Hmm, that seems to indicate that the leader info is not propagated properly from the controller to broker 0. In the state-change log of broker 0, do you see anything related to these 2 partitions when broker 0 comes up?
Thanks, Jun On Tue, Sep 9, 2014 at 9:41 AM, Steven Wu <stevenz...@gmail.com> wrote: > noticed one important thing. topic foo's partition 1 and 2 have empty .log > file on replicas. I suspect replication doesn't create the partition dir on > broker 0 in this case, which then cause the WARN logs. > > > On Mon, Sep 8, 2014 at 11:11 PM, Steven Wu <stevenz...@gmail.com> wrote: > > > sorry. forgot to mention that I am running 0.8.1.1 > > > > > > On Mon, Sep 8, 2014 at 9:26 PM, Steven Wu <stevenz...@gmail.com> wrote: > > > >> did a push in cloud. after new instance for broker 0 comes up, I see a > >> lot of WARNs in log file. > >> > >> 2014-09-09 04:21:09,271 WARN kafka.utils.Logging$class:83 > >> [request-expiration-task] [warn] [KafkaApi-0] Fetch request with > >> correlation id 51893 from client 1409779957450-6014fc32-0-0 on partition > >> [foo,1] failed due to Partition [foo,1] doesn't exist on 0 > >> 2014-09-09 04:21:09,383 WARN kafka.utils.Logging$class:83 > >> [kafka-request-handler-5] [warn] [KafkaApi-0] Fetch request with > >> correlation id 51894 from client 1409779957450-6014fc32-0-0 on partition > >> [foo,2] failed due to Partition [foo,2] doesn't exist on 0 > >> > >> zookeeper shows it is the leader after this new instance come back. > >> > >> > {"controller_epoch":5,"leader":0,"version":1,"leader_epoch":0,"isr":[0,5]} > >> > {"controller_epoch":5,"leader":0,"version":1,"leader_epoch":1,"isr":[0,1]} > >> > >> but we I check the data dirs. those two partitions aren't there. > >> > >> any idea? > >> > >> Thanks, > >> Steven > >> > >> > > >