mesverrum · Observability Architect · ✭✭✭✭✭

It shouldn't be that surprising, Samanage already had over $1.5 mil in monthly recurring revenue before SW bought them and I'd hope they've grown that number since they got it. I'd be surprised if WHD brings in that much revenue in a whole year considering that I was a SW consultant for 5 years and can count the number of…

So you would use the average() and max() functions as normal, but in the group by section you include your downsample. Something like select top 100 downsample(cpu.DateTime ,'00:15:00') as [15m_Block] , cpu.nodeid , avg( cpu.AvgLoad ) as [Average] , max( cpu.AvgLoad ) as [max] from orion.CPULoad cpu group by cpu.NodeID ,…

My company closed down several of our datacenters and has a directive that all new projects are cloud based. We keep a small 2-6 node VMware cluster in every warehouse/shipping facility for the apps they use because latency really sucks for their scanners. We spend a staggering amount of money for it but the leadership is…

Although in your list LEM is not part of Orion and can't integrate really. It's a totally different thing. 

Yes this is the most common way to use the tools, when you run the installer you just check the boxes for which modules you want to include. 

What do you mean by sql components? You need to have a working sql server to connect to with an SA level account to set up the databases but the installer does all that work for you and deploys everything the app servers need. You basically just next next next through the install and it handles what needs to be handled. 

I feel like Orion is pretty robust about not being too worried about specific services. I've killed services 100 different ways over the years, the main trick being that if you have HA you need to disable that first to avoid having it try to bounce you over to the backup server. 

I've not done this exactly, but an idea that came to mind to make a "report" that only runs if a certain condition is true is to use an alert where the action is to email a web page. So you build a report or dashboard that lists out all the information that you think is useful related to node reboots over the last 24 hours…

How quickly do those snmp.getTree operations complete? Also seems like you would want to just pull the community as part of your nodes query and pass it into the second argument of the snmp.open.

This looks pretty close SELECT cs.Component.Application.NodeID ,cs.ApplicationID ,cs.ComponentID ,cs.Component.Application.Node.Caption AS NODE ,cs.Component.Application.Node.DetailsUrl as [_linkfor_NODE] ,cs.Component.name AS [CMPNT] ,cs.Component.DetailsUrl as [_linkfor_CMPNT] ,'/Orion/images/StatusIcons/small-' +…

This seems to work for me SELECT top 1 it.NodeID , it.InterfaceID , it.DateTime , it.InAveragebps , it2.DateTime as previousDateTime , it2.InAveragebps as previousAverageBps , it.InAveragebps - it2.InAveragebps as Change FROM Orion.NPM.InterfaceTraffic it join orion.npm.InterfaceTraffic it2 on it.DateTime >…

Other - self service, company is too big for my team of 3 to service the requests of everyone, learn their tools and explain to them how they should be monitoring their stuff. So we try to make it so those who are able can DIY what they need to free up bandwidth for the ones who need more hand holding. The hard part there,…

You don't connect to the db, you connect the your main Orion server, swis is the solarwinds information service that you'll find running on that. 

So the log DB is structured very differently than most traditional databases. The whole thing is structured based on the timestamps when the data came in and when a table gets big enough they cap it off and generate the next one as needed. So a query to filter by anything except timestamps is going to be tricky and you are…

This should work, assuming it doesn't take too long to calculate. SELECT g.DisplayName as Site , Case When n1.mm = 1 Then 'Jan' When n1.mm = 2 Then 'Feb' When n1.mm = 3 Then 'Mar' When n1.mm = 4 Then 'Apr' When n1.mm = 5 Then 'May' When n1.mm = 6 Then 'Jun' When n1.mm = 7 Then 'Jul' When n1.mm = 8 Then 'Aug' When n1.mm = 9…

Good point, at my job we have a tool that is used by both IT ops and Security (not the IT kind of security) for emergency notifications via calls/sms/etc

Assuming you have SAM yes, local account creation registers an event 4720 in Windows. There is an OOTB template called Domain Controller Security that has a monitor for that event ID, but if you want to monitor it on all servers then you would probably want to copy it out of that template and create a new template that you…

Network discovery works on essentially the same ports as normal monitoring. So from your polling engine you need to have potentially snmp on udp 161 to talk to linux or network appliances, and you would need all the normal WMI ports for WMI based polled, tcp 135 from the poller to initiate but the responses come back…

Don't even need anything fancy for that, the built in report writer would do fine. Select the object as Nodes, the column to display would be Caption. There you go. If you want to get more elaborate you can also add the dns name as a second column, but not all nodes report their dns names so I would rely on caption as the…

^ just confirming, in the past i have had to stand up intermediate instances of sql to bridge us through upgrades. I will say that the extra work and risk of migrating through all those intermediate versions is rarely worth the extra hassle compared to standing up a completely new instance on a second server and just…

^ Seems like it says it all there.

Orion only runs on Windows Server, and those can be run with no license for like 180 days, at this stage I wouldn't stand up a new system with anything less than at least server 2016.

Installing Orion on desktop OS is generally not recommended, https://documentation.solarwinds.com/en/success_center/npm/content/system_requirements/npm_2020-2_system_requirements.htm# 20m to download is slower than I'd expect, but the full install is around 3-4gb depending on which modules you are trying to get so it can…

Assuming you are writing this in sql it would just be a left join from nodes to volumes where volumes.volumeid is null

You say you have npm, have you looked at the web based mapping they rolled out over the last few versions? It shows utilization by default but connections turn red and yellow when there are errors on the link. 

OOTB there is no built in filtering for the volumes unfortunately, what I do is run a script that does the discovery, then edits the results of that discovery job in SQL to delete anything I don't want to bring in, then it executes the auto import of everything that was leftover. It's a bit of a hacky work around but for…

Which SolarWinds product? NPM is exclusively Windows based so you should not expect to be able to SSH into it. Try RDP with a domain credential. There isn't a SW specific password recovery process if you can't login to the server, that would just be standard windows stuff.

Seems doable, just need to find the correct API address to call to get the counter. I tried looking up some aruba API documentation and couldn't find anywhere with it published but it's probably around somewhere.

Piggybacking, I have had to do a lot of custom work over the years when i was consulting for clients to more tightly integrate Orion and IPAM, because natively they pretty much act like they aren't even related to eachother. Discovering an IP in IPAM has no built in capability to add that node to monitoring, and adding an…

This is an example similar to the code I use for muting and managing github.com/.../SupressAlerts.ps1