[
https://issues.apache.org/jira/browse/CASSANDRA-7955?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14137409#comment-14137409
]
Benedict commented on CASSANDRA-7955:
-------------------------------------
That would work, I just favoured letting the user know they were screwing
around with this.
If we do it automatically, we should probably gossip the ring delay and take
the max of all we can see, though, while we're at it. If one node configures a
high write timeout, the whole cluster is potentially permitted to get into an
inconsistent state
> Range movements can violate consistency if RING_DELAY <=
> write_request_timeout_in_ms
> ------------------------------------------------------------------------------------
>
> Key: CASSANDRA-7955
> URL: https://issues.apache.org/jira/browse/CASSANDRA-7955
> Project: Cassandra
> Issue Type: Improvement
> Reporter: Benedict
> Priority: Minor
>
> Cassandra should probably refuse to start if RING_DELAY is too close to (or
> below) write_request_timeout_ms, because we depend on this for
> correctness/consistency during range movements
> We should probably also consider throwing a WriteTimeoutException _even if we
> don't get interrupted by the timeout, since there are reasons due to
> scheduling or system overload this could not happen (though it is unlikely to
> be significant enough to have an impact, it's better safe than sorry)
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
