Currently, we see a massive amount of mails in the queue of IPFire's mailserver. Since the situation needs some further investigation, it would be helpful if the number of mails currently in the queue could be monitored and alerts to Michael and me are eventually triggered.
What would the thresholds be and potential actions?
Treshold: 100 (warning), 250 (critical) Actions: Depends on the situation. If we just have a lot of traffic to a bogus destination (DTAG, for example), we might just have to wait. Most important aspect here is to get notified.
Apologies for being a bit too busy to google right now, but I will have to write something that allows us to extract that data via SNMP. I don't want to distribute SSH keys for running monitoring checks if I can avoid it.
(In reply to Michael Tremer from comment #3) > Apologies for being a bit too busy to google right now, but I will have to > write > something that allows us to extract that data via SNMP. I don't want to > distribute SSH keys for running monitoring checks if I can avoid it. Let me google that for you... :-) Maybe this is helpful: https://conshell.net/wiki/Using_Nagios_with_SNMP (Basically there is a Nagios plugin for monitoring the mail queue, "/usr/local/nagios/libexec/check_mailq", which can be added to SNMP). Could you set this up?
Done.