[
https://issues.apache.org/jira/browse/CASSANDRA-19823?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17873258#comment-17873258
]
Caleb Rackliffe edited comment on CASSANDRA-19823 at 8/13/24 4:46 PM:
----------------------------------------------------------------------
Preliminary testing indicates that CASSANDRA-19830 might be enough to fix this.
If it is, there are still some things we could do to tighten things up in the
patch attached to this Jira...
I think the hypothesis that best fits the data, at this point, is that the
cycle in the object graph above is transient, and only appears when the test
stalls long enough for the detector to see it while a query is in flight. In
any case, profiling revealed that the biggest component of the SAI tests'
runtime was the batch commitlog, and overall test runtime has come down
drastically after moving to periodic/10s.
was (Author: maedhroz):
Preliminary testing indicates that CASSANDRA-19830 might be enough to fix this.
If it is, there are still some things we could do to tighten things up in the
patch attached to this Jira...
> Investigate timeouts in tests extending RandomIntersectionTester
> ----------------------------------------------------------------
>
> Key: CASSANDRA-19823
> URL: https://issues.apache.org/jira/browse/CASSANDRA-19823
> Project: Cassandra
> Issue Type: Task
> Components: Feature/SAI, Test/unit
> Reporter: Stefan Miklosovic
> Assignee: Caleb Rackliffe
> Priority: Normal
> Fix For: 5.0-rc, 5.x
>
> Attachments:
> TEST-org.apache.cassandra.index.sai.cql.intersection.TwoRegularOneStaticPartitionIntersectionTest.log
>
>
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
