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

Flackovic's solution looks good though i still cant test it at the moment. Had a thought though which is that netpathing to your load balancer IP/name/url produces a lovely chart of %traffic going to each endpoint and whatnot. Should work on :443 and :17778/4 

Your SWQL looks about right, except you probably want to set avg response time over a specific window using either a separate element or a custom property, the one used there is over a short range I believe. I'm short a test env for a few days- * Downsample(d, p) - Rounds the supplied timestamp d to the defined time…

Are you checking the box "impersonate..."? You're in the realm of creds and security, there's some org specific stuff that'll kick in now When you test personally it'll be using your creds/the creds you opened PS with You can feed in creds either by impersonating, which is running the script as cred X, or by feeding it…

is Dev-postal1 the polling engine or the remote server?

I'm away from a test environment at the moment - Usually that error means "Add a group by in the UI", you probably want to group by either Caption or Date, maybe both. You can stick all the nodes in a group (dynamically if need be) and then there's grouip availability commited to the DB, with a chart available on the…

Couple bits, changed your script ever so slightly to: $App = "chrome"##$args[0] $proccount = ((get-process $App) | Measure-Object).count Write-Host "Statistic: $proccount" Write-Host "Message: $App was found $proccount times" exit 0 Some of that might just be stylistic, but the quote placement and the .count method bother…

I'm struggling to understand the question, what problem are you having? To "Alert->Run NCM script" part would be adding a trigger action with the NCM action type, you make a service account with NCM rights, and place the script

(You can monitor devices in GCP as any other server or API endpoint though)

1) IIS front end, SQL back end, polling from polling engine on a per-device cycle. Data is polled from the engines on SNMP, WMI, Agent methods, and is collected by log/flow methods. Data works its way into the DB that way from a PE to a DB, and can then be queried by the UI and the alerting service 2) SNMP allows access to…

Actually, to confirm @"ssugar" , this is specifically for the login page?

This is a good question. I dont know a way to do this, I think it'd be doable on classic using the custom html widget and calling some js or something. I may look into this further. Not aware of a way of callling any variables associated with the user session elsewhere.

You can add them on classic with a iframe in a custom html widget, if iframes are disallowed you'd be best to call statuses and link to them for the moment. I think this is a feature coming, but not confirmed

Needs a custom script for the moment, I've got one but not to hand, I think there's other examples already on thwack. Yell if you cant find one. I do strongly reccomend to the solarwinds PM's this would be a quick win Alternately, ask whoever owns the server to move their task up a folder or two A downside of the custom…

I laid it out above really, but let's say you get a P1 where something breaks and a queue fills up, and when the queue is full you have a bad day. You have a session with people involved and figure out what occurred, you place a monitor on the queue and/or whatever broke. You have an alert with an action on it that resets…

Add the idrac ip as a node too

Checks out given the time range stated by OP Your memory is incredible!

Excellent correction thank you!

I dont think those are editable without getting into the IIS server with the webdev hat on (Hoping i'm wrong, might be)

In a custom SWQL you can do ServiceAssignments.tests.rtt > 40

You can indeed scan for event 322 using a SAM event monitor template, which you should have access to if you have HCO You probably dont need to, if you can see it as a scheduled task the sort of solutions mentioned here probably work fine. The OOTB scheduled task monitor doesnt see into folders for some reason, if you cant…

What Vinay said usually, sounds fine I usually like alerting on nodes with hardware sensor issues over hardware sensors, but that's a matter of preference with some tradeoffs You may need to monitor the ILOs/DRACs/Virtualization stuff in addition You can go beyond this but I dont think you'd need to based on your question

There's a few ways to do this You can use SWQL Studio and the WITH QUERYSTATS/QUERYPLAN verb to see the efficiency if you want (It would be best to SUM then ROUND usually) I think the easiest way to do what you want is just to put the 3 interface names in an OR in the WHERE SELECT ROUND((SUM(TotalBytes)/1024/1024/1024),1)…

The time-to-receipt formula will be something like:~ Poll time (max being the frequency, else half for average) + Alert Evaluation Time (max being the frequency, else half for average, usually 1min) + Condition Must Exist time + Delivery time Assuming no Group stuff, the device knowing immediately, and no escalation layer…

Ah glad to hear it! If you can, could you mark one of my replies as correct (maybe yours too)? There is probably an update coming in this space, hopefully soon, a tweak here and there would avoid most of the weirdness we've dealt with there

I think you missed the edit button twice part, you should end up here:

On a modern dashboard, you can create a new widget and it'll ask you to clone an existing one, use that On a classic dashboard you can hit edit twice, then select the element and hit the copy icon in whichever column it's in

(If it helps, if you get option 2 working badly, you can kinda transfer that to an option 1 or 3 fairly easily)

Addressing the other questions: You can address multiple "sites" in one api poller yes, though it's a weird thing to do You use the same credential if you do, which is a faff, and may put you into one of the other options (unless you use creds that dont use the cred section, but then they're probably unencrypted/stealable)…

Start with https://127.0.0.1:17774/SolarWinds/InformationService/v3/Json/Query?query= Change the 4 to an 8 if you're on one of the last few versions of solarwinds Then you want to write a SWQL query that produces a date, and then run it through https://www.url-encode-decode.com/ and paste the encoded result in Ages back I…

eg: