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

No, but usually with SNMP you're defining values on the host OS, then just collecting them with SW. Might be better placed in a mainframe z/OS forum

I've seen a post on there once on this topic, and i've no clue how to find it, but someone's done it publicly What you'd do would basically be 1. Make Classic Dash 2. Make Custom HTML widget 3. Find your mapping product's "Plot stuff and display a map" UI, get used to how that works 4. See if your solarwinds location data…

The short but unhelpful answer is "Topology data", if there's some of that it can be auto-mapped to some extent. You can get that from SNMP routing data, as well as other sources If you want the full network map of everything I dont know of any one-step process to get there, usually you've got to not-do that because it…

https://documentation.solarwinds.com/en/success_center/orionplatform/content/core-agent-requirements-sw476.htm and https://solarwinds.github.io/OrionSDK/swagger-ui/#/ 

In addition to what JM's said below, which is correct, i'd think about it like this: Get the data -> Put it in Solariwnds -> Take actions off it That workflow works for most stuff, to expand, in this case: Command that gets the data (This'll be on the thingboard) -> Put that in solarwinds (This is the stuff I mentioned…

Reboot ya box. Maybe reset the job engine files, check firewalls.

Interesting hyperlink you've got there

That did help, and enabled now thank you!

Seems like the API security hole is bigger than the SNMP security hole Install an agent to send traffic over 1777x Or setup something weird like: Device goes "Hey SWIS i'm ok" Solarwinds says "Device says it's OK", ELSE "Device says it's broke" OR "Device has gone quiet"

Either the API monitors function or the SAM script monitors function are good for this.

Hi Jamie, You can get the cert information in, just have to use one of the script templates. The difference is if the cert is read out to the "console" or to "message: ", where the message value can be recalled but the console one is discarded. Does get fiddily with multi-line messages though This works in a powershell…

Sounds like either you use the status/availability/percentdown stat, or it'll be very hard Probably "flapping" is in the warning/critical sections, itherwise it's not a defined status so you'd need to remove it from one of the others. Same with Unknown and N/A Might need a bunch of case statements and a CTE or two

No You can CTE but i've not seen it documented really. You can stick it in the FROM statement or the JOIN bit, possibly in the WHERE, bracketed in each case No WITH command in the same way though

Had a check in my environment and it looks unaffected in the way yours does. Not sure what'd cause that, but probably a local interaction in there somewhere

as part of a bash script action sure, otherwise you can go curl -> postman -> something else or rebuilt it in powershell or whatever

You can put multiple headers in the url usually with endpoint/.../something

Could you cover how to get alert stack with non-HCO/Moduled/On-Prem-Solarwinds-Platform-style setups?

01349560 - Mine, now I know more it's not got the best title, it'll do, troubleshooting info on it is solid

See if you can catch it outside the "test" page/s. on it's dash or in the DB. If it always works when prompted, but doesnt do it on it's own half the time, that's probably a scheduler/scheduling issue Check your polling rate, completion rate Try deleting the scheduler/jobs engine cache db thing if the above's healthy.…

I do this a bunch This is waaaaaaaaay too big a question, take many steps back Figure out what you want to do Figure out what the components would be to achieve that Figure out where you'd store stuff you need to store Find the API support docs for each system Find working creds for each system Get network access working…

This was my Monday/Tuesday! Could you pass me your refs and what the offending line looked like so I can convince my support tech to bump it far up the chain?

Any idea what these say when they're in the grey status?

If this is a machine learning problem I feel like it should have spotted the "PST" thing by now, maybe something's wrong there Probably something wrong with the reputation system too

Try this - On the Scope or Trigger condition for a broken one, there's a drop down arrow on the right. Press that, export the SWQL Paste into notepad or whatever, and add line-breaks until it's easy to read Compare it to the UI carefully. One of the conditions has probably been translated from UI into code badly. It'll…

I'd check the account access first, then delete the website and rerun config wizard to rebuild it under a change process

Feels like either something's being picked up wrong or missed, hard to say what without being in your environment. I'd maybe see if there's any way to deep dive on any one of the events for further hints. You could uninstall powershell but i'd really really not recommend it.

Delved into my version of this issue over the last few days, there was also a conversion happening where it didnt need to, for me it was a "__00" string into a number or something simmilar.

Hmm... look in SAM, disable custom powershell monitors, disable all the appinsight templates, see how much it's reduced? You could probably uninstall it but that sounds like a disaster waiting to happen

You're trying to use the -ignorecertificateaction thing right?

Rabbit's just queueing stuff This'll be "you have powershell monitoring in your solarwinds platform somewhere. And that is probably fine"