mesverrum · Observability Architect · ✭✭✭✭✭

Simple way to spot check would be to do something like Select count(*) as undp from orion.npm.custompollerassignment That query could be more elaborate but I'm using my phone and can't really test anything more fancy at the moment.

NPM has no visibility into what specifically is using your bandwidth, NTA is the module that tells you who top talkers are. Within NTA there is a feature you can use called IP Address Groups. You can define the public IP address ranges as groups and then enable them to get your data sliced up that way.

${Location} works on my lab, as far as I can tell Atlas will pull any column name listed in the nodes view if you look into the database. -Marc Netterfield Loop1 Systems: SolarWinds Training and Professional Services * LinkedIN: Loop1 Systems * Facebook: Loop1 Systems * Twitter: @Loop1Systems

manage groups > check the box for "Alphabet" > add/remove members From the next screen just set the section on the left to show groups and pick d/e/f from the list to add to the group. Done

Group status is ENTIRELY based on up/down of the members. It does nothing to roll up the alert conditions or any other property, so you need to factor that into anything you expect from them. Like rfroembgen​ pointed out you could use alert logic to force the node down based on whatever you come up with but those might not…

NPM does not have any IPSLA support, but there is a module called VNQM that does. I believe it does not have support for the types of operations you are asking for though. This link shows the types of operations currently supported and the method solarwinds uses to gather their metrics Polling methods VNQM use to poll…

The trick to knowing what variables are available on a details page is to just look that the corresponding SQL database table for that object type. So for a node open the DB and every column on the Nodes table is a variable you can use, for an interface it would be the Interfaces table. That applies to almost all of the…

This is the documentation from the hithub relating to unmanaging objects Unmanaging Entities · solarwinds/OrionSDK Wiki · GitHub

From when I have done varbinds in the past I believe the syntax is more generic, it should be something like this: ${vbData1} Trap variable binding value ${vbName1} Trap variable binding name

I run into this kind of question a lot, there are no built in variables or relationships going from a group alert to the objects inside the groups. You are approaching it from the top down and the UI doesn't really facilitate that, you would have to approach it from the bottom up. For the point of view of Solarwinds a…

I've definitely done what you describe of just having my SAM monitors call local files. As you said its kludgy but does the job. The usual problem I run into here is that server admins write scripts for the purposes of things they need done, and almost never care to think about the format requirements I have in order for…

Looking at those numbers it seems pretty clear that you are not monitoring the same total number of elements between the old and new instances. Node count is just one (and usually the least important) of the things that plays into element counts and polling load. Go to your Settings > Polling Engines and see what you have…

Mobile admin is usually installed to a completely separate system than your NPM, and if you want it to be reachable from people via their phones when they aren't on VPN then yes you would typically put it in a DMZ with a public facing address. Mobile admin is the proxy to NPM.

This is a chunk of scripting I did a while back for a similar purpose, adopted bits from a few other sources to get it there but this script prompts you for your orion server and login creds, then a CSV source file, then for a name and does all the leg work for you of finding all the credentials and adding them to the…

If a device doesn't have snmpv3 configured then Solarwinds would have no way to know if you do or don't have the option to enable it. You most likely could use Solarwinds to build a report of the distinct models and code versions on your hardware but you would have to cross reference that against your vendor's feature…

Did you see that they have a new api polling feature currently in RC stage? Seems like it could be very much in line with what you need. What We're Working On For Server & Application Monitor (Updated July, 2019)

I've run into situations where it says im logging in but just sits there forever, is that the one you have? Dumping your browser cache can fix that issue.

Set up your alert conditions such that vm status =down and host status != down. There may be some cases where the polling doesnt exactly match up and a vm shows as down before Solarwinds realizes its host is down but that should be good enough to quiet the majority of the possible spam. No need to mess the with…

Socket timeout should be able to be tested via the native tcp port monitor for the relevant port, there is also a built in component for https monitoring for your ssl, and the previously mentioned port monitor should also be able cover your port test as well. _ Nagios to Solarwinds transitions are not terribly difficult,…

I want to say that if your server has the same name and IP you can probably just run the scalability installer on the new one and everything will just fall into place. The AWS are a LOT simpler than polling engines. The only thing that I can think of that might bite you is if the license store realizes it's not the same…

Yes, for every process you add to a SAM template you can define thresholds for CPU, Memory, Virtual Mem, and IO loads, i think these default to 80/90 % if you don't touch them but you can adjust them to whatever level is appropriate in your case. There are variables in the alert messages to allow you to pull those metrics…

I don't see why you couldnt just set up alert logic where interface name contains DMVPN tunnel interface and interface status is up and interface total bps < 150000 and interface total bps > 0

_linkfor_ and _iconfor are special functions built into the Custom Query widget, they ONLY work inside that widget and cannot be used anywhere else. After the second _ you put the name of whatever column you want to turn into a link, or an icon and on the backend it sees that and knows to convert whatever you fed it into…

Honestly the default vman and srm views usually satisfy most people's needs. I tend to create a few custom resources to replace some of the out of the box alerts that I prefer to disable (I don't need 900 emails about vms with old tools when I can just run that as a report and deal with them in bulk rather than 1 at a time)

If you unmanage a node it skews availability, because during the unmanaged time we do not poll and so the total number of intervals during the period becomes inconsistent. For example, over a 24 hour period: If the node never goes down outside maintenance, 100% If the node was down for 144 minutes, up the rest of the day,…

Solarwinds is notoriously tight lipped about release schedules and nobody who works for them will give you any kind of timelines for anything until the official announcements. May as well just get your expectations in line with the policy.

The templates you want to use pretty much depend on which software you run on your servers. Almost anything that is out there has a template in the content exchange, some are better than others but a BIG factor is how many licenses you have. If you have a limited number then you tend to need to be pretty ruthless in only…

you have to go to manage nodes, expand out that node, and check the boxes for the phantom child objects you want to delete.

The link you posted mentions that RDS is supported since 2nd quarter of 2018 The tricky part will be determining the correct size to pick, IPAM loads vary dramatically from one client to the next. For example if you have many subnets and actively scan them all at a frequent interval then there is a lot more activity…

SAM will allow you to monitor Linux metrics via SSH with scripts, netbios wouldnt give you anything especially useful from a monitoring perspective and isn't supported by solarwinds.