[
https://issues.apache.org/jira/browse/CASSANDRA-21327?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=18084621#comment-18084621
]
Michael Semb Wever commented on CASSANDRA-21327:
------------------------------------------------
bq.
dtest-latest.client_request_metrics_test.TestClientRequestMetrics.test_client_request_metrics
This test shouldn't be applied. With Paxos v2's optimistic reads, contention
is much less likely to occur because v2 can complete reads in 1-2 round trips
instead of 3-4 and uses optimistic reads that don't require a full Paxos round
unless there's in-progress state.
The test is running 20 concurrent SERIAL reads on the same partition (k=0),
trying to generate contention. That doesn't happen with v2.
> Fix -latest.yml Paxos tests after we made Paxos v2 default in
> cassandra_latest.yaml in CASSANDRA-21316
> ------------------------------------------------------------------------------------------------------
>
> Key: CASSANDRA-21327
> URL: https://issues.apache.org/jira/browse/CASSANDRA-21327
> Project: Apache Cassandra
> Issue Type: Bug
> Reporter: Stefan Miklosovic
> Assignee: Michael Semb Wever
> Priority: Normal
> Fix For: 5.0.x, 6.0-alpha, 7.x
>
>
> TBD which they are
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
