adam.beedell · SRE/Machine Learning Student · ✭✭✭✭✭

This reply should have more love!

I looked into this and couldn't figure it out. You can pull a bunch of metrics though that should be sufficient for almost all purposes

There's probably an awful lot of SQL in text files, if you look through the installation logs you can see a number of them, it's not very human-readable The main thing this is for as far as I'm aware is to create the logs and netflow dbs if you've not got em already. It's not to deal with other stuff installed on the same…

Hi Rotto, Netpaths, while lovely are inherently a bit slow, there's like 10 min calc duration as default and It doesnt go as low as other methods If you want speed, you probably want traditional polling at a high frequency or event-based methods, where the event ones are near-instant There isnt a great way to scale…

2023.2 has a bug in which your alerts can get garbled. If they're garbled you can get errors, if you get errors you can get weird behaviour like this. 2023.3 is much better If you had an alert that was previously not storming, and now is, I'd check them and their SWQL to make sure it's not all messed up You can probably…

Saw reference to Rouge APs being detected by essentially asking the controllers what their view is. I believe that's what goes into the "rouge APs" widget OOTB HCO is like, patch to 2023.3, switch to regular licensing, and up the products to like 8. So if you've got the networkey products already it probably isnt changing…

Have a look in f12 devtools and see what's taking a while It's really surprising to me you'd see this with low RAM. I'd expect SWIS would be ballooning Probably one for the support line really

I think none of this is "supported" as such, but this query from earlier should prove it's possible select top 1 n.NodeID, CASE when n.nodeid = 3 then (select top 1 '1' as one from orion.nodes n) else 99 end as test from orion.nodes n Could you supply either a bit more of the code (with some blanked values) and/or the…

This is a bit of a brainmelter. Does the subquery return exactly 1 "cell". 1 row 1 column? If no, that's your problem. If so.. not sure offhand what the issue is. The error might help. This structure is very unfamiliar to me. I think i'd do something to try and join all the data together (maybe a jazzy right join to…

In an earlier iteration there was a huge red icon on a wallet-looking icon next to the bell on the top menu bar. I dont see the wallet icon any more so I guess that may have been disabled. A drop down for "Resources" or "Learn" or something next to settings, or in the same style would probably be more welcome. Red colours…

You might be screwed there. Classic's would do it. Excellent thing to pop toward the UI people/Featurey channels mostly as an example of flexibility asked for elsewhere

Hi Mike, agreed, this could use an update for sure Often I find for component alerts its better to alert on the application level even if it makes the macros a pain afterwards, just for noise reduction. If you do that you make other problems, but this particular one gets easier Else you can do stuff like trigger a write to…

You probably want to go with option 2 then, creating a report-type page with that time condition in the where, and repeating for each region

This seems like you're working on something awesome. Would love to see some of the PS if it's available, i've struggled with this in the past but havnt tried recently. Offhand i'm not seeing a fix yet, but I might need to work in this space fairly soon

Personally not seeing it, had any errors? Seems like some port exhaustion going around

I'm not familiar with this sort of security thing, not my space, familiar enough with the tool though. I've seen a rogue APs section in solarwinds before. I think there's a report off of NCM or something. Could you elaborate on what you're hoping for it to do>

Yeah this is a very unwanted feature in it's current implimentation, please remove

Interesting one. So the link is what you're putting in in the link field, any clickthrough issue is there. What you need is something to link to that displays the down devices right away. A few options come to mind 1) Build a complex URL - Stuff like perfstack and moderndashboards can have filters and attributes applied in…

Much like the others reply I think I can only offer partial help on a couple bits 1) In the magic restore button - > Database restore to temp location from earlier backup, copy just some stuff from the triggeractions (sp?) table 2) Failing that the Copy or Assign stuff - If you've got a source of an action, you can assign…

Weird. If it helps you can also get any other value produced in an API poller, but its a pain and it puts some restrictions on ya

Upgraded last week, had a "normal" level of difficulty and everything's been fairly smooth since upgrade day. Tentatively marking this down as a stable version internally

SELECT TOP 1 * FROM [dbo].[NetFlowEndpoints]

Click the button you've highlighted, then something like you could do this in a bunch of different ways mind

* Add a where name not like "appinsight for sql%" to your application alert* The application Type is quite nice for this too * Make some other alert setuo for appinsight for sql 

Add a vsphere, usually. You can add an esx in the 'virtualization polling' creds section, or something named similar to that

A phrase in your application alert like "And ApplicationName not like 'NOTPRODUCTIONYET%' should cover your dev actions. Then rename the template when stable.

Doing this directly in solarwinds is unusual, you'd usually do this in whatever system your tickets exist in. If you wanted to, you'd hope there was an api available, and i'd probably stick that in an api poller or SAM component.

I'm a bit confused by your screenshot, but basically a column swql syntax is: " SELECT column, another, anotherifyouwant FROM table " In that it's comma separated things immediately after a select. You can get fancy with it, but I think that's a lesson for another day

working over here. You should be able to see something about where it's failing in f12 dev tools

If you remove the WHERE and GROUP BY the application name in the query above you basically get what you want I think. Technically to reference the template you're probably joining to another table, but it'll report the same unless people are messing with names a bunch. If you've not used queries much you can export them to…