ghub2015
commented
9 years ago I agree.
"Passive checks" or "distributed monitoring" (that has a light footprint on the "remote" monitors, i.e., does NOT require MySQL) would put this app over the top, IMHO.
Also, some way to coordinate checking with other "master" instances (so that you can replicate/distribute the dashboard/UI itself) would just make this awesome.
But, then you would really have to re-examine the cron based approach to checking; it would be better if the various checks could happen at independent intervals (i.e., I could choose to check something every 2 minutes; something else every 30 seconds; something else once a day).
I know this is quite a bit of work, but these are some of the features that would make this comparable with some industrial strength monitoring apps.
It would be great to have an option to declare "passive checks" which would not be directly run by cron on the monitoring server. This could be used by an agent utility running on remote monitored machine which sends a periodical update of the status value (eg. JMX stats). Cron job only should check if the value arrived (while no news are also bad news = offline). The value itself (if it arives) would be checked for limits upon update.
This is necessary especially if there are security limits for accessing the server directly.