Kafka 0.8.2 consumer offset checker throwing kafka.common.NotCoordinatorForConsumerException

[Kartheek Karra <kk...@salesforce.com>]

We recently upgraded kafka in our production environment cluster of 5
brokers from 0.8.0 to 0.8.2. Since then the consumerOffsetChecker script is
unable to fetch offset due to
kafka.common.NotCoordinatorForConsumerException.
Note I'm able to run the 'consumerOffsetChecker' from an older version
0.8.0 successfully without any exceptions against the same upgraded
cluster. Also we haven't migrated to kafka for offsets and are still using
the default zookeeper. Another piece of information here is that kafka is
always picking the same host as coordinator for fetching offset and we had
to reimage that host just after the upgrade. Haven't been able to reproduce
this in our test environments yet.
Any clue what might be wrong here ? Let me know if more details are needed
anywhere.


Thanks,
Kartheek

Re: Kafka 0.8.2 consumer offset checker throwing kafka.common.NotCoordinatorForConsumerException

[Kartheek Karra <kk...@salesforce.com>]

Update here, we resolved this by deleting the kafka-data directory on that
host ( which had file inconsistencies from 'fsck' run log last week in the
kafka-data dir ) and restarting kafka. Note we also never reimaged the host
( that was another host, which we got confused over ).

Thanks,
Kartheek

On Tue, Apr 28, 2015 at 6:36 PM, Kartheek Karra <kk...@salesforce.com>
wrote:

> We recently upgraded kafka in our production environment cluster of 5
> brokers from 0.8.0 to 0.8.2. Since then the consumerOffsetChecker script is
> unable to fetch offset due to
> kafka.common.NotCoordinatorForConsumerException.
> Note I'm able to run the 'consumerOffsetChecker' from an older version
> 0.8.0 successfully without any exceptions against the same upgraded
> cluster. Also we haven't migrated to kafka for offsets and are still using
> the default zookeeper. Another piece of information here is that kafka is
> always picking the same host as coordinator for fetching offset and we had
> to reimage that host just after the upgrade. Haven't been able to reproduce
> this in our test environments yet.
> Any clue what might be wrong here ? Let me know if more details are needed
> anywhere.
>
>
> Thanks,
> Kartheek
>