I see what the issue is – trying to resolve – looking at the consumer offset’s,
the ones
that are logging this message are all not in sync for this broker. I read that
setting
this parameter - replica.lag.time.max.ms , to a time before the failure may
allow for
the consumer offset to get back in sync. I’m letting this run.
Jim
////////////
On 2/25/22, 7:09 PM, "Jim Langston" <jlangs...@resolutebi.com> wrote:
Hi all , after upgrading to 2.8.1 and restarting the brokers, one of the
brokers is continually logging messages similar to this
Feb 25 23:58:17 bigdata-worker2.dc.res0.local WARN: (Logging.scala:70)
[ReplicaFetcher replicaId=2, leaderId=0, fetcherId=0] Reset fetch offset for
partition __consumer_offsets-6 from 45647 to current leader's start offset 0
Feb 25 23:58:17 bigdata-worker2.dc.res0.local INFO: (Logging.scala:66)
[ReplicaFetcher replicaId=2, leaderId=0, fetcherId=0] Current offset 45647 for
partition __consumer_offsets-6 is out of range, which typically implies a
leader change. Reset fetch offset to 45647
I have let this run without interruption hoping it would clear on its own (
I have been rotating /var/log/syslog with fills up), but
the messages just keep logging.
Is there something I have missed during the upgrade ? Is there something
that can be done to stop the messages ( I have
found several blogs/user group references to the message but none have
successfully stop the message). I have started
and stop the brokers several times.
Thanks,
Jim
