Is it possible to get an alert on a service that will not restart.? We use monitoring to restart services if they stop and then send an alert. However, we do not receive an alert if the service failed to start.
If your monitoring for a service, it will alert you that it is not running. So if the service does not start up, Kaseya will see the service as not running and the monitor alert will trigger. You can also use Event Log monitoring to watch for this. I hope this helps, let me know if you would like more details though.
Jon, thanks for the reply. I must be missing something obvious. Our monitoring sets will attempt to restart a failed service and is usually successful. We always receive that alert. However, we do not receive an alert if the service will not start. For example, we had an issue where someone modified the SQL Server service authentication. The monitor set attempted to restart (3) times but it would never start because of the change. We were not alerted that it was not running. Thanks again.
Hmm, I have never had an issue with the alert not being triggered in cases similar to what you outlined unless there was a suspend alarm set at the time and the rearm period of the monitor was set to several hours.
To counter this, Event Log Monitoring is the best method. You can watch for specific events (such as service failed to start) or you can monitor all events and build up an extensive ignore list for what you do not want to see (my preferred method).
We're in the same boat as you. Ideally we'd like the alarm to trigger only when the service fails to restart. We don't want alerts when the service goes down as long as it restarts fine, that just generates noise that we'd prefer to avoid. Did you end up using Event Log Monitoring for this or did you come up with another solution?
bascheew - we don't have an answer yet. I have a conference call scheduled with K next week and this is one of our first questions. I don't know if we are missing anything obvious. I don't want to create multiple event log alerts - there are just too many to pick from.
For the monitor set, is it set to create an alarm, create a ticket, or send an email? Is this happening on multiple service monitors or only noticed on the one case mentioned with the logon change?
Send an email. It happens on any service that fails to (re) start. We are notified of the restart attempt but not the failure.
I have this working for me. Here's my solution....
under the monitor sets build a monitor set that includes the service you're trying to monitor. use the "Service Check" button and add the service from the list. Set your attempts to what you'd like (I use 3) and your restart interval to greater than 0, I use 1 min. Then your re-arm should be less then the total restart interval.
So if you restart attempts every one min and tries 3 times, that's 3 min's. Set your re-arm to say 5 min. You'll get an alert only when the service failed to start 3 times.
Let me know if you have any trouble. I can screen shot my monitor set if you need to view it.
danrche - that looks interesting. I will have to give that a try. Thanks.
Does that not mean it will alarm when it first occurs, then again after 5 minutes if it fails? Rather than only alarming after it has failed to restart?
I just tried what danrche suggested. Stu_u2k you are correct, this results in getting an email when it first sees the service is down, followed by emails every 5 minutes to remind you that it's still down!
I don't think Kaseya's email alert will do what we want so I am going to start working on a script to be triggered when a service goes down or use a combination of a service monitor set with event log monitoring.
so you're getting an email on the 1st check that it's down? that strange. I only get it when it can't restart it after the 3rd time. But if you're looking for something to verify the service, #ln# is the variable used for the monitor, so you could do something like pass that to a script that checks for that service status. up/down result and have that email you.
this tread may be helpful to you, kcears has a nice method. I've included a vbscript on the thread as well that can verify and email the service state, but if you're looking to check multpile services, try Kcears 1st, I think his method is better then mine.
There is another way to do it which I will post on 13442 too...
Apply a monitor set for a group of services you want to be alerted on the 3rd occasions for, tell it to write a Event Log entry on occurrences which says something like "Monitor Set Standard 2003 Services Failed" and apply another set to watch the event logs, this set then emails you the alert only if 3x occurrences of this EL entry appears in a predefined period, a little messy and it means you have to apply extra sets which is why I think KCEARS possibly has the better solution here but my method will work for any machine, Linux, Mac, Windows etc. where as KCEARS requires VBS support.
I found this which looks similar to stu_u2k's solution (I think) - macnamaratech.blogspot.com/.../monitoring-services-using-kaseya.html