mesverrum · Observability Architect · ✭✭✭✭✭

When I was building out that bit I just created a few in the gui until I felt like I knew how to make what I wanted in SQL, always backup your db.

That's why I suggested scripting it, should take about 10 lines to re-automate it with the api until solarwinds figures out their bug. If you are spending more than 1 man hour on this you are throwing money away.

I dont have my solarwinds server handy to get in and play with it right now and see about reproducing your problem, I just moved across the country last week and havent got around to setting it back up yet. Off the top of my head my first thoughts would be using the wrong credentials inside the tool, should be the same…

I actually happened to be loading up my custom widgets for this use case into a client environment when I noticed this thread got updated. This is the SWQL I use, a little bit less leg work for the end user and it attempts to clean up some of what makes traps super ugly SELECT MessageDateTime , let.Name as Type…

I vpn'ed into the lab to check, whatever you were doing with your datetime where condition, that seems weird. Are you trying to get entries just for yesterday, or the last 24 hours? I guess what I'm seeing in your original query is a data point for every hour between the start of the day yesterday and now? This is what i'd…

To be honest, building a 3rd party integration into Solarwinds only through direct API edits is pretty challenging, I would say nearly impossible. You will need to have a very deep understanding of the Solarwinds application, the database (they do not publish db information so you will basically only learn by trial and…

Unless you mean you want to combine multiple interfaces into a single widget to get an aggregate. That's doable as well, I'd have to sit down at a computer and write the process out though.

This block of code would be set up to execute on the orion server (or a system that is pointing at the orion server), it doesn't matter if the node you want to delete is linux, windows, or a router. You define the conditions that you want to use to determine which nodes you are deleting on line 2

The easiest way to tell if someone is giving you SWQL or SQL is that the table names in SWQL almost always have "orion." something in them but you would never see that in SQL. It is a pretty commonly seen issue for the people who do a lot of custom SWQL that the Custom Table/Report validator for SWQL has some bugs with how…

I am not on VPN right now to confirm, but you might want to make sure that the n.lastboot value is not in UTC since you are using SWQL and it converts nearly everything to UTC. if it is, then just change your getdate() to getutcdate() to be accurate. A few hours probably doesn't matter in a 30 day reboot report but other…

CPU Monitoring With Process Snapshot (PowerShell)

I use this feature all the time with no delimiter at all, the "any" options seem to natively use spaces as their delimiter, and the all would be the one you select if you want to exclude a string that includes spaces. I'll typically do something like interface description does not contain any keywords: loopback -0000 null…

No that last chart style is the new perfstack feature. So far you can't embed these into other dashboards and such except as a link to load a saved version of the chart, I get the feeling that as they rebuild more of the web console in the new format this would become the default charting tool but for now you have to load…

Just an FYI, I am extremely wary of installing the trials into prod systems. I recently had to scrape all the old references to APM v 3.x from a client's db before we could install their new license for SAM 6.6 almost a decade later. Safer to run the demos completely self contained with their local sql express install so…

Sounds like you might want to include some notification indicating when that throttle is being applied, otherwise users have no idea what's wrong

I've honestly never tried, is this not for a work environment where you can just spin up a windows vm using eval licenses? I mean technically you could even do that from your linux workstation with something like KVM but I would assume that if this is for work then they already have infrastructure in place to make that all…

You are also likely to run into issues when you get to the notifications because there is a special syntax you need to use on these X or more objects type of alerts. For the alert if you want it to list off the names of each down server you need to surround your virtual server name variable with <<< >>> Anything inside the…

I've done this using custom properties, or if you just want to do it based on group membership then you get into having to do it with custom SWQL. I've attached one example of a SWQL query I use on my Group Details page to bring up all the major node metrics. You could make it a top 10 by ordering by whatever column you…

This Custom Query widget can be added to the node details page and it will get you a list of all the maps that the node appears on. Making a caption searchable form of it will be a bit trickier so I'll get to that later on when I have a real need for it. SELECT --AccountID, Data, DisplayName as Name , '/ui/maps/viewer/' +…

Errrr just realized this is a hyper v thread. In which case it falls under the requirements for monitoring hardware health on any other windows server. Dell/IBM/HP all work if you have the appropriate agent from those vendors installed on the server.…

So I've seen some display glitches where if you copy things from a "base" map and then redo all the nodes some views within solarwinds still retain info about the old nodes that used to be on the map. Kind of a nuisance, so I just got into the habit of building myself map "templates" by right clicking the white space and…

Interesting, probably need to dig into what exactly the corruption you are encountering is, must be environmental otherwise there'd be hordes of angry upgradees blowing up the thread

That isn't a SQL report, it's SWQL. Flip the radio button on the custom report to SWQL and it should execute better The easiest way to recognize SWQL from SQL is the SWQL will have most table names starting with orion.something, SQL doesn't.

You can actually just blank out the thresholds as well

Tim, I think you may be misunderstanding how muting works. If your NOC team mutes a node then no new alerts will trigger from that node at all until the mute is cleared. Muting isn't intended to hide alerts, it suppresses new alerts from triggering on that node. In the example I gave the Node down alert wouldn't show up…

Wouldnt it be fairly easy to just point your script at the windows event log and check for the timestamps of the 7035/7036 event that matches your service? Only return warning or critical when it gets above 5/10 mins or whatever. -Marc Netterfield Loop1 Systems: SolarWinds Training and Professional Services * LinkedIN:…

Try this info out: PSEXEC to start remote process -Marc Netterfield Loop1 Systems: SolarWinds Training and Professional Services * LinkedIN: Loop1 Systems * Facebook: Loop1 Systems * Twitter: @Loop1Systems

To be honest, in contrast with the direction things have been going in Orion, I drive my clients away from having multiple variations on the Node Details views. I've always found that it is not that hard to build a single node details view that covers all cases, although it would be a bit easier if more of the built in…

CPUCount was added to the nodes table in a recent release, you shouldn't need to join anything anymore

Bah I forgot to include a datetime, thats what I get for doing these off the top of my head. Here's some screenshots of how I got it working. Start by adding the custom chart resource to the view > Uncheck 'Use current network object...' and click Select Datasource From the drop down selecft advanced database Query, make…