mikegale · Systems Administrator · ✭✭✭✭✭

We're only just now looking at Office365 API monitoring so I'll have more feedback soon.

Thank you tomasv and aLTeReGo​ for the speedy responses. Y'alls help is always greatly appreciated. I spoke to the security chaps again and the script is not a problem now that they know it's legit. What had set of their alerts was a process that created a vbs script and then ran it from a 'temp' location. Is that correct,…

Sorry it took this long. Here's a solution that could work for you: Find a string in a logfile using a counter file This script uses a counter to determine if there are more matches on the current poll than there were on the last poll. If there's more errors on the current poll it will return a DOWN status. I'm still…

Did this KB get created? I'm having the sane problem and was curious about reading a little more about it.

Hey aLTeReGo, I know this is an older post but I have a related question. We updated SAM in our test environment yesterday from 6.24 to 6.5 and this morning we received a call from our security dept saying they had to block "GetPendingUpdates.vbs" over night because it appeared suspicious to them. Has anything changed in…

That makes sense. Thanks for your help.

Slightly old post but still relevant. Did you ever find the correct fix for this problem? I run in to this occasionally and when one gets the "Get Output Failed" message one doesn't get a Statistical value or the Message returned. I understand that it goes down and one could alert on that but what if I need the value or…

That is definitely a better solution.

We're almost there. Just to clarify, the script could search for the term "Scheduler jobToBeExecuted" or could that message be anything? Mike

Thanks to both of y'all for the info. That's exactly what I needed to know. Balki1429: Is that from one of the setup docs in STM? If so, I apologize for not reading through it more carefully.

That's good and rules out that it's a credential issue. I run the Recorder from my laptop and it works fine so it is possible. I just fired up Wireshark and captured traffic while connecting and the only port it is hitting on the other end is 17777.

Is one method better than the other? The only benefit in using the exit code that I can see is that there are more levels of severity than warning and critical.

That would definitely prove that the server is running correctly but what if I want to know if a user clicks an icon that should launch a citrix app but it just pops up and goes away for example or doesn't even do that? Do we now start monitoring the desktop for the xenapp.exe (not sure what the actual exe is) and alert if…

Which version of NPM are you using? Is your Advanced Alert builder a part of the web portal or is it a stand alone app on one of your polling engines?

You'll probably need to load the Snapin in your script so you're sure that the cmdlets are available. Here's one that I use when I'm connecting back to Orion: if (!(Get-PSSnapin -Name "SwisSnapin" -ErrorAction SilentlyContinue)) { Add-PSSnapin SwisSnapin -ErrorAction SilentlyContinue }

My bad. I see what you're saying. If it's a windows system then it can be done in powershell. The only catch I can think of is, you'd need to be able to write the first time it occurred to a file or db. So, the application runs and finds the Service has stopped, it writes that time to a temp file and returns "exit 0" for…

I can certainly try and help but as small as that difference may seem it's quite a bit more complex. Searching from the bottom up in powershell is a little tedious but that's not the real problem. SAM doesn't alert as things happen, it polls at a given polling interval. We don't use LEM (Log & Event Manager) but maybe it…

We are on 11.01 but are looking at upgrading in the very near future. I will try reinstalling the JobEngine and the Collector and see if that works for us. Thanks

If you want each service to have it's own output, you could write the services that are down to an array. Then finish with a for loop that increments the Message/Statistic index number by one for every down service and uses array.length to know how many times to loop through. The output would be something like this:…

I think you'll find that if you are adding your Windows nodes using WMI for the "Polling Method", you will probably have most of the metrics you're looking for out of the box. Don't hesitate to post questions here if you need help with custom Powershells later on.

Running as Admin resolved the problem for me too. Thanks

Good to see you back. Thanks for the link.

Thanks a lot. That's a great resource. I guess I didn't realize one would have access to the entire selection of variables. For some reason I thought there were only a set number of variables available in the argument field of a script component.

I did not use the Unmanage Utility in the end. I added some code to the powershell scripts so it only alerts at certain times. Scheduling Powershell Component Monitors

It would give me the correct polling times but the rest of the time it won't be green on the console. Wouldn't it be nicer to have the application appear "UP" when it's not actually failing?

That's exactly what we've done to get all our "Line-of-Business" views on one page. Great minds think alike.

Also, what happens when you run the Recorder from the server running WPM?

I don't remember having to open any ports but the documentation states to be sure that ports 80 (HTTP traffic), 443 (HTTPS traffic) and 17777 (Solarwinds traffic) are open. Run 'Test-NetConnection "Hostname or IP" -port 17777' from a powershell window to make sure the port is accessible. Which message do you receive when…

Thanks aLTeReGo‌. That did it.

Thank you adatole​. The batch file worked fine after having the security team allow powershell.exe to be launched by cmd.exe.