mesverrum · Observability Architect · ✭✭✭✭✭

You don't have to handle it, just set your UNDP to a tabular poller and it should be good to go. Those IP's are just the "index" of the snmp table in this case.

I use this SWQL in the custom query widget to see how long all my alerts were active. github.com/.../Alert SLA

You can rework the script mentioned in this thread so instead of using the ${Node.DNS} variable you just hard code the DNS name of the linux server you want it to try and access. thwack.solarwinds.com/.../does-solarwinds-linux-agent-monitor-certificate

I think the problem you will run into here is that SW alerts require you to tie an alert to a currently monitored object. If you were monitoring those servers themselves you would be alerted on the server. I dont remember there being any table in the DB that would keep historical MAC info except in the UDT module. Assuming…

The way I expect this will play out is some time over the next few months SAM will be updated to query wherever iSM stores this data. If you are in a hurry and dont want to wait for the product team to figure it out you can probably so some exploration to find the relevant WMI calls yourself and create a powershell script…

So in this case since there isn't a verb to do exactly what you intend in a single step you need to create the subnet, then do a set-swisobject on the new subnet and set its parentid to the relevant group. Takes a couple extra steps but thats how that part of the API works github.com/.../IPAM-Observability-2022.2

So the part that confusing you is that a hierarchy group is not just the folders you see there. It represents this sort of logical divide so that they can allow overlapping ip spaces. If you havent gone out of your way to create a heirarchy group you probably dont have any. Check this thread, you probably just need to…

My guess is that your login is not mapped into the flow database correctly, this stack overflow might help stackoverflow.com/.../the-server-principal-is-not-able-to-access-the-database-under-the-current-securi

Functionally any time you do new swisobject you are doing an insert to the db, and set swisobject is an update. If there is a verb then those do the work for you as if you were using the gui, I don't recall if there is an add interface verb on that table though. 

if you had some external source of truth about the interface then you could make something like this work, but the big issue to me is that most devices don't expose a way to reliably know their SNMP indexes unless you actually poll the device with SNMP. You could pretty much brute force it by running a whole lot of…

Based on the swagger im expecting it might need to be closer to this, but im not 100% and cant test today <span>[</span> {<span> </span><span class="hljs-attr">"Key"</span><span>: </span><span>"enabled"</span><span>,</span><span> </span><span class="hljs-attr">"Value"</span><span>: "false"</span> } ]…

Im not even sure if removing their group completely would trigger removing their individual entry, for example ive had occasional problems in the past where a user moved between groups in AD that both had access but with different permissions and their Orion account got stuck with the old permissions until I manually…

When orion first sees a user log in then i will create a local entry in the db for that user to manage their preferences and login history and such. If that user disappears from AD SW will never know because it only updates the users when they are able to log in. No AD entry means failed login attempts if someone tries to…

Yeah, lots of ways to make that happen. For example you could modify the query to have something like where p.customproperties.yourproperty = 'yourvalue' If I'm using them in dashboards i hate having to edit every query one by one so I'd most likely just apply a dashboard limitation based on the property or a group…

Even if you could display a timeseries or won't let you load 500 rows. That's asking for a page to take 40 seconds to load. Any time I'm working on a scale of hundreds or thousands i move to a model of only showing broken things. Wouldn't be too hard to display a table of nodes where the current response was above the…

Which brand of load balancer are you using? I don't remember exactly the names of the settings I had to use but I've had this problem in the past and had to make the user sessions sticky to the balancer they initially connect to

Have you tried to snmp walk the device directly to see what the raw snmp status code is showing? Vendors have been known to have snmp bugs where they report incorrect data and SW is only going to show you what it's seeing when it polls the device.…

When I was the in house guy for a big enterprise I worked on a model where almost anything i ever got asked to do became a script that ran in scheduled tasks. I don't like being woke up for on call issues or having a lot of tickets coming in. For a long time Ive felt that the best thing about Orion was that if there was…

I havent tested that attribute specifically, but my gut guess would be that avgresponsetime might be the rolling average across the last 10 polls, similar to how they do percentloss. It's also possible that it is just the latest response time value straight up with no math, there are plenty of places in the DB where they…

@"antonis.athanasiou" posted a modified version of the custom query widget a few years ago that allows you to do all kinds of style sheet modifications in the widget you are using. Its a bit of a chore to get it added to your environment, but i love it Custom Query resource with colors and styles Otherwise the OOTB custom…

If you need to span multiple object types you are best set by using groups, since they already allow you to contain the majority of the categories. You should be able to use the group logic to come up with basically any additive logic for types of monitored objects. Alert and Report limitations are actually their own…

Assuming there are not too many charts in the perfstack you are able to "build" a perfstack on the fly by manipulating the URL, so a script could translate all the object ID's by doing a bunch of lookups to find nodeid in systemA and replace it with equivalent nodeid in systemB. This thread from @"wluther" got pretty in…

I've wrestled this use case before and just opted not to use neighbor scanning. Wrote my own powershell tooling to use ncm and show commands to get what I needed from the devices and then pushed that back into ipam in a way where I was able to dictate exactly what it would and wouldn't do. The ootb behavior is pretty much…

This swql or variations gets used a lot for that github.com/.../Nodes w Polling Issues

Youd adjust them to match your needs and run them in SSMS or the database manager, these are making changes to the database directly. There is not a way to set different tooltips for different parts of the GUI, its going to be the same anywhere in the UI. In older versions with the Atlas diagrams there used to be a way to…

The real time widget cannot be modified to show a different time range, they are all hard coded to display the current rolling 10m window starting from the time you load the page.

I didn't like the limitations around those verbs, so I use variations on this example to take my devices, push them through a batch of discovery jobs, when those finish it runs sql to delete the junk i dont want from the discoveredinterfaces/discoveredvolumes/etc, and then I import what's left over. Of all the methods I…

The appinsight for exchange template monitors most of the relevant exchange processes but you will often find that seeing exchange use up 100% cpu doesn't actually tell you why or what you could do to mitigate it. You'll probably be looking at the performance counters and queues to see what activity it is that spikes just…

WLC gets polled every 10m assuming default intervals.

Assuming we are talking about custom properties you are incorrect. When they rolled out the feature to add custom props to the tooltip it grabs the first 5 properties for that entity type. Of course that is usually the OOTB props and a lot of people didn't realize we can control which ones show up. I wrote this sql query a…