[
https://issues.apache.org/jira/browse/FLINK-17040?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17077773#comment-17077773
]
Zhijiang commented on FLINK-17040:
----------------------------------
Thanks for reporting this [~wenlong.lwl]!
It is indeed caused by FLINK-16537. But it is strange that the Azure shows
success before the respective PR merged, and actually the Azure did not execute
any tests in that PR before.
The fix should be easy to only return empty list in
SavepointEnvironment#getAllWriters, instead of throwingÂ
UnsupportedOperationException in some dummy methods. The root cause for Azure
should be investigated further afterwards.
> SavepointWriterITCase broken
> ----------------------------
>
> Key: FLINK-17040
> URL: https://issues.apache.org/jira/browse/FLINK-17040
> Project: Flink
> Issue Type: Bug
> Reporter: Wenlong Lyu
> Assignee: Yingjie Cao
> Priority: Major
>
> I think it is because of the change of flink-16537 which create partition
> writer in beforeInvoke [~zhijiang]
> Caused by: java.lang.UnsupportedOperationException: This method should never
> be called
> at
> org.apache.flink.state.api.runtime.SavepointEnvironment.getAllWriters(SavepointEnvironment.java:242)
> at
> org.apache.flink.streaming.runtime.tasks.StreamTask.lambda$beforeInvoke$0(StreamTask.java:439)
> at
> org.apache.flink.streaming.runtime.tasks.StreamTaskActionExecutor$1.runThrowing(StreamTaskActionExecutor.java:47)
> at
> org.apache.flink.streaming.runtime.tasks.StreamTask.beforeInvoke(StreamTask.java:433)
> at
> org.apache.flink.streaming.runtime.tasks.StreamTask.invoke(StreamTask.java:454)
> at
> org.apache.flink.state.api.output.BoundedOneInputStreamTaskRunner.mapPartition(BoundedOneInputStreamTaskRunner.java:76)
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
