If you have false positives you can customize the negative cache time to be less than the retry interval.
The function OID has the following description.
returns value of given snmp OID. Only first parameter mandatory. First parameter - oid string, second - cache time - default 5 seconds (5.0), third - negative cache time - default 5 minutes (300.0), forth - ip address (overrides context device), fifth - snmp profile (overrides context device)
oid_column has the same parameters.
In one probe that I usually get false positives on I set the negative cache time to 29 seconds and the cache time to 10 seconds… I had negative cache time set to 60 but that was still giving a few false positives. This seems to have cleared up failed snmp reads of my cisco cpu probe.
array_size(oid_column(“1.3.6.1.4.1.9.2.1.57”, 10, 29))
The problem is I will need to modify all my oid reads to fix every probe. I use the default polling of 30 seconds with 3 retries. A single failed read sets the negative cache timer (300 seconds) and the next two successful retries don’t clear the negative cache time. Negative cache time should not be set unless all retries fail. That way ONLY if you have all retries fail the device will stay down for 5 minutes. Or allow the negative cache time to be set by configuration…
Any thoughts?
Thanks,
SD