You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@kafka.apache.org by "Manikumar Reddy (JIRA)" <ji...@apache.org> on 2015/06/14 07:46:00 UTC

[jira] [Commented] (KAFKA-2218) reassignment tool needs to parse and validate the json

    [ https://issues.apache.org/jira/browse/KAFKA-2218?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14584945#comment-14584945 ] 

Manikumar Reddy commented on KAFKA-2218:
----------------------------------------

[~charmalloc] I am not able to reproduce the issue with below json. Can you confirm affected kafka version? 

JSON :
{"partitions":
[
{"topic": "EVENT", "partition": 0, "replicas": [a] }
],
"version":1
}


Output:
Partitions reassignment failed due to Partition reassignment data file /tmp/reassignment.json is empty
kafka.common.AdminCommandFailedException: Partition reassignment data file /tmp/reassignment.json is empty


> reassignment tool needs to parse and validate the json
> ------------------------------------------------------
>
>                 Key: KAFKA-2218
>                 URL: https://issues.apache.org/jira/browse/KAFKA-2218
>             Project: Kafka
>          Issue Type: Bug
>            Reporter: Joe Stein
>            Priority: Critical
>             Fix For: 0.8.3
>
>
> Ran into a production issue with the broker.id being set to a string instead of integer and the controller had nothing in the log and stayed stuck. Eventually we saw this in the log of the brokers where coming from 
> 	
> me	11:42 AM
> [2015-05-23 15:41:05,863] 67396362 [ZkClient-EventThread-14-****ERROR org.I0Itec.zkclient.ZkEventThread - Error handling event ZkEvent[Data of /admin/reassign_partitions changed sent to kafka.controller.PartitionsReassignedListener@78c6aab8]
> java.lang.ClassCastException: java.lang.String cannot be cast to java.lang.Integer
>  at scala.runtime.BoxesRunTime.unboxToInt(Unknown Source)
>  at kafka.controller.KafkaController$$anonfun$4.apply(KafkaController.scala:579)
> we then had to delete the znode from zookeeper (admin/reassign_partition) and then fix the json and try it again



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)