mesverrum · Observability Architect · ✭✭✭✭✭

I think there probably needs to be some clarification. Specifically what kind of change are you looking to identify? You mention a config change report. Are you talking about NCM? The way NCM does config changes it doesn't have any way to know who made a change. Most network devices don't record things like "user abcd…

The values exist in Orion because they create min/max values for historical data, so after data ages out of the detailed polling period (7 days default) it will have min/max/avg values calculated for hourly intervals and then daily intervals.

There is no native capability to do what you are asking. Where would you be expecting to see the message on the web console? With SAM there are many places you could rig up a script to show up an output, linux nodes have the option to extend their SNMP to allow you to gather custom data, WMI based nodes can return numeric…

SolarWinds Network Performance Monitor Training: Create a Manual Discovery Job - YouTube Sounds like you may want to take a spin through the Solarwinds youtube channel or look through the training section in the Success Center for each module you own. Network Performance Monitor (NPM) Training - SolarWinds Worldwide, LLC.…

This is the SQL i use to display parents and child relationships for nodes as part of my down nodes messages, just add it to your email and test it against some of those VG nodes to see what you get.. It might work for your use case as-is, but you might need to tweak it It displays the name and description of the…

Richard, I looked at this and realized it would be more useful if it included a direct link to editing the node, so I've updated my original post with a segment of code to streamline that.

I don't use OOTB reports, for reasons just like that. Its probably not useful data. To be frank I always put a big * near any availability number I see on Orion because there are dozens of ways that those numbers could be inaccurate or misleading. If you didn't create the report then you don't know what assumptions are…

Strange, a way you can debug which widget is causing the pain is to add another tab to the view and move the widgets to the second tab one by one until your page stops generating those queries.

First, did you test them and make sure they returned values from your UPS? If that was good, did you check the box that says keep historical data (can't chart it out of we aren't logging it over time). Did you take those UNDP's and actually assign them to the UPS? Hit poll now and wait a few minutes after assigning them,…

Unfortunately there is no option to dynamically add groups to other groups, I recall reading a post a long time ago that the devs were concerned about the possibility of having groups recursively adding to each other and causing performance issues. As it is, having a large number of dynamic groups can pretty much cripple…

Just a guess, but maybe the localhost address isn't being allowed to query snmp? Can you try to launch an snmpwalk from the server to itself? If thats not the problem then I'd be looking for log files relating to the agents, either on the polling engine or locally on the node itself.

This should get you in the correct ball park Configure the managed computer inventory task - SolarWinds Worldwide, LLC. Help and Support One of the things I had struggled with in PM was realizing there were so many different types of jobs that you set up and it was easy for them to slip through the cracks. It is a pain but…

When you say it is not getting fetched, is there an error message, does the report time out?

What that exe is doing is launching a poll against your node to collect the top processes and then it adds a note to the alert that lists them out. You have to leave enough time between launching that process and escalating to send the message for that whole thing to complete and you need to include the alert notes…

Wilson99, There is an auto-dependencies feature, you enable it from the web console under Settings > Polling Settings > Enable Auto Dependencies. In order for the auto dependencies to work Solarwinds has to be able to read the topology tables from whatever switches/routers etc you have which can be hit or miss depending on…

Do you monitor the switches using SNMP v3? Lots of the requires OID's are not normally available in v3 by default.

From my vantage point working with lots of large clients all over the country, at this stage in the game nearly all sysadmin tasks for linux servers should be running through some kind of centralized server management scheme. I see a lot of Ansible out there running in the wild these days. cron – Manage cron.d and crontab…

Since I wrote this I have done a bunch more alerts using custom SQL and I had switched to a model where I just always build the logic in such a way that it always gets at least 1 row, and if its not the row I NEEDED then I just isnull the value, so in this case I would do something like SELECT top 1…

Which is exactly why SW doesn't provide a list, these SID's are completely under the control of the vendors who you are getting data from. It would be impractical for someone at SW to try to stay ahead of all the possible event id's that come out from Cisco/Microsoft/Barracuda/Symantec/etc.

All NPM statistics are subject to the same rollups, other modules sometimes have their own retention settings but they basically all fall under the same model Node response time, loss, cpu, memory are all rolled up using the following process. details - raw data, weight is 1, min/max/avg will all be the same value hourly -…

just add to the where condition and ac.category='NOC'

This is the powershell script i use on my server to sync up nodes to the engines used on their discoveries. <#------------- CONNECT TO SWIS -------------#> # load the snappin if it's not already loaded (step 1) if (!(Get-PSSnapin | Where-Object { $_.Name -eq "SwisSnapin" })) { Add-PSSnapin "SwisSnapin" } #define target…

So this maybe is harder than it has to be. If the parent interface is not a group of interfaces, you could skip creating the group for it and just use the interface directly as the parent of the child groups. Parent and child objects can be any orion object, no reason to make groups with only 1 member. For the rest I would…

No, the active/standby relationship requires a 1 to 1 pair

In Sam you can do something similar by setting up a Linux script monitor that would gather the output you are looking for. You may have to tweak it a bit since Sam script requires an integer result but you can just do something where if you get the desired output the script writes a 1, and a failure gets a 0 or something.

^ That report above would show you, but only if the alert was created within the last 30 days of the event retention period

I also use both of these techniques with clients, i show everyone the alert engine method in case they aren't super strong in powershell, and then if they have skills with powershell I show them how to use the API to set properties. Another option I've used to scale up the alert method is to build one out by hand, then…

I see what you are getting at, and I cannot think of a simple way to make that happen off the top of my head. The db doesnt treat the config files as their own objects that can be tagged as much as they are an accessory to the nodes. -Marc Netterfield Loop1 Systems: SolarWinds Training and Professional Services * LinkedIN:…

New link to same article https://support.solarwinds.com/SuccessCenter/s/article/Monitor-alert-and-restart-a-Windows-service-from-SAM

Npm only, ncm and sam and nta have their own tests now IIRC and the architecture exam crosses multiple modules but it's really just focused on thre stuff that's in the scalability guide.