[
https://issues.apache.org/jira/browse/NIFI-12659?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
David Handermann updated NIFI-12659:
------------------------------------
Affects Version/s: (was: 2.0.0)
> If Controller.py process is killed, nifi is not recover from it
> ---------------------------------------------------------------
>
> Key: NIFI-12659
> URL: https://issues.apache.org/jira/browse/NIFI-12659
> Project: Apache NiFi
> Issue Type: Bug
> Affects Versions: 2.0.0-M1, 2.0.0-M2
> Reporter: Zoltán Kornél Török
> Priority: Major
> Labels: python
> Attachments: Screenshot 2024-01-18 at 9.04.18.png
>
>
> When you first add a python processor to canvas / restart nifi, for the
> releated python processor a Controller.py processor is start, which handle
> communication between python / java, etc..
> If you kill this processor (simulate an error in it, or some other reason the
> processor die), nifi is not restart it and eventually processor became
> invalid:
> !Screenshot 2024-01-18 at 9.04.18.png|width=684,height=496!
> When this happens, the only solution is to restart nifi.
> This is not good for production use cases (imagine if you realize this error
> 1 week later) so we need to extend pythonBridge logic to:
> * periodically check status of the started process
> * If process is not running, try to restart it and log the problem into nifi
> log
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
