I have a WF with Python node to fetch Azure AD token, It was working fine and there were no issues with Python node and code inside it.
Now I am facing some strange issue with Python node, Usually Python node will execute in less than 30 sec, but now it is executing for hours and it is not coming to an end.
Not sure what is the issue with the server/ executor.
If I run the WF in local, It is working fine no issues
If I run the WF on server (dev1), The node is executing for hours.
If I run the same WF on server (dev2), The node gets executed within seconds. It is working fine.
dev1 server is acting strange.
I had tried to kill the Python using PID and then ran the WF again but no use. The Python node is executing and it is not showing any error in logs to resolve the issue.knime-log.txt (15.0 KB)
dev1 and dev 2 server are using
a. Server - 4.11.4
b. KNIME Analytical Platform - 4.2.4
It is a blocker for me, suggestions are really appreciated
I am afraid, I do not really see the connection between your issue and the post you referenced. Do you observe an excessive number of open Python processes or high resource usage on the server? Or was that just about terminating the node execution?
Your knime.log does not show anything suspicious, I agree. That is why it is currently hard to tell whether this is a problem with KNIME or some external factor (you mentioned Azure AD: so maybe network problems, changed configurations, or the like?)
I think the simplest way to further diagnose this problem would be to execute your Python script outside of KNIME using the Python interpreter directly, and see whether the long runtime persists. Could you try doing that? Please let me know if you need any assistance with this step.
