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

Why would you want to change an ID?

Could you mark the answer as correct?

You want a WHERE condition or several Are you happy exploring the UI? Am I correcting in thinking you want 4 KPI numbers per section (Down/Critical/Warning/up) and 8 sections (Office routers, Office FWs, DC A x3, DC B x3) for a total of 32 numbers? (If so, seems like a lot, could do 3 Pie charts?)

Really? Spicy

There's an SNMP component which inherits I believe

こんにちは！行ったり来たりの翻訳をしましたが、日本語は少ししか話せませんし、漢字はまだ読めませんが、何度か訪れたことがあり、この国が大好きです。 日本では太陽風に対する熱意はあるのでしょうか？そこに英語を話すソーラーウィンズスタッフが欲しいですか、それとも必要ですか?将来のある時点で訪問や短期契約を正当化したいと思っています。 そこでコミュニティを構築しているのを見るのは素晴らしいことです。

I'm not sure of the specific thing you're looking to output, but each block would be a query, and your queries are probably selecting from the Nodes table. You can either use the UI, or write the queries by hand. Writing by hands means these probably include a COUNT(ID) ... GROUP BY STATUS or something. Having a number per…

It's in the database already in the Hosts table, though offhand i'm not sure if it's a VMAN or NPM derived field

The answer is definitely "yes" but not sure exactly what'd be ideal. The answer is something like "whatever url and headers are needed" on the SW end and something a lot more specific on the juniper end Headers get url encoded too, so there's some character replacement going on. I feel like I should write something up but…

Beaut!

Seems fine from a SWQL perspective, that error doesn't look right, not sure what's happening on your end, but I'd say check the code/spelling/aliases etc

Yeah that's a good way to go Availability metrics: -Site available/Not primarily http errored -Db responds to basic select statement -Key Servers up -Key function X works -Key services all running -Each major component connected to each other major component etc Performance metric - Count of records per time of important…

Is that SQL or SWQL?

I'm literally using it in prod, so it does work, might just need to work with it a bit, work the query etc.

You're using an ID not a URI, gotta switch over

If line 35 fails there's a problem I can only describe as "ODBC-like" in which a earlier version of the SQL processor is default and not supported, I wrote this with the intention of making that sort of problem as unlikely as possible

My understanding of the issue connecting to azure SQL MI is independent of any IP-based issues, so maybe those exist too, but if there's no IP issues you'd still have problems You could always do something like set a monitor against both IPs, group em, alert on if both down, for example. Could build an LB into the PS I'd…

### Module-less SQL (Azure SQL MI) template [Net.ServicePointManager]::SecurityProtocol = [Net.ServicePointManager]::SecurityProtocol -bor [Net.SecurityProtocolType]::Tls12 Function SecureStringToString($Value) { [System.IntPtr] $BSTR = [System.Runtime.InteropServices.Marshal]::SecureStringToBSTR($Value); try {…

I've got a powershell monitor for this, it doesnt work with the OOTB sql monitor as far as I'm aware

Getting the data right from the report is fine too, but just not how i'd ever do it so idk

You want https://solarwinds.github.io/OrionSDK/swagger-ui/#/Query/Query /query or Get-SwisData depending on if you've a preference for modules or not eg $swis = connect-swis $params $query = " @ whatever the query on the jobsdata report is @ " $results = get-swisdata -swis $swis -query $query

Calling it now, someone hit an install button somewhere

You've posted in the wrong forum for NPM FYI If you've got no other connection to the site, i'd probably try and monitor the next neighbours for the ISP's kit, and/or some way of asking the ISP directly. You can then ping the site, and the firewall interface on your site If site down AND ISP down, then likely ISP Is the…

It's not all doom and gloom dont worry, probably someone's done most of these components before, or something similar Website -> Pingdom/WPM/SAM probably does it DB -> Sam/DPA/SQl Sentry probably does it etc etc Sometimes a database responding slowly or being in a strange health status or the date of a specifically…

Aah shoulda said you were doing it for doing cert monitoring, loads of those around Much of the time, though probably not for this, i'd default to invoke-restmethod over invoke-webrequest, the former's better when receiving structured data and the latter's better when you need a cookie or something more web-ey. There is a…

Easiest answer is to stick em in a perfstack There's plenty of classic charts or even 3rd party stuff which can accept multiple objects, and there is a multiple object classic chart called that I believe. The UX of it across the platform should be improved, I think perfstack and modern dashboards were the attempt at that…

Not sure what else to tell ya other than it's there and that's what it's called, try the swql search bar top left I suppose you could just use a CASE statement in your report given the table above sans join

(I think the classic availability chart is a histogram, right?)

I have an equivalent of this script, and i've not run your script yet to validate, some notes: The Write-hosts that dont start Message or Statistic i'd comment out unless testing, solarwinds tries to parse the output, so while they should get ignored it's all potential downside in prod You dont exit in the catch block, so…

Keep an eye out for the different types of azure SQL, it makes a difference DPA cant run on Azure SQL MIs right now