[
https://issues.apache.org/jira/browse/FLINK-22007?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17310470#comment-17310470
]
Zhu Zhu commented on FLINK-22007:
---------------------------------
Hi [~pnowojski], thanks for reporting this problem!
We also noticed the problem. It is actually a performance regression issue
caused by changes of FLINK-21326.
However, the change is needed for further optimization which will finally
dispose the performance regressions.
For this issue(PartitionReleaseInBatchJobBenchmark), FLINK-21332 will fix it.
We will try to fix it as soon as possible.
> PartitionReleaseInBatchJobBenchmarkExecutor seems to be failing
> ---------------------------------------------------------------
>
> Key: FLINK-22007
> URL: https://issues.apache.org/jira/browse/FLINK-22007
> Project: Flink
> Issue Type: Bug
> Components: Benchmarks, Runtime / Coordination
> Affects Versions: 1.13.0
> Reporter: Piotr Nowojski
> Priority: Blocker
> Fix For: 1.13.0
>
>
> Travis CI is failing:
> https://travis-ci.com/github/apache/flink-benchmarks/builds/221290042
> While there is also some problem with the Jenkins builds for the same
> benchmark.
> http://codespeed.dak8s.net:8080/job/flink-scheduler-benchmarks/232
> It would be also interesting for the future to understand why the Jenkins
> build is green and try to fix it (ideally, if some benchmarks fail, partial
> results should be still uploaded but the Jenkins build should be marked as
> failed). Otherwise issues like that can remain unnoticed for quite a bit of
> time.
> CC [~Thesharing] [~zhuzh]
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
