[ 
https://issues.apache.org/jira/browse/KAFKA-305?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13235295#comment-13235295
 ] 

Prashanth Menon commented on KAFKA-305:
---------------------------------------

Thanks for review, Jun.

1. Will do.
2. So that test actually exposed the issue to begin with - the initial send 
would fail and then hang forever when attempting to refresh the topic metadata. 
 Regardless, I'll create a separate more direct test for timeouts.  On my local 
machine, this test seems to be unpredictable around 30% of the time.  In these 
cases, it seems like the ephemeral broker nodes aren't removed from ZK and 
bringing back up a broker after shutdown throws a "Broker already exists" 
exception.  Is anyone else experiencing it or just me?  Increasing the wait 
time after shutdown helps but not 100%.
3. 1,2,3 Sounds fair.

I should be able to get a patch in for this by Friday.  Then continue on 
KAFKA-49 over the weekend and get it in on Saturday or Sunday should the review 
go okay.  Apologies for the delays :(
                
> SyncProducer does not correctly timeout
> ---------------------------------------
>
>                 Key: KAFKA-305
>                 URL: https://issues.apache.org/jira/browse/KAFKA-305
>             Project: Kafka
>          Issue Type: Bug
>          Components: core
>    Affects Versions: 0.7, 0.8
>            Reporter: Prashanth Menon
>            Priority: Critical
>         Attachments: KAFKA-305-v1.patch
>
>
> So it turns out that using the channel in SyncProducer like we are to perform 
> blocking reads will not trigger socket timeouts (though we set it) and will 
> block forever which is bad.  This bug identifies the issue: 
> http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=4614802 and this article 
> presents a potential work-around: 
> http://stackoverflow.com/questions/2866557/timeout-for-socketchannel for 
> workaround. The work-around is a simple solution that involves creating a 
> separate ReadableByteChannel instance for timeout-enabled reads.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: 
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira

        

Reply via email to