Hi
We run kafka 0.10.0.1 on the server
And recently upgraded a client that used the kafka 0.8.2 java client to use
kafka client 0.9.0.1
We previously had fetch.message.max.bytes set to 11534336
This ended up crashing in production with
org.apache.kafka.common.errors.RecordTooLargeException: There are some
messages at [Partition=Offset]: {conversations-3=80824497} whose size
is larger than the fetch size 1048576 and hence cannot be ever
returned. Increase the fetch size, or decrease the maximum message
size the broker will allow.
and we realized this property had been changed to max.partition.fetch.bytes
however as soon as i set this property. The log would be full of
Attempt to heart beat failed since the group is rebalancing, try to re-join
group.
errors and no one would ever end end up owning a partition. The where all
unassigned.
Can anyone offer any suggestions on how to solve this ?
--jelmer
