I have an alert setup for any interface errors greater than 25 to alert our team. We just got one, and I cleared the counters on that switch, but libre wont recover from it.
Ive searched the forum and only found 2 other threads with this similar issue. Something about how there is no snmp reset on cisco, and you have to delete it possibly from librenms database? I have no clue, and I have no idea how to even do that.
Thoughts?
What does your alert rule look like?
Are the error counters on the switch increasing since you cleared the counters?
If you turn the rule off and on again, does it clear?
no errors are not increasing, especially now since the user left and the PC is not being used. So libre should clear this alert you would think. Weird.
Yes, I have tried turning it off and on again
I believe ifInErrors and ifOutErrors just read the raw SNMP values, which in most switches isn’t the same as the CLI counters and wouldn’t clear without a reboot or SNMP service restart. So if there were >25 out error OR >10 in errors on the interface basically from the time the switch booted, the last 2 rules will alert.
I would try ifInErrors_delta which should be # of errors between the last two polls (ifInErrors - ifInErrors_prev)
…or…
ifInErrors_rate which should be errors per sec between the last two polls ( (ifInErrors - ifInErrors_prev) / polling interval )
