Details
-
Bug
-
Resolution: Unresolved
-
Trivial
-
None
-
None
-
None
-
None
Description
There are disabled port down alerts except network ports with ifAlias included 'p_mon' tag ('Port down alert checker.PNG' image in attach)
There are few ports which are down for a long time and they are in this state (sfp module not inserted).
Also there is high traffic on interface alert checker ('High traffic alert checker.PNG' image in attach).
These ports alertsare suppressed by "Ignore untill ok".
So on some reason such port which is down for a long time gets alert about high traffic on interface (but port is down!), then port marked as up by this alert and then we have alert that port is down ('Alert log.PNG' in attach).
I checked counters just in case, but there are no counts (screen in attach 'Port data.PNG')
Attachments
Activity
Hi all, we have such issue aswell. I noticed it with Cisco NCS-5502-SE running 6.5.3. Sometimes the devices are not responding to ifIndex SNMP request and not returning any interfaces. Observium things that those interface simply disappeared from the devices and marks them as "DELETED". When the next poller.php cycle gets executed our NCS is returning all available interfaces and Observium is removing the "DELETED" state. Now we are receiving all alerts that are related to this interface even if "ignore_until_ok" was set before.
I'm not sure what the issue here is, but there are no high traffic alerts.
It's telling you that the check has SUCCEEDED, which is odd, as it's never telling you it failed, but constantly succeeding.
It's difficult to say what the issue is, perhaps you could run a debugged ports poll :
./poller.php -h <hostname> -m ports -dd >> debug.txt
It might be a failing SQL query or something, because it doesn't seem to know that the state is OK (you didn't post a screenshot of the actual alert instance)
Hey. Do you have any ideas?