mesverrum · Observability Architect · ✭✭✭✭✭

select n.nodeid, n.caption, n.ip_address from orion.nodes n where n.uri not in ( SELECT distinct d.parenturi </code><code> ) and n.uri not in ( SELECT distinct d.childuri )

There's not really anything in the API that would give you those test capabilities, the way we historically have done this kind of thing is to just create it how we think it should be, wait a few minutes and check if it turns unknown or up or down.

if the account has access to the orion server that tells me it is a Windows account, you don't use the (v3) logon for Windows accounts at all, it is specifically just local orion accounts. That's your problem, not a bug, WAI.

The group by functionality in the web report/report writer is not at all the same thing as doing a SQL/SWQL group by. Thats really just a presentation layer kind of thing and there is no SQL equivalent to do that.

Looking more closely at the error it might not even be on the Orion API side, its complaining about my use of a replace so that might have to do with older versions of powershell needing different syntax for that

So you need to do the aggregation in a subquery instead of left join orion.alerthistory ah on ah.AlertActiveID=aa.AlertActiveID and ah.EventType in (2,3) i would try left join ( select count(*) as number, AlertObjectID from orion.alerthistory hist where EventType in (0) --this is the event for a new alert trigger, which…

I've been frustrated with that behavior on agents forever. In many cases you update the agent via the web console and they go down and you check them on the server and it says it's still pointed at the former polling engine. I've found that installing the agent again from the web console is one way to get around it,…

@"bmallon" I will say I never trust anything a first line support rep tells me. They are often the last to know anything real but many view it as part of their job to sound like they know what's going on. *Now I am speculating here, but I'm not pretending its official from anything* From the detailed breakdown announced so…

https://docs.aws.amazon.com/AmazonRDS/latest/UserGuide/USER_SQLServerWinAuth.html Haven't used AWS RDS but I've done a few others, you need to make sure your account has at least ALTER ANY LOGIN and probably ALTER ANY USER, not sure exactly how AWS doles those permissions out but that will get you past the point you are…

Feels like all you really need for auditors is select * from nodes

@"mrxinu" hit most of my build list as well, all of those are indispensable for me. Tacking on for good measure, * mRemoteNG or a similar session client - My tools server also functions as a jump host into a few other segregated networks that I can't hit from my workstation. * SSMS - You can also get by with the vs code…

For anyone that didn't already see it, HF2 was released a few hours ago. I just finished my upgrade, so far so good but I'm testing now to confirm the security changes didn't break anything for me.

That's the spot, when you import the results of the discovery just make sure the box doesn't get checked.

Most likely you are running discoveries and have left the box checked for the appinsight templates. When you set up the discoveries you just uncheck those and you should be good.

I don't use SAM to monitor those myself, but I expect the best method would be the API poller against Cloudwatch. https://documentation.solarwinds.com/en/Success_Center/SAM/Content/SAM-API-Poller-How-It-Works.htm

https://community.spiceworks.com/topic/2118422-enable-snmp-on-workstations

in the past when hammering at my server with scripts I have maxed out the number of connections from the sql connection pool. It is set to 1000 by default and the timeout is a randomized value of 2-4 minutes that can't be shortened. You can increase the max number of connections per the below documentation. I typically run…

You don't really even need to do all the agents stuff if you don't want to. Each APE would be using their local routing and dns resolution, you would just need to turn on the settings to allow duplicate nodes and it may get pretty confusing in your manage nodes screen, but it'll work.…

The separate version of DPA is essentially a dead platform and has been for a few years now. VMAN could be installed in a separate server, but the Orion based version is really designed 100% around the assumption that all your Orion and VMAN data would be integrated. Separating it makes it so you wouldn't see any of the…

Yep you got it, those legacy widgets and views are all hard coded to look at the old database tables that are no longer in use and nobody thought to remove them from the product at the time.

SNMP always has been disabled by default on all versions of Windows. https://docs.microsoft.com/en-us/troubleshoot/windows-server/networking/configure-snmp-service

That is indeed the process, won't affect licenses unless you change your hostname I believe. The agent may complain if you use server based polling, but if you use agent initiated it will not have an issue.

It sounds like APM is the term for what you are asking for. SolarWinds doesn't sell an APM module for Orion, but they do offer it as a SaaS with their Appoptics platform.

Have to upgrade to 2020.4 https://support.solarwinds.com/SuccessCenter/s/article/SEM-Legacy-Console-Adobe-Flash-End-Of-Life?language=en_US

good luck on that wishlist, I did a lot of migrations over the years and I was able to extract most of what you are asking for, but it took me a long time to get that intimately familiar with the db and api to get all the parts in and out. I don't have all those code samples handy but you can check my github and there's at…

I always felt like that part of the url naming was a bit deceptive, AAT: comes from the alert objectid if I remember correctly. Something to watch out for is that value is not unique to each time the alert fires, it is uniquely generated based the combo of the alert and the object being alerted on, so you will notice that…

I'm fairly certain the CW would reinstall it any time you ran it. I've never taken the extra step to try and actively remove IIS from my primary, i just dont add it to the load balancing pool. I also like keeping it around to help me troubleshoot times when the LB is being weird.

I've used both of the solutions you describe in the past, these days I don't mess with the service accounts and just write my powershell to use the secret store https://github.com/PowerShell/SecretStore https://github.com/PowerShell/SecretManagement/tree/master/ExtensionModules/CredManStore

Have you connected to your AIX servers with SNMP already? There are a lot of AIX specific things that people have a hard time getting but process and file system usage should be available.

Have you installed SWQL studio anywhere? Initially i would just hunt around in there, but a couple years ago they added a search to SWQL studio, so a few tries of "it's probably called something like this" would get me there. Looks like it would be orion.volumes if you are just interested in the right now numbers, or…