Home | About | Sematext search-lucene.com search-hadoop.com
NEW: Monitor These Apps!
elasticsearch, apache solr, apache hbase, hadoop, redis, casssandra, amazon cloudwatch, mysql, memcached, apache kafka, apache zookeeper, apache storm, ubuntu, centOS, red hat, debian, puppet labs, java, senseiDB
 Search Hadoop and all its subprojects:

Switch to Threaded View
Kafka >> mail # user >> problems reasignating partitions


Copy link to this message
-
Re: problems reasignating partitions
What you saw is not normal. In sync replicas should always be a subset of
assigned replicas. Did that anomaly happen before or after reassigning
partitions? Is that reproducible?

Thanks,

Jun

On Tue, Sep 3, 2013 at 12:16 AM, Pablo Nebrera <
[EMAIL PROTECTED]> wrote:

> Hello
>
> I have a cluster with:
> * three nodes: rb01, rb02 and rb03
> * three topics: rb_event, rb_monitor, rb_flow
> * three partitions per topic
> * replicas 2
>
>
> The current status is:
>
> Topic      Partition  Leader  Replicas             In-Sync replicas
>
> -------------------------------------------------------------------------------
> rb_event   0          2       2,1                  2,0
>            1          2       1,2                  2,0
>            2          2       1,2                  2
>
> -------------------------------------------------------------------------------
> rb_flow    0          2       1,2                  2,0
>            1          2       2,0                  2,0
>            2          0       0,1                  0,1
>
> -------------------------------------------------------------------------------
> rb_monitor 0          2       1,2                  2
>            1          2       2,1                  2,0
>            2          0       0,2                  2,0
>
> -------------------------------------------------------------------------------
>
>
> I see something strange. For example:
>
> * rb_event -> partition 0 has replicas 2,1 but In sync replicas are 2,0
>  -> that is normal ???
>
> And the worst it when I try to reassign propertly the partitions:
>
> The kafka-reassign-partitions.sh  tries to write
> to /admin/reassign_partitions the following:
>
>
> {"partitions":[{"topic":"rb_event","partition":0,"replicas":[0,1]},{"topic":"rb_event","partition":1,"replicas":[1,2]},{"topic":"rb_event","partition":2,"replicas":[2,0]},{"topic":"rb_monitor","partition":0,"replicas":[2,0]},{"topic":"rb_monitor","partition":1,"replicas":[0,1]},{"topic":"rb_monitor","partition":2,"replicas":[1,2]}]}
>
>
> When I access to zookeeper I do see:
> [zk: rb02:2181,rb01:2181,rb03:2181(CONNECTED) 3] get
> /admin/reassign_partitions
> { "partitions":[ { "partition":2, "replicas":[ 1, 2 ], "topic":"rb_monitor"
> }, { "partition":0, "replicas":[ 2, 0 ], "topic":"rb_monitor" }, {
> "partition":1, "replicas":[ 0, 1 ], "topic":"rb_monitor" }, {
> "partition":0, "replicas":[ 0, 1 ], "topic":"rb_event" }, { "partition":2,
> "replicas":[ 2, 0 ], "topic":"rb_event" } ], "version":1 }
>
>
> And keep there for ever
>
> What can I do ?
>
> Thanks
>
>
>
> Pablo
>

 
NEW: Monitor These Apps!
elasticsearch, apache solr, apache hbase, hadoop, redis, casssandra, amazon cloudwatch, mysql, memcached, apache kafka, apache zookeeper, apache storm, ubuntu, centOS, red hat, debian, puppet labs, java, senseiDB