Intermittent SNMP "Item not found in monitoring data" errors

Evaluating checkmk 2.0.0p1 CFE running on Ubuntu 20.04 and trying to monitor Cisco Catalyst 9300 and 9500 switches running IOS XE 16.12.03a using SNMP v2. I’m seeing “item not found…” events at random intervals - always on interfaces where it’s pulling SFP DOM data (temp, rx/tx power). They just come and go for no apparent reason.

Loving what I see so far with checkmk and I’m looking to replace both my Solarwinds instances, but if I can’t reliably monitor switch interface data it’s just not going to work for me.

BTW - I sent an email to your marketing address two days ago asking what support resources are available during an evaluation and never heard back, so here I am. Might want to look into that as first impressions do matter.


1 Like

Hi texx,

I just checked with our C9300s and it looks like I’m seeing that too. It’s pretty rare (about once per switch in 3 weeks) and always only on the 25Gig in 10Gig interface.

How often are you seeing this?

Random, but at least 6-8 times per day for each of the two switches I’m currently testing with.

Same message but on a local check doing some magic on a local api. We set check interval to 5 minutes for machines giving issues and it went away. Could it be the check takes too long / times out and thus the data is “not found in the monitoring data”

I guess that’s not the reason for me. The check-duration seems pretty solid at the time the problem occurs:

Did you eventually get a response on this?
Or maybe found a way to filter out these messages?

From what I see the device does not return the data. All other plugins (OIDs) work find, just this one is missing.

I’d also go for SNMP troubleshooting, i.e. timeout of the request, SNMP timing settings.
We have a bunch of Cisco devices (switches, APs…,>11000 in sum) and a bunch of them are Catalyst, so we had to deal with SNMP performance alot. Also what i would check is: how often is Check_MK Discovery set to run?

BR Thomas