Re: aborting a repartition assignment

[Jason Rosenberg <jb...@squareup.com>]

As Lance mentioned, the best course of action in such a case (since version
0.8.X) is to keep the failed broker down, and bring up a new node (with the
same broker id as the failed broker), and it will automatically re-sync its
replicas (which may take some time).  You don't want to try to reassign
partitions, because, as you say, the one broker is failed and so that won't
succeed.

Jason

On Thu, May 28, 2015 at 7:10 PM, Foo Lim <fo...@vungle.com> wrote:

> Thx for the reply. Can't I just reassign the partition to the new broker in
> zookeeper manually? What zookeeper path should I change?
>
> TIA
>
>
> On Thursday, May 28, 2015, Lance Laursen <ll...@rubiconproject.com>
> wrote:
>
> > Hey,
> >
> > Try clearing out /admin/reassign_partitions on your zookeeper.
> >
> > Additionally, your best bet might be to bring up a new broker with the
> same
> > broker ID as your failed broker. It'll join the cluster and carry on,
> > though I'm not sure what effect having a now-empty partition is going to
> > have.
> >
> > On Thu, May 28, 2015 at 12:38 PM, Foo Lim <foo.lim@vungle.com
> > <javascript:;>> wrote:
> >
> > > Hi,
> > >
> > > I'm using kafka 0.8.2 & one of my kafka servers died (no way to recover
> > the
> > > data on the disks). There is a topic, with replication of 1, with one
> of
> > > the partitions on the dead server. I thought a reassignment would move
> > the
> > > metadata for that partition to a new server without needing the data,
> but
> > > the reassignment is stuck to in progress.
> > >
> > > I ran:
> > >
> > > $ /opt/kafka/kafka/bin/kafka-reassign-partitions.sh --zookeeper
> > > myzookeeper.my.com --reassignment-json-file new_assignment.json
> --verify
> > > Status of partition reassignment:
> > > Reassignment of partition [topicX,1] is still in progress
> > > This will never succeed since the dead server is never coming back.
> > >
> > > In the new server's log, I saw:
> > >
> > > [2015-05-28 06:25:15,401] INFO Completed load of log topicX-1 with log
> > end
> > > offset 0 (kafka.log.Log)
> > > [2015-05-28 06:25:15,402] INFO Created log for partition [topicX,1] in
> > > /mnt2/data/kafka with properties {segment.index.bytes -> 10485760,
> > > file.delete.delay.ms -> 60000, segment.bytes -> 536870912, flush.ms ->
> > > 9223372036854775807, delete.retention.ms -> 86400000,
> > index.interval.bytes
> > > -> 4096, retention.bytes -> -1, min.insync.replicas -> 1,
> cleanup.policy
> > ->
> > > delete, unclean.leader.election.enable -> true, segment.ms ->
> 604800000,
> > > max.message.bytes -> 1000012, flush.messages -> 9223372036854775807,
> > > min.cleanable.dirty.ratio -> 0.5, retention.ms -> 259200000,
> > > segment.jitter.ms -> 0}. (kafka.log.LogManager)
> > > [2015-05-28 06:25:15,403] WARN Partition [topicX,1] on broker 4151132:
> No
> > > checkpointed highwatermark is found for partition [topicX,1]
> > > (kafka.cluster.Partition)
> > > [2015-05-28 06:25:15,405] INFO [ReplicaFetcherManager on broker
> 4151132]
> > > Removed fetcher for partitions  (kafka.server.ReplicaFetcherManager)
> > > [2015-05-28 06:25:15,408] INFO [ReplicaFetcherManager on broker
> 4151132]
> > > Added fetcher for partitions List()
> (kafka.server.ReplicaFetcherManager)
> > > [2015-05-28 06:25:15,411] INFO [ReplicaFetcherManager on broker
> 4151132]
> > > Removed fetcher for partitions  (kafka.server.ReplicaFetcherManager)
> > > [2015-05-28 06:25:15,413] INFO [ReplicaFetcherManager on broker
> 4151132]
> > > Added fetcher for partitions List()
> (kafka.server.ReplicaFetcherManager)
> > > Is there a way to force it to complete or abort the reassignment
> action?
> > >
> > >
> > >
> >
> http://stackoverflow.com/questions/30500004/aborting-kafka-reassign-partition-action
> > >
> > > Thanks!
> > >
> >
>