kafka-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Ryan Williams <rwilli...@gmail.com>
Subject Re: LeaderNotAvailableException
Date Tue, 12 Aug 2014 05:37:37 GMT
Thanks for the heads up on attachments, here's a gist:

https://gist.githubusercontent.com/ryanwi/84deb8774a6922ff3704/raw/75c33ad71d0d41301533cbc645fa9846736d5eb0/gistfile1.txt

This seems to mostly happen in my development environment, when running a
single broker. I don't see any broker failure in the controller log.
Anything else to look for with the topics reporting 0 replicas?




On Mon, Aug 11, 2014 at 9:31 PM, Guozhang Wang <wangguoz@gmail.com> wrote:

> Ryan,
>
> Apache mailing list does not allow attachments exceeding a certain size
> limit, so the server logs is blocked.
>
> From the controller log it seems this only broker has failed and hence no
> partitions will be available. This could be a soft failure (e.g. long GC),
> or the ZK server side issues. You may want to take a look at your
> controller log to see if there is any entries like "broker failure" before
> the offline leader selection process.
>
> Guozhang
>
>
>
>
> On Mon, Aug 11, 2014 at 5:08 PM, Ryan Williams <rwilliamz@gmail.com>
> wrote:
>
> > The broker appears to be running
> >
> > $ telnet kafka-server 9092
> > Trying...
> > Connected to kafka-server
> > Escape character is '^]'.
> >
> > I've attached today's server.log.  There was a manual restart of kafka,
> > which you'll notice, but that didn't fix the issue.
> >
> > Thanks for looking!
> >
> >
> >
> >
> >
> >
> >
> >
> > On Mon, Aug 11, 2014 at 4:30 PM, Guozhang Wang <wangguoz@gmail.com>
> wrote:
> >
> >> Hi Ryan,
> >>
> >> Could you check if all of your brokers are still live and running? Also
> >> could you check the server log in addition to the producer /
> state-change
> >> /
> >> controller logs?
> >>
> >> Guozhang
> >>
> >>
> >> On Mon, Aug 11, 2014 at 12:45 PM, Ryan Williams <rwilliamz@gmail.com>
> >> wrote:
> >>
> >> > I have a single broker test Kafka instance that was running fine on
> >> Friday
> >> > (basically out of the box configuration with 2 partitions), now I come
> >> back
> >> > on Monday and producers are unable to send messages.
> >> >
> >> > What else can i look at to debug, and prevent?
> >> >
> >> > I know how to recover by removing data directories for kafka and
> >> zookeeper
> >> > to start fresh.  But, this isn't the first time this has happened, so
> I
> >> > would like to understand it better to feel more comfortable with
> kafka.
> >> >
> >> >
> >> > ===================
> >> > Producer error (from console produce)
> >> > ===================
> >> > [2014-08-11 19:32:49,781] WARN Error while fetching metadata
> >> > [{TopicMetadata for topic mytopic ->
> >> > No partition metadata for topic mytopic due to
> >> > kafka.common.LeaderNotAvailableException}] for topic [mytopic]: class
> >> > kafka.common.LeaderNotAvailableException
> >> > (kafka.producer.BrokerPartitionInfo)
> >> > [2014-08-11 19:32:49,782] ERROR Failed to collate messages by topic,
> >> > partition due to: Failed to fetch topic metadata for topic: mytopic
> >> > (kafka.producer.async.DefaultEventHandler)
> >> >
> >> > ===============
> >> > state-change.log
> >> > ===============
> >> > [2014-08-11 19:12:45,312] TRACE Controller 0 epoch 3 started leader
> >> > election for partition [mytopic,0] (state.change.logger)
> >> > [2014-08-11 19:12:45,321] ERROR Controller 0 epoch 3 initiated state
> >> change
> >> > for partition [mytopic,0] from OfflinePartition to OnlinePartition
> >> failed
> >> > (state.change.logger)
> >> > kafka.common.NoReplicaOnlineException: No replica for partition
> >> [mytopic,0]
> >> > is alive. Live brokers are: [Set()], Assigned replicas are: [List(0)]
> >> >     at
> >> >
> >> >
> >>
> kafka.controller.OfflinePartitionLeaderSelector.selectLeader(PartitionLeaderSelector.scala:61)
> >> > [2014-08-11 19:12:45,312] TRACE Controller 0 epoch 3 started leader
> >> > election for partition [mytopic,1] (state.change.logger)
> >> > [2014-08-11 19:12:45,321] ERROR Controller 0 epoch 3 initiated state
> >> change
> >> > for partition [mytopic,1] from OfflinePartition to OnlinePartition
> >> failed
> >> > (state.change.logger)
> >> > kafka.common.NoReplicaOnlineException: No replica for partition
> >> [mytopic,1]
> >> > is alive. Live brokers are: [Set()], Assigned replicas are: [List(0)]
> >> >     at
> >> >
> >> >
> >>
> kafka.controller.OfflinePartitionLeaderSelector.selectLeader(PartitionLeaderSelector.scala:61)
> >> >
> >> > ===============
> >> > controller.log
> >> > ===============
> >> > [2014-08-11 19:12:45,308] DEBUG [OfflinePartitionLeaderSelector]: No
> >> broker
> >> > in ISR is alive for [mytopic,1]. Pick the leader from the alive
> assigned
> >> > replicas:  (kafka.controller.OfflinePartitionLeaderSelector)
> >> > [2014-08-11 19:12:45,321] DEBUG [OfflinePartitionLeaderSelector]: No
> >> broker
> >> > in ISR is alive for [mytopic,0]. Pick the leader from the alive
> assigned
> >> > replicas:  (kafka.controller.OfflinePartitionLeaderSelector)
> >> >
> >>
> >>
> >>
> >> --
> >> -- Guozhang
> >>
> >
> >
>
>
> --
> -- Guozhang
>

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message