Hi guys. Wondering if/how your handle this…
As part of our proactive network monitoring, we have a bunch of Services that we monitor for, to make sure they are running on the servers for our managed clients. If they are not running, we want to be notified so we can proactively take action. We have been monitoring for more and more Services over the past 4+ years. As I have been looking into our ServiceDesk and NOC activities, I am finding that we are regularly getting a bunch of service tickets created for bogus scenarios, often for services not running immediately following server reboots, services set to have a Delayed start, etc.
I don’t personally do a lot of our Kaseya administration anymore, but was I was looking around today, it seems that Kaseya->Monitor->Monitor Sets are simply monitoring for a service not running. There is no option (that I see) to say that the service has to be not running for X minutes before the alarm is generated (or email is sent).
For a few services, we created some scripts. We monitor for the service not running, and then set the action to execute script #1. Script #1 simply schedules script#2 to run in 5 minutes. When script #2 runs, it checks to see if the service(s) is/are running, or not, and then takes action from there. Seems like a lot of extra gymnastics to go through, and its a lot of extra work when you scale across all of the various services that we might want to monitor for.
I am wondering how others so this. Do you guys monitor for many stopped services? And if so, do you run into many problems with unnecessary tickets that get generated following a reboot? Anyone have any magic for how to accomplish the end results (get notified if a service isnt running for more than X minutes, but not get a bunch of noise and bogus tickets)?
Thanks in advance!
I really like your idea for executing a script to check in 5 minutes. Sadly, it would take a hell of a lot of work to re-apply all of our monitor sets in this manner. I would love to see an easier way to get this done and perhaps submitted as a feature request to Kaseya.
yeah tis a bit of a challenge with Kaseya - we ended up building our own servicemonitoring.exe program - it runs on each server and monitors all automatic services minus a global exclusion list we created - it looks at the service if its stopped and automatic it starts it and logs informational only to event log - then if it stops again it starts but only does this so many times within a given period then it alerts via event log (error) So we have a monitor set that looks for the servicemonitoring.exe and we have an event set that looks for the error in the event log - all gets applied by template.
Has worked well for us means we don't care what server / role it is running on if it is automatic and not on our exclusion list we are monitoring it.
That was our solution
Thanks for sharing. That approach sounds very interesting. You mentioned you built the servicemonitoring.exe program. DO you mean you actually wrote the source code yourself? Or was this somethign commercially available? I'd love to check it out.
yeah we built it ourselves, I built it using auto IT but we are currently converting to a C# service application. My head was wrecked with the limitations to how kaseya managed services so I figured do it yourself.
It appears that this functionality is avaialble in KNM module. There is a discussion in another thread
Vote for this feature many it will be released in a future kaseya release: helpdesk.kaseya.com/.../219200048-Monitoring-Automatic-services-with-a-time-dealy-for-ones-that-start-as-needed-or-have-a-delay-on-start-up-