[
https://issues.apache.org/jira/browse/CASSANDRA-14447?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16478409#comment-16478409
]
Jason Brown commented on CASSANDRA-14447:
-----------------------------------------
TL;DR The dtest failures in my branches are almost exactly the same ones as in
the HEAD for 3.0/3.11. Thus I feel my patch did not introduce any new
regressions.
HEAD
- 3.11 dtest fails: no-vnodes & vnodes - both 6 fails (5/6 are the same between
the two runs)
- 3.0 dtest fails- no-vnodes & vnodes - both 14 fails (all are the same between
the two runs)
14447 branches
- 3.11 dtest fails: no-vnodes & vnodes - 7 and 6 fails (most are overlap), and
are the same as 3.11 HEAD
- 3.0 dtest fails: no-vnodes & vnodes - both 14 fails, and seem to be the same
as the 3.0 HEAD
> Cleanup StartupClusterConnectivityChecker and PING Verb
> -------------------------------------------------------
>
> Key: CASSANDRA-14447
> URL: https://issues.apache.org/jira/browse/CASSANDRA-14447
> Project: Cassandra
> Issue Type: Improvement
> Reporter: Jason Brown
> Assignee: Jason Brown
> Priority: Minor
> Fix For: 3.0.x, 3.11.x, 4.0.x
>
>
> This is a followup to CASSANDRA-13993. After an internal review,
> [~iamaleksey] had some concerns wrt to the VERB choices as was committed;
> this was discussed on CASSANDRA-13993, after commit. Further, he pointed out
> (and provided) some optimizations for {{StartupClusterConnectivityChecker}}
> itself. While testing some of the proposed changes, I also discovered a small
> bug where the timeout for waiting for the response to the PING message was
> dramtically shorter than the overall timeout that is configured for
> {{StartupClusterConnectivityChecker}}.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
