Monitoring a server process (.exe)

Hi,

I would like to monitor whether two particular process on all of our windows servers are down.

I am new to this, but starting to get more familiar with creating reports using report writer and also adding some filtering to the built in monitors.

I just can't quite seem to find the right area to on either report writer or a monitor on the web console to filter these results.

Hope this makes sense.

Thanks

Mike

Find more posts tagged with

Accepted answers

All comments

NKKswitches

Maybe someone has better instructions or better yet some screenshots. But real quick, I just recently did this. And maybe there is a better way.

Here is the way I did it. I stumbled across it while doing something else.

Go to the Node Details page that has the service you want to monitor. If you don't already have a "Top 10 Processes by.." list on there, add one. This will give you a button for the "Real-Time Process Explorer". If there is a faster way to get to this for the selected Node, I would love to know how.

Click on "Real-Time Process Explorer", A new window appears. Click "Show all" at the bottom if you don't see the process you want.

In this list, you will see a column called "Assigned Application". If there is already a template that covers the process, you will see the name of that Template. Otherwise you will see "Start Monitoring". Click "Start Monitoring".

That will open another new window for a custom Application Monitoring template. Do the usual name, description, creds, etc.

I found this post while looking for the next logical step to this. Alerts for this newly made monitor. I now have a way to view historical CPU usage, the status up/down/warning, and more using this custom monitor method. But now I want to get an email when the status is not UP. Which would be easy if I could figure out how to make any alerts for my new monitor template.

In case I am close, I will tell you what I did for the Alerting portion of this process.

I created a new alert and for the Trigger Conditions I chose:

ALL:

System Name is equal to <Node Name of my server>

Node Status is not equal to Down (if it was down, another alert would occur and I couldn't check the processes anyway)

Application Name is equal to Nrouter.exe (the exe I created the template for)\

Application Status is not equal to Up

Now I think this is correct and works. I need to try with a test server as I can't take this service down to really test. But my issue is with the Message Variables. I simply want to have the message give me the process name, status, and cpu usage. LastTimeUp would be nice as well. But mostly I need the process name given and the status. Sure, I could simply type out the process name in this case as it's 1 process and I know it's Nrouter.exe here. But I would like to know what variable has that info. But I must have the Status variable. Sure I could say "Nrouter.exe is not Up" since that is what I am testing. But I would like to see if it's down or warning. And then of course the CPU usage.

Anyway. That should be enough info to get you in the right direction. I will have to really test the alert to see if I have the trigger setup correctly.

UPDATE: Of course this is probably more than you wanted to do. I was thinking you could then create a custom monitor Template, 1 for each of the 2 processes you want to monitor. And then use those 2 templates to alert against each of your server Nodes.

aLTeReGo

If I'm understanding you correctly, you want to be alerted when two processes are not running on all of your servers at the same time? Or are there two important processes running on each of your servers that you want to ensure are always running, and you'd like to be alerted if both of these services are stopped on any server?

aLTeReGo

Maybe someone has better instructions or better yet some screenshots. But real quick, I just recently did this. And maybe there is a better way.
Here is the way I did it. I stumbled across it while doing something else.
Go to the Node Details page that has the service you want to monitor. If you don't already have a "Top 10 Processes by.." list on there, add one. This will give you a button for the "Real-Time Process Explorer". If there is a faster way to get to this for the selected Node, I would love to know how.
Click on "Real-Time Process Explorer", A new window appears. Click "Show all" at the bottom if you don't see the process you want.
In this list, you will see a column called "Assigned Application". If there is already a template that covers the process, you will see the name of that Template. Otherwise you will see "Start Monitoring". Click "Start Monitoring".
That will open another new window for a custom Application Monitoring template. Do the usual name, description, creds, etc.

A simpler method might be to use the Find Processes, Services, and Performance Counters Wizard under [Settings -> SAM Settings]

I found this post while looking for the next logical step to this. Alerts for this newly made monitor. I now have a way to view historical CPU usage, the status up/down/warning, and more using this custom monitor method. But now I want to get an email when the status is not UP. Which would be easy if I could figure out how to make any alerts for my new monitor template.
In case I am close, I will tell you what I did for the Alerting portion of this process.
I created a new alert and for the Trigger Conditions I chose:
ALL:
System Name is equal to <Node Name of my server>
Node Status is not equal to Down (if it was down, another alert would occur and I couldn't check the processes anyway)
Application Name is equal to Nrouter.exe (the exe I created the template for)\
Application Status is not equal to Up
Now I think this is correct and works. I need to try with a test server as I can't take this service down to really test. But my issue is with the Message Variables. I simply want to have the message give me the process name, status, and cpu usage. LastTimeUp would be nice as well. But mostly I need the process name given and the status. Sure, I could simply type out the process name in this case as it's 1 process and I know it's Nrouter.exe here. But I would like to know what variable has that info. But I must have the Status variable. Sure I could say "Nrouter.exe is not Up" since that is what I am testing. But I would like to see if it's down or warning. And then of course the CPU usage.
Anyway. That should be enough info to get you in the right direction. I will have to really test the alert to see if I have the trigger setup correctly.
UPDATE: Of course this is probably more than you wanted to do. I was thinking you could then create a custom monitor Template, 1 for each of the 2 processes you want to monitor. And then use those 2 templates to alert against each of your server Nodes.

The out-of-the-box alert named "Alert me when an application goes down" should work for your needs by default. If there is some specific information you'd like included as part of this alert, a full listing of Alert variables is available in the SAM Administrators guide.

michaelbensusan

alerego, you are spot on!

two particular processes (they are not windows processes), antivirus and also backup agent are the processes i want to see if they are down\stopped.

I cannot seem to find any references for these processes when i have looking around the template creator.

Thanks

aLTeReGo

two particular processes (they are not windows processes), antivirus and also backup agent are the processes i want to see if they are down\stopped.
I cannot seem to find any references for these processes when i have looking around the template creator.

First you'll need to ensure these processes are monitored. The easiest way to do this I outlined above, using the Find Processes, Services, and Performance Counters wizard to create an application template. Then you can apply this template to any Orion managed node that's running these processes.

As for alerting, the out-of-the-box alert "Alert me when an application goes down" will trigger when either of these processes aren't running. If you need to configure an alert that only triggers when both of these processes are stopped, below is an example of what that might look like as a Trigger Condition in the Advanced Alert Manager.

lionkin

hello, when I open 'alert me when an application goes down' it doesnt give me the 'Component Status' to choose I can do 'Node Status is equal to ...'

also I cant add 'Process name', cant find it on the list, am i missing anything here?