Identification sections in pattern failed: section: discovery, error: All command implementations (SNMP_TABLE) failed on host 10.142.11.8. Host might be down or unreachable.

[Sha_pan]

Hi everyone,

When I was trying to discover a number of IP Switches, I'm getting below error for some:

"Identification sections in pattern failed: section: discovery, error: All command implementations (SNMP_TABLE) failed on host 10.142.11.8. Host might be down or unreachable."

It was showing Classified, Identiying but whole process got aborted after that.

I tested discovery on a number of IP Switches with the same model and they're being updated successfully.

Not sure what to look here. I assume it's not the credential as the it passed that stage already and device is definitely not down.

Hope some could shed a light here.

Thanks,

Shirley

[Rahul Priyadars]

did you tried running - Pattern in Debug mode for the given IP?

I assume its reaching to pattern?

Regards

RP

[Sha_pan]

Yes, it's reaching pattern.

So the discovery log is showing it stopped after the "Get entPhysicalTable1" step.

But when I tried debugging the pattern for that one particular IP showing that same error, I was able to past through successfully until the last step "39.21 Union cmdb_ci_ip_switch_stack_backup and cmdb_ci_ip_switch"

[SiD2]

Hi @Sha_pan,

This seems to be a known issue for now and the fix might be provided soon in this PRB1378161.

But this known issue happens when some of the SNMP query is failing as mentioned in the PRB. Try debugging the pattern and see which SNMP query was failing and try to fix that by adding the proper credential. This is said based on my previous experience with same issue with some customer.

If you couldn't fix the step, try disabling tentatively till the fix is done.

There are many customers who are facing this, hope the fix will be provided soon by respective product team.

Please mark Helpful / Accept Solution so that it helps others with similar questions.

[tlsimmons88]

Hi @Sha_pan or anyone else running into this issue.  

It now looks like there is a workaround KB for this:  KB0866577

Note: you must be logged into HI to see the KB.

Hope this helps.

Travis