[
https://issues.apache.org/jira/browse/CASSANDRA-4840?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13480468#comment-13480468
]
Brandon Williams commented on CASSANDRA-4840:
---------------------------------------------
I believe this may be evidence of the issue I've heard reported where nodes are
still trying to connect to dead IPs that were removed. I suspect a message
might be getting stuck in OTC and causing this.
As far as the FD goes, we definitely remove it there in
Gossiper.removeEndpoint, so something must be adding it back.
> remnants of removed nodes remain after removal
> ----------------------------------------------
>
> Key: CASSANDRA-4840
> URL: https://issues.apache.org/jira/browse/CASSANDRA-4840
> Project: Cassandra
> Issue Type: Bug
> Components: Core
> Affects Versions: 1.0.10
> Reporter: Jeremy Hanna
> Priority: Minor
>
> After nodes are removed from the ring and no longer appear in any of the
> nodes' nodetool ring output, some of the dead nodes show up in the
> o.a.c.net.FailureDetector SimpleStates metadata. Also, some of the JMX stats
> are updating for the removed nodes (ie RecentTimeoutsPerHost and
> ResponsePendingTasks).
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
