[
https://issues.apache.org/jira/browse/CASSANDRA-16387?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17279066#comment-17279066
]
Caleb Rackliffe edited comment on CASSANDRA-16387 at 2/5/21, 7:11 PM:
----------------------------------------------------------------------
Given the clean test runs, I'm posting the 3.0 and 3.11 versions of the
solution that avoids globally pushing mutations from the distributed system
keyspace schema changes at startup. (Nodes still passively announce their
schema version.)
While this change appears to consistently fix the problem in local test runs
(3.0 -> trunk in-JVM upgrade tests), the next step would be modifying the
CircleCI config to use my 3.0 and 3.11 branches rather than the base branches
to build dtest JARs.
|[CASSANDRA-16387-3.0|https://github.com/apache/cassandra/pull/885]|[CircleCI|https://app.circleci.com/pipelines/github/maedhroz/cassandra/223/workflows/e60f131d-326e-4241-92d4-f51b9043eb62]|
|[CASSANDRA-16387-3.11|https://github.com/apache/cassandra/pull/886]|[CircleCI|https://app.circleci.com/pipelines/github/maedhroz/cassandra/225/workflows/0bea0ed5-9dd6-4a7e-8cae-3a6c5ea7942d]|
UPDATE: The trunk version also looks not to have created any new test failures.
|[CASSANDRA-16387-trunk|https://github.com/apache/cassandra/pull/887]|[CircleCI|https://app.circleci.com/pipelines/github/maedhroz/cassandra/227/workflows/93a9d7d5-df87-41bd-a4be-5ea689416f30]|
In any case, judging from these results, either it *is* safe to avoid the
global push, or we just lack the tests/in-code documentation to make it obvious
that it isn't.
was (Author: maedhroz):
Given the clean test runs, I'm posting the 3.0 and 3.11 versions of the
solution that avoids globally pushing mutations from the distributed system
keyspace schema changes at startup. (Nodes still passively announce their
schema version.)
While this change appears to consistently fix the problem in local test runs
(3.0 -> trunk in-JVM upgrade tests), the next step would be modifying the
CircleCI config to use my 3.0 and 3.11 branches rather than the base branches
to build dtest JARs.
|[CASSANDRA-16387-3.0|https://github.com/apache/cassandra/pull/885]|[CircleCI|https://app.circleci.com/pipelines/github/maedhroz/cassandra/223/workflows/e60f131d-326e-4241-92d4-f51b9043eb62]|
|[CASSANDRA-16387-3.11|https://github.com/apache/cassandra/pull/886]|[CircleCI|https://app.circleci.com/pipelines/github/maedhroz/cassandra/225/workflows/0bea0ed5-9dd6-4a7e-8cae-3a6c5ea7942d]|
UPDATE: The trunk version also looks not to have created any new test failures.
|[CASSANDRA-16387-trunk|https://github.com/apache/cassandra/pull/887]|[CircleCI|https://app.circleci.com/pipelines/github/maedhroz/cassandra/227/workflows/c8f12d79-8f8c-420c-965c-d8188199822b]|
In any case, judging from these results, either it *is* safe to avoid the
global push, or we just lack the tests/in-code documentation to make it obvious
that it isn't.
> UpgradeTest sporadically failing on schema updates
> --------------------------------------------------
>
> Key: CASSANDRA-16387
> URL: https://issues.apache.org/jira/browse/CASSANDRA-16387
> Project: Cassandra
> Issue Type: Bug
> Components: Test/dtest/java
> Reporter: Caleb Rackliffe
> Assignee: Caleb Rackliffe
> Priority: Normal
> Fix For: 3.0.x, 3.11.x, 4.0-rc
>
> Time Spent: 1.5h
> Remaining Estimate: 0h
>
> We’ve observed {{UpdateTest}} failing during what appears to be a schema
> change:
> https://app.circleci.com/pipelines/github/maedhroz/cassandra/192/workflows/ed5305e6-e4f9-420e-9f0a-6153333746dc/jobs/1068
> It almost looks like the Gossiper can’t find its own endpoint state in the
> endpoint state map, and the failure is not consistent, which might suggest a
> race.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
