The MaxLag mbean is only valid for an active consumer. So while the consumer is actively running, it should be accurate (as far as telling you the lag on the partition that you are consuming that has the most lag). However, if you look at the output from the ConsumerOffsetChecker, you see that the partition you are looking at (0) has no owner. That means that there is no consumer in the group that is actively consuming the partition. That means that MaxLag will not be accurate.
Now, the other thing that could be going on here is that you could have a consumer that has lost its Zookeeper connection, but is still consuming messages. In that case, MaxLag would be correct for that consumer, but with offsets not being committed, it means that when you stop that consumer and another one picks up the partition, it will start from the last committed offset, which means it¹s going to duplicate a lot of messages. -Todd On 7/26/14, 8:03 AM, "ravi singh" <rrs120...@gmail.com> wrote: >The Max lag Mbean is defined as "Number of messages the consumer lags >behind the producer". Now when I read the Mbean value it give me the count >as 0 (and occasionally some value like 130 or 340 ) > >ConsumerFetcherManager.test-consumer-group-MaxLag count = 0 > > >But when I use the kafka.tools.ConsumerOffsetChecker I get following >as Lag value > >*Group Topic Pid Offset >logSize *Lag* Owner* >test-consumer-group kafka-test 0 275985215 >276195685 *210470* none > > >Are both the lags not related? Or is it like I am capturing the lag MaxLag >incorrectly? > >-- >*Regards,* >*Ravi*
