mesverrum · Observability Architect · ✭✭✭✭✭

Is topology available on the IR- side as well, usually printers won't support the necessary OID's to get that to display in Solarwinds and NPM doesn't like it if both sides of a connection don't match up in that way. I've never seen an environment where NPM could 100% map out their topology unless the only devices they had…

Do you have scheduled inventories of your wsus server and all the computers? I find that inventory gaps are the main cause of missing report data

I was able to adapt some of my old queries to do this much of it pretty quickly, obviously mine is not 100% identitical to your request but hopefully you will be able to tailor it as you need. There are a lot of assumptions built in here that may not apply in your case so I want to clarify them in case you need to modify…

Lots of ways to approach this but the thing I think would be simplest is just to add a custom table resource to the view and set it to display the full name of all hardware health sensors that are in warn/ crit/ down status. Make sure to enable the link to the details page for the object. If you want to get more detailed…

above is right, but also those variables aren't correct for a component alert, they are for nodes as the triggering object. if you hit the insert variable button and look for the info you want to include you should be able to see how the syntax differs

If you scroll to the bottom of the screen you showed you will have a button that says something like "Not finding your OID?" and it lets you set up a custom OID. In this case you will want to set up a custom OID for 1.3.6.1.4.2.1.47.1.1.1.1.13.1 which will pull the first row of the table. You may have to experiment with…

In my case I ended up rebooting the server and it started working again afterward. Not sure if I had done something previously that helped but when in doubt, give it a reboot?

When you only have a hammer all problems look like nails. What I'm trying to say is that the reason it seems difficult for you is because you are trying to use solarwinds to handle scheduled tasks and that's not really what the monitoring tool is intended to do. Linux has natively built cron functions that could do this…

By performing a Create on the node table you are kind of just putting the shell there. All the polling data needs to be created on the pollers table as well. Check this example and notice that it has two phases, initially creating the node, then adding all the pollers OrionSDK/CRUD.AddNode.ps1 at master ·…

I had to rework a few of my own sql based charts as well, you are correct that the problem is not that the query broke it's that they made some changes to how the custom chart resource works and the sticking point seems to be those two data grouping options that used to be optional and are now required. After mucking with…

I think it would primarily boil down to the specifics of how that network will be built. If the in and out segments can both talk to each other then just put the poller closer to where the majority of the devices live (probably in fabric if this is a network centric instance). If the networks can't easily route then you…

Gone, unless you have older backups of your database and can recover it from one of those.

Solarwinds only reports what it gets via SNMP for those interface stats, so when troubleshooting erratic reporting from Cisco hardware I always take snmp walks of the hardware in question. I personally have experienced 5 different situations where specific models and firmwares will show completely different values in SNMP…

If I remember correctly the alert objects table is going to be historical, as in objects only get an entry there if they have already had each alert. This won't help you with listing out all the alerts that an object could fire but havent yet. Ive danced with this problem a few times and there doesn't seem to be a handy…

No, in a standard scenario the APE goes down and all data collection stops for the nodes assigned to it. There is no built in function to roll nodes to another polling engine, but in concept it could be made to happen with a script and the API launched from an alert action.

Try this for a custom report Data source is Windows Nodes Add the columns you need, but set the install date to use Max aggregation

I was never on a project where we did it, but I heard through the grapevine that the last time one of my coworkers tested it out there were problems getting acceptable performance from cloud db offerings for the SQL side. This was a while back and I know I've seen announcements from the big cloud providers touting that…

Have you installed SWQL studio in your environment yet? If you intend to get anything done in SWQL you won't get far without it, its basically a prerequisite. The tables in SQL have almost identical names and layouts to the ones in SWQL so it shouldn't take much work at all if you load Studio up and have a look. Also worth…

Taking those questions one step at a time, So the question is, is it possible to assign nodes to a group based on actual assigned polling engine? The groups dynamic query UI does not have an option in their pull down for engineid, but what you can do is set up a node alert that is always true (example nodeid is not null)…

There's an export button at the top of the main Log Viewer page, i don't have any data in my lab so I'm not entirely sure about the format it pumps it out in, but I'd start there. 

First, make sure that your routers/firewalls and various other bottlenecks are sending netflow where it is available. Then just go to the netflow summary page, look under the sources resource, expand the node that you are interested in checking, click the link to interface you are interested in and you should mostly be…

So it sounds like you may have turned on an alert to notify you regarding any changes to services? Tracing an alert event back to the rule that triggered your message is kind of a beast in LEM, you might be able to figure it out by looking at the internal lem events filter on the monitor tab to see when the alert fired and…

What you are showing doesn't look normal to me. Can you check your scheduled jobs list and see if there is duplicate entries in there? The way it should work is that each node completes the NCM jobs based on the poller that it is assigned to, one job should still cover all pollers.

So this query is pretty likely to time out just do to the sheer amount of data it would have to crunch unless you environment is very small. The interface in/out total bytes value is the total byes since the last time we polled. So to get to your situation we are going to have to add them all up for each interface from the…

One extra server for each system you want to protect with HA, so 5 new standby servers for everything you just described.

you can see exactly what it does by reviewing the logs for it, referenced in this kb Troubleshoot Orion database maintenance issues found by Active Diagnostics - SolarWinds Worldwide, LLC. Help and Support In short, it takes all the data that has aged out of the retention periods and rolls them up into the next larger…

In order to populate that data Solarwinds needs you to be monitoring those interfaces on both the server side and the switch side, and both devices have to support some specific SNMP OID's. Depending on the devices involved it can be pretty hit or miss getting those relationships to automatically populate. Topology related…

Does your query have any kind of order by at the end? the Custom Query widget gets mad when you use certain functions and don't tell it how to sort them. -Marc Netterfield Loop1 Systems: SolarWinds Training and Professional Services * LinkedIN: Loop1 Systems * Facebook: Loop1 Systems * Twitter: @Loop1Systems

yeah, if your report includes the orion.nodes table then add this to your selection nodes.detailsurl as [_linkfor_Caption] That linkfor function is case sensitive and if you have chosen a different name for the caption field then you need to use the same one on the linkfor Same sort of method can be used for any table with…

So the min/avg/max values are actually all the same on data as far back as you have for your "detailed" interface retention in the polling settings, defaults to 7 days. They only start to spread out once your data is getting condensed to the hourly/daily values. Where things can get a bit more complicated with the charts…