INFO Marking the coordinator XXXX dead

[stam...@gmail.com]

Hey,

I'm running a connector on kafka 0.9.0.0 distributed.

Every once in a while I see in the log a lot of "coordinator XXXXXXXXX dead" Infos and everything continue work fine.

What does this error means? Is it ok?

INFO org.apache.kafka.connect.runtime.WorkerSinkTask@522d24b3 Committing offsets (org.apache.kafka.connect.runtime.WorkerSinkTask:187)

INFO org.apache.kafka.connect.runtime.WorkerSinkTask@f5c6c04 Committing offsets (org.apache.kafka.connect.runtime.WorkerSinkTask:187)

INFO Marking the coordinator 2147483638 dead. (org.apache.kafka.clients.consumer.internals.AbstractCoordinator:540)

INFO Marking the coordinator 2147483638 dead. (org.apache.kafka.clients.consumer.internals.AbstractCoordinator:540)

INFO Marking the coordinator 2147483638 dead. (org.apache.kafka.clients.consumer.internals.AbstractCoordinator:540)

INFO Marking the coordinator 2147483638 dead. (org.apache.kafka.clients.consumer.internals.AbstractCoordinator:540)

INFO Marking the coordinator 2147483638 dead. (org.apache.kafka.clients.consumer.internals.AbstractCoordinator:540)

INFO Marking the coordinator 2147483638 dead. (org.apache.kafka.clients.consumer.internals.AbstractCoordinator:540)

INFO Marking the coordinator 2147483638 dead. (org.apache.kafka.clients.consumer.internals.AbstractCoordinator:540)

INFO Marking the coordinator 2147483638 dead. (org.apache.kafka.clients.consumer.internals.AbstractCoordinator:540)

INFO Marking the coordinator 2147483638 dead. (org.apache.kafka.clients.consumer.internals.AbstractCoordinator:540)

INFO Marking the coordinator 2147483638 dead. (org.apache.kafka.clients.consumer.internals.AbstractCoordinator:540)

INFO org.apache.kafka.connect.runtime.WorkerSinkTask@585e3210 Committing offsets (org.apache.kafka.connect.runtime.WorkerSinkTask:187)

INFO org.apache.kafka.connect.runtime.WorkerSinkTask@522d24b3 Committing offsets (org.apache.kafka.connect.runtime.WorkerSinkTask:187)

INFO org.apache.kafka.connect.runtime.WorkerSinkTask@f5c6c04 Committing offsets (org.apache.kafka.connect.runtime.WorkerSinkTask:187)

Thanks,

Stam

[Ewen Cheslack-Postava]

For future reference, in general the reason this is logged at INFO level is that it's not inherently a problem. Sometimes this can happen when there is no issue, e.g. the group member tries to heartbeat and is informed that the coordinator is unavailable/the broker contacted is not the coordinator. This can happen even in healthy failover scenarios, so do not indicate anything is wrong.

If this continues to happen repeatedly, then there is cause for concern (you cannot reach the real leader), but for the message reported here it is not an issue.

-Ewen

--
You received this message because you are subscribed to the Google Groups "Confluent Platform" group.
To unsubscribe from this group and stop receiving emails from it, send an email to confluent-platform+unsubscribe@googlegroups.com.
To post to this group, send email to confluent-platform@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/confluent-platform/d199c528-4311-4c60-8d4a-40036379beeb%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.