Hi,
Is it possible to create a push monitoring (a url that an appliaction can ping to show activity)?
I would like to monitor my periodic backups by pinging that url after an sucessful run. Or is there maybe a different approach to monitor something like that?
It is not exactly what you are looking for, but might also be helpful:
You can monitor the execution of a cron job by using the mk_job utility which ships together with the Checkmk agent. See
Could you explain what you mean?
For me its not clear what you want to achive but the way @Heavy describes sounds the correct one if backups scheduled by cron.
regards
Michael
Some monitoring software provides a check type that generates a URL for each service. Applications that are otherwise cumbersome to monitor can call their URL to tell the monitoring that they were successful with their job/action.
OK, thats AFIK not available out of the box but there are many other options for you. Do you have a checkmk agent installed on the host where the backup is running?
Do we talk about Windows or Linux?
Any chance of the app popping a line into a log file? Check MK logwatch can then pick this up and it will appear in the event console, which has very powerful rule-based options for alerting on logs.
My current solution is checking a file which contains the time of the last successful backup run which is fine for me. I just wondered if there is something like a push-like check type. 
No - SNMP traps are the only other way out of the box.
You could also try to “inject” the service status in the local agent’s spool directory.
I might be stating the obvious here, but as this comes up so often, and especially these days, let me just say this:
Event-based monitoring is never a good idea. Unless there really is no other humanly conceivable alternative (I am looking at you, “I can only do SNMP traps” network devices).
If you can, go state-based, which is what Checkmk does and recommends.
Regarding the backups, there were several approaches mentioned, which should all work, but possibly the easiest way for you could be a local check. Of course depending on your backup solution there might already be plugins that can monitor what you are looking for, or you could write your own plugin, but to get started, local checks should be good.
I can confirm what Robin wrote, event based monitoring is always incomplete and creates a lot of confusion. If you are able to run a checkmk agent on that host build a local check or a spooled local check as suggested by @martin.schwarz. Thats simple to do and depending on your backup software it could run as a post backup job.
regards
Michael
This topic was automatically closed 365 days after the last reply. New replies are no longer allowed. Contact an admin if you think this should be re-opened.