[
https://issues.apache.org/jira/browse/NIFI-3923?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16014187#comment-16014187
]
ASF GitHub Bot commented on NIFI-3923:
--------------------------------------
Github user mcgilman commented on the issue:
https://github.com/apache/nifi/pull/1817
Thanks @markap14. This has been merged to master.
> When clustered, nodes can start processors before repositories have finished
> initializing
> -----------------------------------------------------------------------------------------
>
> Key: NIFI-3923
> URL: https://issues.apache.org/jira/browse/NIFI-3923
> Project: Apache NiFi
> Issue Type: Bug
> Components: Core Framework
> Reporter: Mark Payne
> Assignee: Mark Payne
>
> When a node starts up, if configured to be part of a cluster, the
> StandardFlowSynchronizer starts/stops processors by calling
> ProcessGroup.startProcessor and ProcessGroup.stopProcessor instead of going
> through the FlowController. This can result in a Processor being started
> before the Provenance/FlowFile/Content Repositories are fully initialized. We
> should only be starting or stopping Processors through the FlowController.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
