NiFI Process Group Got Corrupted If a node on the server gets disconnected & connects back automatically

[#1]

Hi All,

I'm having a 6 node on prem nifi cluster which has certain process groups. It had a node disconnect once and got auto connected back. Post this I tried to stop/disable a process group which was in started state in the UI, but it threw me an error stating -

Node xxx is unable to fulfill this request due to : [xxxx, xxxxxxxxxxx, xxxxxxxxxxxxx] is not the most up-to-date revision. This component appears to have been modified.

Although I didn't modified any component in the past, only a node disconnect and auto connect was the event which happened in between. 

As per my previous experiences, I had to restart whole nifi from Cloudera Manager UI and then restart the flows which would resolve the problem which I stated above. But I was looking for an alternate solution other than restarting whole NiFi from CM UI as this is my production cluster and I can't get downtime so easily.

Any help would be highly appreciated.

--Adding the snapshot of the error below which i got while stopping/disabling the process group from NiFi UI post a node disconnect and auto connect happened.

[#2]

@mechtech007, Welcome to our community! To help you get the best possible answer, I have tagged our NiFI experts @cotopaul @MattWho @SAMSAL @ckumar  who may be able to assist you further.

Please feel free to provide any additional information or details about your query, and we hope that you will find a satisfactory solution to your question.

[#3]

Hey Vidya, thanks for reverting back. Have attached every possible details to my query.

@cotopaul @MattWho @SAMSAL @ckumar , anyone of you would love to assist me here?

[#4]

@mechtech007 

Providing details such as the exact version of CFM NiFi you are using is very important in getting help as NiFi evolves over time, those who help want to know as much as possible in order to avoid providing irrelevant feedback that may not pertain to the version you are using.

You could rename the flow.xml.gz and flow.json.gz (present only if new enough version of CFM) and restart the node.  The node will inherit the flow from the cluster nodes when it tries to join the cluster.

Depending on how old your CFM version is, you could be hitting one of these bugs:
https://issues.apache.org/jira/browse/NIFI-7920
https://issues.apache.org/jira/browse/NIFI-8204

If you found any of the suggestions/solutions provided helped you with your issue, please take a moment to login and click "Accept as Solution" on one or more of them that helped.

Thank you,
Matt