[ 
https://issues.apache.org/jira/browse/CASSANDRA-17594?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17539811#comment-17539811
 ] 

Ekaterina Dimitrova commented on CASSANDRA-17594:
-------------------------------------------------

{quote}CI runs are
{quote}
Thanks! I think we miss only the python upgrade tests for 3.0
{quote}Because back porting that fix to 3.x would be its own effort, so I felt 
it was out of scope for this ticket. That behavior has been around for a long 
time so if you were bitten by it... upgrade to 4.0? =D
{quote}
While we often agree to not backport something, I am not sure that is what the 
community wants here? If we are going to silently ignore this bug in our tests, 
I would like to see some visibility and common agreement on that.

My concerns – the original ticket is critical and mentions the bug affects 
availability but it doesn't even mention that the issue exists on previous 
versions. I is said it is there since 4.0? While I absolutely do not expect 
fixing the problem or porting the patch here as part of this ticket, I am not 
sure we should just decide to silence it. Probably [~maedhroz] can comment 
better as the person most familiar with that problem. 

> Fix flaky python-tests due to connection getting closed
> -------------------------------------------------------
>
>                 Key: CASSANDRA-17594
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-17594
>             Project: Cassandra
>          Issue Type: Bug
>          Components: Test/dtest/python
>            Reporter: David Capwell
>            Assignee: David Capwell
>            Priority: Normal
>             Fix For: NA
>
>          Time Spent: 40m
>  Remaining Estimate: 0h
>
> We log unknown exceptions at the networking level, which includes case where 
> the remote side closes the connection (such as the cases caused by shutting 
> down), depending on how quickly the instances shutdown, this could cause 
> python-dtest to fail for random tests with a message such as
> {code}
> Unexpected error found in node logs (see stdout for full details). Errors: 
> [WARN  [epollEventLoopGroup-5-9] 2022-05-03T16:47:03,800 
> ExceptionHandlers.java:134 - Unknown exception in client networking
> io.netty.channel.unix.Errors$NativeIoException: writeAddress(..) failed: 
> Connection reset by peer, WARN  [epollEventLoopGroup-5-9] 
> 2022-05-03T16:47:03,800 ExceptionHandlers.java:134 - Unknown exception in 
> client networking
> io.netty.channel.unix.Errors$NativeIoException: writeAddress(..) failed: 
> Connection reset by peer]
> {code}



--
This message was sent by Atlassian Jira
(v8.20.7#820007)

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

Reply via email to