Hey! We need to investigate with http://ci.hive.apache.org/job/hive-flaky-check/ and fix it, there is no other way. Recently we made HIVE-27717 <https://issues.apache.org/jira/browse/HIVE-27717> about improving precommit tests, but Zoltan is right about that we can improve the flaky check itself first to give better logs, so any tickets under HIVE-27717 <https://issues.apache.org/jira/browse/HIVE-27717> should be considered about flaky check job (too).
I can see that the pipeline script of flaky check <http://ci.hive.apache.org/job/hive-flaky-check/configure> is in Jenkins and I cannot see/recall any seed job that generates this test job, so we should consider adding this to the hive repo - similarly to the main precommit job's Jenkinsfile <https://github.com/apache/hive/blob/master/Jenkinsfile> - if possible. Regards, Laszlo Bodor Stamatis Zampetakis <zabe...@gmail.com> ezt írta (időpont: 2023. okt. 4., Sze, 10:53): > Hello, > > I am looking into this as part of HIVE-27695. Till we commit the fix > feel free to ignore the tests that are failing with OOM related errors > in TestMiniTezCliDriver. > > Best, > Stamatis > > [1] https://issues.apache.org/jira/browse/HIVE-27695 > > On Wed, Oct 4, 2023 at 9:44 AM Zoltán Rátkai > <zrat...@cloudera.com.invalid> wrote: > > > > Hi Everyone, > > > > the hive build environment is really unstable lately. I get many times > > exceptions like this, or test just fail randomly, then next time another: > > * java.lang.OutOfMemoryError: GC overhead limit exceeded* > > > > see here: > > > http://ci.hive.apache.org/blue/organizations/jenkins/hive-precommit/detail/PR-4690/16/tests/ > > > > What can we do to avoid this? It takes a lot of effort to always > retrigger > > the build and hope next time it succeeds. > > > > Regards, > > > > Zoltán Rátkai >
