mrxinu

Did this ever get resolved? -- Steven Klassen Developer Analyst @ Loop1 Systems, Inc. http://www.loop1systems.com/

Then if you pull the current statistics on that custom poller you should get 1..n rows indicating which routes are present. If you had another table with the routes you expect to see (custom, created in the same database, but with a unique enough name that it doesn't interfere with the stock tables) you could do a LEFT…

Uninstall .NET and try again?

Good question. The report is the only thing left on the filesystem so it might make sense that they'd cache the pulls of the report definitions for a few minutes at a time. I'm not sure.

Yep, sending you another link.

By work I assume you mean it starts the service? Can you add some debugging to your script to write the arguments out to a file somewhere so we can make sure they're interpolating properly? If you're winding up with blanks, there's a problem with the variables. If you're getting values but they're the wrong ones, that…

Ah well, there you go - it's missing the PollerType key.

I understand that the template works. What I was suggesting was removing SolarWinds from the equation for this single system by running the WMI query manually to see if you can get a more specific error (if one exists). Sometimes applications will give generic errors on exceptions they weren't expecting to have to catch.

It's my pleasure, sir. Give me a shout if you come up with any more of these.

Ah, sounds like that's the case. It was a typo, though - ^.*?set admintimeout 480 was what I was shooting for.

That's fine, too, but using a custom property would allow you to add/remove systems that should be part of that alert without affecting the definition of the alert. Something like a boolean custom property called "TemporarilyIgnore" that they could set true/false on a per-node basis. Then your trigger condition could be:…

And to you my friend. =)

Usually if there's an interesting statistic to be returned I'll go that route. The only time I'll skip the statistic (setting it to 0 every time) and go with the up (exit 0) and down (exit 1) is when the thing I'm testing only has two interesting states: up and down. The number of active print jobs would be a good example…

With syntax highlighting: select c.NODE ,'/Orion/NetPerfMon/NodeDetails.aspx?NetObject=N%3a' + ToString(c.NodeID) AS [_LinkFor_NODE] , c.ComponentStatisticData as Total , '/Orion/APM/MonitorDetails.aspx?NetObject=AM%3a' + ToString(c.ComponentID) AS [_LinkFor_Total] , d.ComponentStatisticData as Success ,…

Alright, it's out there!

You got it. Let me get clear of my quick launch and I'll hack something up.

If you drop this into a report you can have >1 report element. I would have one at the top that only lists the engines and their element counts and then put the original query from up top in as the one below that.

You are very welcome! I'm glad it fit the bill!

I try to stick to SWQL when I can only because it abstracts a lot of the complexity away and insulated my work from changes in the underlying database schema. For example, to do any work with custom properties it involves a minimum of 4 joins.

Do you have access to your database? I don't have an install handy, but I've never used the 'Rate' for the custom poller in an alert. There are two other options - one of them is prefixed with 'Numeric'. I want to stay it's either 'Numeric Statistic' or 'Numeric Status'. That's the one you want to use. I'm not sure what…

Screenshots?

I noticed it defaults to 'no reports' now. Annoying when you're not used to looking at it, but I guess least-privilege is the way to go. I'm glad you sorted it out. =)

Awesome! I'm glad it got sorted out.

Okay, then we're not matching our if/elseif conditions. Try this: # this is kind of a throwaway value since we're using the exit status for up/downWrite-Host "Statistic: 0"if ($indexDetails.State -eq 'Active') { Write-Host "Message: Search Index is active!" Exit 0}Elseif ($indexDetails.State -eq 'Degraded') { Write-Host…

You're very welcome - here you go. $ErrorActionPreference = 'Stop'# database configuration for SolarWinds NPM$dbHost = '<servername>\<instance>,<port>' # or just the <servername> if not instanced$dbName = '<database_name>'$dbUser = '<username>'$dbPass = '<password>'# create the SQL objects we need$sqlConnection =…

I can't find any reference to the sys_GetPerformanceCounterDataByName stored procedure - is that something created by your application team by chance? Maybe there's another counter aside from EmailProcessorPendingEmails that you could be querying that would give you the information. Something like, and I'm totally making…

Do you have a NPM custom property called 'Owner'?

Hi oludads​ - you mention an outage duration in minutes and then a time format (HH24:MI:SS). Can you tell me what you'd envision that last column looking like? What would an outage duration in HH24:MI:SS look like? -- Steven W. Klassen Programmer Analyst @ Loop1 Systems http://www.loop1systems.com/…

After that sunny reply, how could I resist? The following gets us most of the way there, but the half hour on the end time has me thrown. I'm used to dealing with whole hours but the datepart routine on the hour doesn't work in your case. Change the bold portion below and see if that gets you what you're looking for.…

Odd. I sent you a private message. If we can fire up a gotomeeting maybe I can sort out why you're seeing this.