[
https://issues.apache.org/jira/browse/CASSANDRA-18766?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17873298#comment-17873298
]
Brandon Williams edited comment on CASSANDRA-18766 at 8/13/24 9:45 PM:
-----------------------------------------------------------------------
I was hoping this would provide a smoking gun that I could use to bisect and
finally find the culprit, so I converted this to a dtest
[here|https://github.com/driftx/cassandra-dtest/commit/68cb71920c9242aeb0e0cc3eb7e998175d9b5ccb].
Unfortunately however, this doesn't reproduce for me, on any of 4.0, 4.1, or
5.0 when run on isolated hardware. I also checked the raw SR numbers from each
and they are comparable across the branches.
-[~aswinkarthik] was this run on isolated hardware and does this consistently
reproduce for you?-
edit: I modified the test to check SR on all the nodes, and now I'm seeing
failures.
was (Author: brandon.williams):
I was hoping this would provide a smoking gun that I could use to bisect and
finally find the culprit, so I converted this to a dtest
[here|https://github.com/driftx/cassandra-dtest/commit/68cb71920c9242aeb0e0cc3eb7e998175d9b5ccb].
Unfortunately however, this doesn't reproduce for me, on any of 4.0, 4.1, or
5.0 when run on isolated hardware. I also checked the raw SR numbers from each
and they are comparable across the branches.
[~aswinkarthik] was this run on isolated hardware and does this consistently
reproduce for you?
> high speculative retries on v4.1.3
> ----------------------------------
>
> Key: CASSANDRA-18766
> URL: https://issues.apache.org/jira/browse/CASSANDRA-18766
> Project: Cassandra
> Issue Type: Bug
> Components: Consistency/Coordination
> Reporter: Ivans Novikovs
> Assignee: Brandon Williams
> Priority: Urgent
> Fix For: 4.1.x
>
> Attachments: screenshot-1.png, screenshot-2.png, signature.asc,
> signature.asc, v4.0.png, v4.1.png
>
>
> There are up to 10+ times higher speculative retries for reads on 4.1.3
> comparing to 4.0.7 and 4.1.2 when using QUORUM and default setting of 99p.
> On 4.1.3 after upgrade I see speculative retries for up to 35% of all reads
> for specific table. Latency for reads is stable around 500 microseconds.
> java 1.8.0_382 is used
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
