[
https://issues.apache.org/jira/browse/NIFI-6984?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17021196#comment-17021196
]
Shawn Weeks commented on NIFI-6984:
-----------------------------------
What's happening is NiFi continues to try and load balance to nodes that are no
longer part of the cluster. It's easy to reproduce this. Stand up a 3 node
cluster and then drop one of the nodes. Restart the other 2 and let elections
complete. You will find that load balancing still tries to contact the old
nodes even though they're aren't in the current cluster. The old nodes are
still stored in zookeeper and aren't ever reconciled.
> Flow files are struck on Load Balancing with Partition with AttributeID
> -----------------------------------------------------------------------
>
> Key: NIFI-6984
> URL: https://issues.apache.org/jira/browse/NIFI-6984
> Project: Apache NiFi
> Issue Type: Bug
> Affects Versions: 1.10.0
> Environment: Nifi 1.10 with Docker Image in Kubernetes (3 Node
> cluster)
> Reporter: venugopal
> Priority: Major
> Attachments: Capture1.PNG
>
>
> Flow files are struck on Load Balancing with Partition with AttributeID
>
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
