Hi,

We are continuing testing profiler. Thank you for your help, but we have more 
questions.

After your recomendations we increased memory for profiler topology and set 
"topology.max.spout.pending" to null as default and set "profiler.workers" to 1 
as default too. But with such configuration we had very slow event processing 
pace and we tried to set "topology.max.spout.pending" to 20000 and also set 
"profiler.hbase.batch" to 1000. 

So processing pace was good and kafka lag for profiler consumer began to 
decrease. But in traffic peak and we had this  error again:

"org.apache.kafka.clients.consumer.CommitFailedException: Commit cannot be 
completed since the group has already rebalanced and assigned the partitions to 
another member. This means that the time between subsequent calls to poll() was 
longer than the configured max.poll.interval.ms, which typically implies that 
the poll loop is spending too much time message processing. You can address 
this either by increasing the session timeout or by reducing the maximum size 
of batches returned in poll() with max.poll.records. at 
org.apache.kafka.clients.consumer.internals.ConsumerCoordinator.sendOffsetCommitRequest(ConsumerCoordinator.java:798)
 at 
org.apache.kafka.clients.consumer.internals.ConsumerCoordinator.commitOffsetsSync(ConsumerCoordinator.java:681)
 at 
org.apache.kafka.clients.consumer.KafkaConsumer.commitSync(KafkaConsumer.java:1416)
 at 
org.apache.kafka.clients.consumer.KafkaConsumer.commitSync(KafkaConsumer.java:1377)
 at 
org.apache.storm.kafka.spout.KafkaSpout.commitOffsetsForAckedTuples(KafkaSpout.java:534)
 at org.apache.storm.kafka.spout.KafkaSpout.nextTuple(KafkaSpout.java:293) at 
org.apache.storm.daemon.executor$fn__10149$fn__10164$fn__10197.invoke(executor.clj:660)
 at org.apache.storm.util$async_loop$fn__1221.invoke(util.clj:484) at 
clojure.lang.AFn.run(AFn.java:22) at java.lang.Thread.run(Thread.java:745)"

And after this error (or maybe this error it isn't the reason of that) we 
didn't see any event processing, and the kafka lag for profiler began to 
increase. As we undesrstood worker was restart after the error. 

So why did processing stopped? 

Thanks.

Reply via email to