[
https://issues.apache.org/jira/browse/CASSANDRA-17609?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17545438#comment-17545438
]
Andres de la Peña edited comment on CASSANDRA-17609 at 6/2/22 12:01 PM:
------------------------------------------------------------------------
Indeed, and probably also against 4.0. Here are the patches and CI after and
before applying the fix:
||PR||CI without patch||CI with patch||
|[4.0|https://github.com/apache/cassandra/pull/1660]|[j8|https://app.circleci.com/pipelines/github/adelapena/cassandra/1659/workflows/6c384b4e-f1f8-4825-adaf-927d277e9c05]
[j11|https://app.circleci.com/pipelines/github/adelapena/cassandra/1659/workflows/80844337-5701-4461-b809-c2474d6f8be3]|[j8|https://app.circleci.com/pipelines/github/adelapena/cassandra/1661/workflows/622c57ae-6db9-4975-808c-fb82f8901d69]
[j11|https://app.circleci.com/pipelines/github/adelapena/cassandra/1661/workflows/3f0d0e9e-20a6-4058-adf9-84ca003733de]|
|[4.1|https://github.com/apache/cassandra/pull/1661]|[j8|https://app.circleci.com/pipelines/github/adelapena/cassandra/1658/workflows/6200e170-d025-46f9-8472-0fc5a6d47329]
[j11|https://app.circleci.com/pipelines/github/adelapena/cassandra/1658/workflows/435c318f-b4ac-49a3-817a-391f27f7a6a8]|[j8|https://app.circleci.com/pipelines/github/adelapena/cassandra/1660/workflows/281e2b21-c834-4f80-8cf7-c590655091e2]
[j11|https://app.circleci.com/pipelines/github/adelapena/cassandra/1660/workflows/122ceb75-218f-4a30-b08e-da4aefadeae5]|
|[trunk|https://github.com/apache/cassandra/pull/1657]|[j8|https://app.circleci.com/pipelines/github/adelapena/cassandra/1641/workflows/7dd9bb59-3e8f-4889-ab84-bcfe4aca6297]
[j11|https://app.circleci.com/pipelines/github/adelapena/cassandra/1641/workflows/4c70c9b7-4faa-4877-b874-b89a5d6f3a4a]|[j8|https://app.circleci.com/pipelines/github/adelapena/cassandra/1656/workflows/f911bcd3-2ce9-4778-81ae-650f964253c4]
[j11|https://app.circleci.com/pipelines/github/adelapena/cassandra/1656/workflows/c7d673c3-aed6-4a21-8cb6-ba867d319d0d]|
Note that I haven't been able to reproduce the test failures on 4.0. I suspect
that might be due to the different flushes introduced in 4.1 by
CASSANDRA-17034, which often flushes the entire test keyspace instead of the
tested column family. Flakiness is quite low so the problem might still be
there, but reproducing less frequently. Thus, I think I'd also apply the
proposed fix to 4.0. wdyt?
was (Author: adelapena):
Indeed, and probably also against 4.0. Here are the patches and CI after and
before applying the fix:
||PR||CI without patch||CI with patch||
|[4.0|https://github.com/apache/cassandra/pull/1660]|[j8|https://app.circleci.com/pipelines/github/adelapena/cassandra/1659/workflows/6c384b4e-f1f8-4825-adaf-927d277e9c05]
[j11|https://app.circleci.com/pipelines/github/adelapena/cassandra/1659/workflows/80844337-5701-4461-b809-c2474d6f8be3]|[j8|https://app.circleci.com/pipelines/github/adelapena/cassandra/1661/workflows/622c57ae-6db9-4975-808c-fb82f8901d69]
[j11|https://app.circleci.com/pipelines/github/adelapena/cassandra/1661/workflows/3f0d0e9e-20a6-4058-adf9-84ca003733de]|
|[4.1|https://github.com/apache/cassandra/pull/1661]|[j8|https://app.circleci.com/pipelines/github/adelapena/cassandra/1658/workflows/6200e170-d025-46f9-8472-0fc5a6d47329]
[j11|https://app.circleci.com/pipelines/github/adelapena/cassandra/1658/workflows/435c318f-b4ac-49a3-817a-391f27f7a6a8]|[j8|https://app.circleci.com/pipelines/github/adelapena/cassandra/1660/workflows/281e2b21-c834-4f80-8cf7-c590655091e2]
[j11|https://app.circleci.com/pipelines/github/adelapena/cassandra/1660/workflows/122ceb75-218f-4a30-b08e-da4aefadeae5]|
|[trunk|https://github.com/apache/cassandra/pull/1657]|[j8|https://app.circleci.com/pipelines/github/adelapena/cassandra/1641/workflows/7dd9bb59-3e8f-4889-ab84-bcfe4aca6297]
[j11|https://app.circleci.com/pipelines/github/adelapena/cassandra/1641/workflows/4c70c9b7-4faa-4877-b874-b89a5d6f3a4a]|[j8|https://app.circleci.com/pipelines/github/adelapena/cassandra/1656/workflows/f911bcd3-2ce9-4778-81ae-650f964253c4]
[j11|https://app.circleci.com/pipelines/github/adelapena/cassandra/1656/workflows/c7d673c3-aed6-4a21-8cb6-ba867d319d0d]|
Note that I haven't been to reproduce the test failures on 4.0. I suspect that
might be due to the different flushes introduced in 4.1 by CASSANDRA-17034,
which often flushes the entire test keyspace instead of the tested column
family. Flakiness is quite low so the problem might still be there, but
reproducing less frequently. Thus, I think I'd also apply the proposed fix to
4.0. wdyt?
> Fix
> org.apache.cassandra.db.compaction.CompactionsCQLTest.testLCSThresholdParams-cdc
> (from org.apache.cassandra.db.compaction.CompactionsCQLTest-cdc)
> -----------------------------------------------------------------------------------------------------------------------------------------------------
>
> Key: CASSANDRA-17609
> URL: https://issues.apache.org/jira/browse/CASSANDRA-17609
> Project: Cassandra
> Issue Type: Bug
> Components: CI
> Reporter: Ekaterina Dimitrova
> Assignee: Andres de la Peña
> Priority: Normal
> Fix For: 4.1-beta
>
>
> {color:#172b4d}[https://ci-cassandra.apache.org/job/Cassandra-4.1/5/testReport/org.apache.cassandra.db.compaction/CompactionsCQLTest/testLCSThresholdParams_cdc_2/]{color}
> h3.
> {code:java}
> Error Message
> expected:<50> but was:<51>
> Stacktrace
> junit.framework.AssertionFailedError: expected:<50> but was:<51> at
> org.apache.cassandra.db.compaction.CompactionsCQLTest.testLCSThresholdParams(CompactionsCQLTest.java:419)
> at java.base/jdk.internal.reflect.NativeMethodAccessorImpl.invoke0(Native
> Method) at
> java.base/jdk.internal.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62)
> at
> java.base/jdk.internal.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
> {code}
>
--
This message was sent by Atlassian Jira
(v8.20.7#820007)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
