[
https://issues.apache.org/jira/browse/CASSANDRA-18902?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17794202#comment-17794202
]
Jacek Lewandowski commented on CASSANDRA-18902:
-----------------------------------------------
I changed the following things:
- the version is not re-read from the properties files - once it has been read,
its value is memoized
- migration coordinator adds some small backoff delay after a failed fetch from
other node (this allows to avoid flooding logs)
- fixed one system property that wasn't cleared in the test after being set
- fixed dtest framework to not overwrite system properties which were
explicitly set in the test case
- fixed a nit timing issue in MigrationCoordinator (unit test)
Also, regarding that failure with address being unavailable, I managed to
reproduce the problem - it looks like a dtest framework problem so I created
https://issues.apache.org/jira/browse/CASSANDRA-19181 to address that.
So far, I've run the whole MigrationCoordinator class 1000 times and also other
1000 time the test case which failed alone
> Test failure:
> org.apache.cassandra.distributed.test.MigrationCoordinatorTest.explicitEndpointIgnore
> ---------------------------------------------------------------------------------------------------
>
> Key: CASSANDRA-18902
> URL: https://issues.apache.org/jira/browse/CASSANDRA-18902
> Project: Cassandra
> Issue Type: Bug
> Components: Test/dtest/java
> Reporter: Jacek Lewandowski
> Assignee: Jacek Lewandowski
> Priority: Normal
> Fix For: 5.0-rc, 5.x
>
>
> Repeated run from `cassandra-4.1`
> [https://app.circleci.com/pipelines/github/jacek-lewandowski/cassandra/941/workflows/46fc6cb7-135e-4862-b9d3-6996c0993de8]
>
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
