04-06-2021 02:22 AM
I have a python service that is called through RESTCONF, and ran into an issue where the service was failing to create certain instances, but did not log any details of the failure in the service python vm log. When I re-created the issue through the NCS CLI instead of RESTCONF, I got a CLI error message showing that the device was rejecting the configuration when NSO attempted to apply it rather than there being a problem with the service itself. Is there a way to add logging for this kind of commit error within the python service code?
04-16-2021 09:43 AM
Hello,
I'm not an NSO expert but please allow me to shine some light on some similar discussion, which may be of relevance in your case:
Turn off console logging in NSO
NSO Dev Days: Tips and Tricks for Troubleshooting with NSO (without screen capture)
NSO Developer Days Troubleshooting presentation
NSO Configuration Errors and Rollbacks Due to Device Prompt Following Write Command
I'm fairly certain none of these are a perfect fit, but they may give you some ideas and/or you can see who in the community is responding and solving these issues.
Finally, I would recommend checking for device / software bugs using the Cisco Bug Search Tool.
Hope this helps!
04-19-2021 06:51 AM
the problem is that the error happened after the create() code was executed, that is why in neither cases you get something on the python-vm log. The same message you see in the CLI, you should have received in the RESTCONF JSON/XML body together with a en error HTTP code.
Was that not the case?
From a log perspective, you will see the ERROR message in the develop.log file.
Discover and save your favorite ideas. Come back to expert answers, step-by-step guides, recent topics, and more.
New here? Get started with these tips. How to use Community New member guide