We are actually using this to monitor our Private Last Mile Synthetic machines to be sure they are "up" and we create a dashboard to easily see the status of each location (this idea came from Dynatrace). My concern is that because these machines
are not "monitored hosts" (no agents on them), we have to specify the host as the App Mon Server. I tried setting it to a collector but it fails. They only run when the AppMon server is specified as the host. I am worried about the overhead on the AppMon server. All we are doing is a ping, but we have a lot of PLM machines and this runs every 10 minutes. Anybody have any idea of the impact of something like this?
I looked at some other plugins but there were known issues with 6.5 AppMon
I'm not seeing the same behavior that you are. I just tried running a simple ping and it worked when I set it to something random (google.com). For a plugin/monitor like this what you enter in hosts shouldn't matter at all if the command isn't using it at all. The only "real" load is on the collector you set in the schedule (apart from having to handle the resulting measures).
The only reason you have to set a host at all is because of the requirements for all monitors even when it is irrelevant. If you look in the logs for when it fails do you see any information? Regardless it shouldn't have any additional impact on the server though I'm not sure why you must set it as the server, I've never encountered that.
Correction to my post. The next scheduled run would fail when I changed the host to the collector. But a subsequent run did process properly. All looks to be okay. Thanks for the input. I was concerned because coincidentally when I set a number of these up, the server experienced some performance issues. Thankfully, those too have since subsided.