Thanks to guozhang.
According to your suggestions，I found my new patch to kafka 0.9.0.0 may casue
the problem,
In delayedfetch.scala, I include import
org.apache.kafka.common.errors.NotLeaderForPartitionException but not import
kafka.common.NotLeaderForPartitionException for intelij auto import
Json,
As you mentioned yourself the "NotLeaderForPartitionException" thrown
from getLeaderReplicaIfLocal
should be caught in the end, and hence I'm not sure why the reported stack
trace "ERROR: ..." throwing the NotLeaderForPartitionException should be
seen from "tryComplete". Also I have checked
it make the cluster can not provide normal service，which leades some producer
or fetch fail for a long time before I restart current broker.
this error may be come from some formerly fetch operation which contain this
partition，which leads many fetch response error.
The delayFetch's tryComplet
Hi all,
I use Kafka 0.9.0.0, and we have a cluster with 6 nodes, when I restart
a broker，we find there are many logs as below，
[2016-10-24 15:29:00,914] ERROR [KafkaApi-2141642] error when handling request
Name: FetchRequest; Version: 1; CorrelationId: 4928; ClientId:
ReplicaFetcherThre
