You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@cassandra.apache.org by "Andres de la Peña (Jira)" <ji...@apache.org> on 2022/07/01 16:09:00 UTC

[jira] [Commented] (CASSANDRA-17232) org.apache.cassandra.db.commitlog.GroupCommitLogTest tests failing on trunk

    [ https://issues.apache.org/jira/browse/CASSANDRA-17232?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17561519#comment-17561519 ] 

Andres de la Peña commented on CASSANDRA-17232:
-----------------------------------------------

I've rebased the patch and [run CI again|https://app.circleci.com/pipelines/github/adelapena/cassandra/1830/workflows/36bb7b87-ea27-43ca-906d-ef15c5ddc6bb]. It seems that the repeated runs still hit multiple timeouts. We could either split them here or separately, hoping that we are not missing something and the timeouts aren't due to a deadlock.

As for [retrying the calls to {{awaitTermination}} in a loop|https://github.com/apache/cassandra/pull/1651/commits/541300b10724d4547af6b89196fad2c0eab69ff0], maybe we can just add timeout parameters to that method and call it a single time with a longer timeout, more or less [this way|https://github.com/adelapena/cassandra/commit/caeca078d38ef0fe39c1db19a23aba67095bc83c].

Regarding [catching and ignoring any {{UncheckedIOException}}|https://github.com/apache/cassandra/pull/1651/commits/ffccbfcb779831a87b39bc6528e61fd15ef18228], I understand that those are raised when the call to {{awaitTermination}} hasn't finished properly. Maybe we can use the boolean returned by {{executor.awaitTermination}} to throw an assertion error if it hasn't finished within the timeline, so we don't continue to the segment deletion if we haven't properly stopped the executor. wdyt?

> org.apache.cassandra.db.commitlog.GroupCommitLogTest tests failing on trunk
> ---------------------------------------------------------------------------
>
>                 Key: CASSANDRA-17232
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-17232
>             Project: Cassandra
>          Issue Type: Bug
>          Components: CI
>            Reporter: Ekaterina Dimitrova
>            Assignee: Berenguer Blasi
>            Priority: Normal
>             Fix For: 4.1-beta, 4.1.x, 4.x
>
>
> I saw [here|https://jenkins-cm4.apache.org/job/Cassandra-devbranch/1344/testReport/junit/org.apache.cassandra.db.commitlog/GroupCommitLogTest/replayWithDiscard_3_/] and [here|https://ci-cassandra.apache.org/job/Cassandra-trunk/881/testReport/junit/org.apache.cassandra.db.commitlog/GroupCommitLogTest/replayWithDiscard_3__compression_3/] the same assertion failure:
>  
> {code:java}
> org.apache.cassandra.db.commitlog.GroupCommitLogTest.replayWithDiscard[3]
> Error Message
> expected:<204> but was:<0>
> Stacktrace
> junit.framework.AssertionFailedError: expected:<204> but was:<0> at org.apache.cassandra.db.commitlog.CommitLogTest.replayWithDiscard(CommitLogTest.java:883)
> {code}
>  
> Then I find that there are many other tests from GroupCommitLogTest failing with No such file found.
> Example:
> https://jenkins-cm4.apache.org/job/Cassandra-devbranch/1344/



--
This message was sent by Atlassian Jira
(v8.20.10#820010)

---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@cassandra.apache.org
For additional commands, e-mail: commits-help@cassandra.apache.org