[ 
https://issues.apache.org/jira/browse/NIFI-1963?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15995586#comment-15995586
 ] 

Jeff Storck commented on NIFI-1963:
-----------------------------------

I attempted to recreate this issue on a 3-node cluster running on my mac.  If a 
node (let's say node 1) is disconnected from the cluster, and then the penalty 
duration is edited for a processor (for example, changing the default 30 
seconds to 90 seconds), node 1 will be able to reconnect to the cluster with 
the penalty duration showing the correct value from the cluster (30 seconds). 
If an attempt is made to change the penalty duration on node 1 while it's 
connected to the cluster, the following error occurs:
{quote} Node localhost:8441 is unable to fulfill this request due to: [2, 
cf4afa51-015b-1000-fcb9-09b5aace642f, bb6ce87e-015b-1000-b1b1-ff86492b6428] is 
not the most up-to-date revision. This component appears to have been 
modified{quote}
If, at that point, node 1 is disconnected again, the previous value entered (90 
seconds) for the penalty duration will be shown.
After reconnecting node 1 a second time, the penalty duration can be changed on 
node 2 or 3 (nodes that haven't been disconnected), and the change is reflected 
properly on node 1.  If node 1 is disconnected again at this point, the error 
mentioned earlier still occurs if an attempt is made to change the penalty 
duration on node 1.

> Cluster - Flow inheritability
> -----------------------------
>
>                 Key: NIFI-1963
>                 URL: https://issues.apache.org/jira/browse/NIFI-1963
>             Project: Apache NiFi
>          Issue Type: Bug
>          Components: Core Framework
>            Reporter: Matt Gilman
>            Assignee: Jeff Storck
>
> I disconnected a node and changed the penalty duration. The attempted to 
> reconnect the node. The node joined successfully with an inconsistent penalty 
> duration.
> Need to reconsider which fields need to belong in the flow fingerprint to 
> ensure consistent configuration throughout the cluster.



--
This message was sent by Atlassian JIRA
(v6.3.15#6346)

Reply via email to