cancel
Showing results for 
Search instead for 
Did you mean: 
Create Post

Monitor Processes Regardless of Status

Monitor Processes Regardless of Status

Sometimes it's important to monitor resource consumption of a process that's not running at all times. It's also nice to be able to quickly identify when the process was running and report upon how long the process ran. Regularly scheduled tasks such as generating reports are a good example of this.

Because these processes aren't running all the time they currently appear "down" in the SAM interface drawing unneeded attention from management and helpdesk staff. It also clutters up the interface by introducing unnecessary UI noise, possibly blinding you to important issues in your environment. Lastly, if alerts aren't properly configured to ignore these processes you may find yourself receiving a barrage of useless notifications.

SAM should implement some option to ignore process state (running/not running) similar to how NPM deals with unplugged status for access port interfaces. These processes should be able to be changed to a "not running" or similar state that is not "down".

Tags (1)
24 Comments
Level 16

If this had history as well it would be terrific for troubleshooting.  That way we can look back at 8am and see that a process was spawned and that is what caused XYZ to happen.

Level 15

This could be useful for other components as well.  We have some queries ran specifically for historical performance data.  Doesn't matter if they fail from time to time.  While we don't alert on these they still flag the screens as down which causes the NOC to question them. 

Level 16

I totally agree, we use the "events filtered" custom resource to filter a TON of events that go into unknown or miss one poll or what ever....

Level 13

I was coming to add this as a feature request right now.  This just happened last night in my environment.  We have been pushing SW down our Support Centers throats to start paying attention to it and action items and use it for troubleshooting.

Last night they did this finally!!!  We have a bunch of different processes across multiple applications that are low priority and like to batch out or restart at the end of the night.  This cause yellow warnings on nodes in NPM, and critical status notifications on SAM.  They jumped all over this last night and woke every on call person up.   No alerts were triggered since we don't alert on these low priority processes, but NPM/SAM react to component status.

This feature would be excellent, so we can continue to monitor metrics, but have priority listings per component.

Maybe along this logic.

Low Priority = Stat Collection, Node/App status not affected by component reporting status

Medium Priority = Stat Collection. Node/App status affected and changes logged to NPM/SAM Events

High Priority = Stat Collection. Node/App status affected and changes logged to NPM/SAM Events.  All alerts (basic/advanced) actioned upon if built.

No Priority = Functions as currently intended today.  Default setting.

Level 10

This feature would really help us in monitoring our IIS servers. Statistics for app pools is an ask from our web development team, but thanks to Windows 2008 R2 these are often down and the console is cluttered. Cleaning up the clutter using filters seems to be a lot of work and concerns me that other issues might get masked. Not cleaning it up causes a numbing effect when looking at the console. At this time we just make app monitors when requested. Not optimal.

Level 16
Level 8

I'm looking for the same feature (Monitoring a process that will spawn multible times depending on demand). Example process zxy.exe spawns and while it runs I get basic data captured like CPU/Disk/memory usage. When the process spawns more processes of the same name exampe zxy.exe#2 zxy.exe#3 etc this is also captured.

This would be great for finding a fault we are trying get data on right now. Also I know from the past that it will be a really useful feature to have, so dear SolarWinds can you give this some priority please.

Level 13

Please also add an option to monitor that a process stays down.  We have some services that get restarted & shouldn't be.  I tried to set the thresholds to > 0, but it still flagged it as being down, which was an error.

Level 15

While I agree and would like to see this done in the product.  There is a way to accomplish this now if it's worth the effort to you.  You write a script instead to detect a process.  Have it return a 0 if down and a 1 if up.  Alert greater than 0.  I've used similar scripts in the past to get around this.

Level 13

Thanks.  I haven't written a script yet.  Can someone provide the script & explain how I enter it into SolarWinds (or point me to the documentation)?.  This is a WMI component.  I know it would be in a template component, but how would I fill out each field?