You are viewing a plain text version of this content. The canonical link for it is here.
Posted to jira@kafka.apache.org by "Yogesh BG (Jira)" <ji...@apache.org> on 2020/08/11 02:13:00 UTC
[jira] [Commented] (KAFKA-10381) Add broker to a cluster not
rebalancing partitions
[ https://issues.apache.org/jira/browse/KAFKA-10381?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17175147#comment-17175147 ]
Yogesh BG commented on KAFKA-10381:
-----------------------------------
One more observation is that when i restart the leader node for that partition, it picks up and issue is reolved. But we can not do restart in real scenario - will be having data loss during restarts
> Add broker to a cluster not rebalancing partitions
> --------------------------------------------------
>
> Key: KAFKA-10381
> URL: https://issues.apache.org/jira/browse/KAFKA-10381
> Project: Kafka
> Issue Type: Bug
> Affects Versions: 2.3.0
> Reporter: Yogesh BG
> Priority: Major
>
> Hi
> I have 3 node cluster, topic with one partition. when a node is deleted and add another node. Topic goes on unknown state and not able to write/read anything, below exception is seen
>
> {code:java}
> [2020-08-10 00:00:00,108] WARN [ReplicaManager broker=1004] Leader 1004 failed to record follower 1005's position 0 since the replica is not recognized to be one of the assigned replicas 1003,1004 for partition A-0. Empty records will be returned for this partition. (kafka.server.ReplicaManager)
> [2020-08-10 00:00:00,108] WARN [ReplicaManager broker=1004] Leader 1004 failed to record follower 1005's position 0 since the replica is not recognized to be one of the assigned replicas 1003,1004 for partition C-0. Empty records will be returned for this partition. (kafka.server.ReplicaManager)
> [2020-08-10 00:00:00,108] WARN [ReplicaManager broker=1004] Leader 1004 failed to record follower 1005's position 0 since the replica is not recognized to be one of the assigned replicas 1002,1004 for partition A-0. Empty records will be returned for this partition. (kafka.server.ReplicaManager)
> [2020-08-10 00:00:00,108] WARN [ReplicaManager broker=1004] Leader 1004 failed to record follower 1005's position 0 since the replica is not recognized to be one of the assigned replicas 1002,1004 for partition B-0. Empty records will be returned for this partition. (kafka.server.ReplicaManager)
> [2020-08-10 00:00:00,108] WARN [ReplicaManager broker=1004] Leader 1004 failed to record follower 1005's position 0 since the replica is not recognized to be one of the assigned replicas 1003,1004 for partition A-0. Empty records will be returned for this partition. (kafka.server.ReplicaManager)
> [2020-08-10 00:00:00,108] WARN [ReplicaManager broker=1004] Leader 1004 failed to record follower 1005's position 0 since the replica is not recognized to be one of the assigned replicas 1003,1004 for partition A-0. Empty records will be returned for this partition. (kafka.server.ReplicaManager)
> {code}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)