Preferred configuration to monitor Desktops

  • Hi everybody,

    I've been using Icinga 2 to monitor my (mostly) Ubuntu servers for a while now and it really makes my everyday life easier. Now, I'd like to use Icinga 2 to monitor some Ubuntu desktop machines that don't run 24/7 to get information about disk usage and apt updates. Usually I would configure this via timeperiods, but the computers are shared by multiple users who don't come to office on a regular basis that could be defined easily. However, I think I can assume that each computer is turned on at least once a week. So my solution would be to turn the checks' timeout to 7 days, which looks like a bad practice to me.

    Is it possible to run checks and only send status notifications when the host is up, without sending notifications for the up- and downtime itself?

  • sure, you have to setup dependencies, so your checks will not fail once the host goes down.

    You should also exclude those machines from the regular notifications and create new ones for them.

    Linux is dead, long live Linux

    Remember to NEVER EVER use git repositories in a productive environment if you CAN NOT control them

  • You could use the cluster config sync to let the clients execute checks on their own. Once they are up, Icinga 2 will run the checks and send them back to the master. Disadvantage: Check results on the master will be "stale", but no action is performed (you'll see them in late check results listing though). If you add an additional cluster-zone check on the master without notifications, you can track when these clients are online. You could write an eventhandler, which auto-acknowledged the cluster-zone check once it turns into a NOT-OK state (send a request to the REST API action).