------------------------------------------
>
> Key: FLINK-8816
> URL: https://issues.apache.org/jira/browse/FLINK-8816
> Project: Flink
> Issue Type: Bug
> Components: Distributed Coordination
> Affects Versions: 1.5.0
> Reporter: Sihua Zhou
> Assignee: Sihua Zhou
> Priority: Major
> Fix For: 1.5.0
>
>
> We should remove the oldWorker only after starting newWorker successfully in
> {{ResourceManager#registerTaskExecutorInternal}} and javadoc for
> {{SlotProvider#allocateSlot}} is outdated.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
