[ 
https://issues.apache.org/jira/browse/CASSANDRA-17810?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17578554#comment-17578554
 ] 

Josh McKenzie commented on CASSANDRA-17810:
-------------------------------------------

||Item|Link|
|PR|[link|https://github.com/apache/cassandra/pull/1782]|
|JDK8 
CI|[link|https://app.circleci.com/pipelines/github/josh-mckenzie/cassandra/267/workflows/9e091d31-cbf6-4462-8c99-eee43dc8cfd5]|
|JDK11 
CI|[link|https://app.circleci.com/pipelines/github/josh-mckenzie/cassandra/267/workflows/24b9380f-3f05-41f8-b12f-c51da06e8129]|

This is a fairly rare edge case with no risk of correctness errors so I'm 
personally comfortable targeting just 4.2 with it. If anyone else feels 
strongly about it going to prior branches and wants to backport it I'd be happy 
to review that effort.

> Revise timeout handling on queries triggering timeout introduced by 
> CASSANDRA-7392
> ----------------------------------------------------------------------------------
>
>                 Key: CASSANDRA-17810
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-17810
>             Project: Cassandra
>          Issue Type: Bug
>          Components: Messaging/Internode
>            Reporter: Josh McKenzie
>            Assignee: Josh McKenzie
>            Priority: Normal
>
> In situations where a replica times out with the code introduced in 
> CASSANDRA-7392, they drop the timed-out query on the floor without notifying 
> the coordinator; this doesn't become apparent coordinator-side until we go to 
> serialize the response. While the results of this (i.e. missed data) should 
> mostly be fixed by short read protection on repeat requests, the coordinator 
> can in some cases throw an exception like {{IllegalStateException: 
> UnfilteredRowIterator for keyspace.table has an open RT bound as its last 
> item}} which is decidedly unfriendly.
> We should introduce a specific exception to allow replicas to notify a 
> coordinator something has timed out and been dropped on the floor so the 
> coordinator can appropriately handle that scenario and abort the query at 
> that time.



--
This message was sent by Atlassian Jira
(v8.20.10#820010)

---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org
For additional commands, e-mail: commits-h...@cassandra.apache.org

Reply via email to