fop · 3rd Level Techromancer · ✭✭✭✭✭

Hi Kevin, nice to be visited by an Actual SolarWinds Person :) The reason I created this request was that I was investigating non-standard SolarWinds API pollers, in this case the Azure Application Gateway. Using the MS docs and an initial script for authentication from someone here I came up with a script that worked…

I did try getting SAM to execute pwsh but not in the way you're doing it there. From a PS5.1 window my script worked fine, but I couldn't get it to play nice in SAM then ran out of time.

When we took on a load of Dell servers for HW monitoring I was amazed to find that they just don't exist in the Orion world. As a purely HPE house previously we had the luxury of iLO integration which is an excellent thing.

We have had someone desperate to back up their workstation because their monitor was on fire. That was a fun call.

Congrats everyone, and greetings all. May the next 12 months be very Solarwindy.

I was talking about this in a meeting just yesterday, much more useful than a page of numbers.

I came here to ask a similar question, in my case 'are the Dashboard database entities available to a widget', so I could use something like SELECT n.caption as [Nodename], n.IPAddress as [IP address] FROM orion.nodes n WHERE n.customproperties.customer_code like '%${DASHBOARD_TITLE}%' which would mean the dashboard title…

I like the idea of a Cache Blessing Ritual, maybe sacrifice some Thwack points to guarantee that the cache improvements it recommends are watertight.

Bah, the 'tell us' option on today's question didn't bring up a text box, so the answer is: 'A rogue DBA from the TLA department causing a MIM because their IDX was thrown out by a CEO claiming it was SEP.'

...or not. I've discovered that the issue is the 'apma.Components.MultipleStatisticData.NumericData' line so now I need to work out why that's seen as a Bad Thing. I get the same fault if I link into apma.Components.CurrentStatistics.ComponentStatisticData as well. I can get the same data if I go straight to…

Aaah wait, you can expand the fault box AND copy the error to the clipboard, this may help.

We keep OME separate to Solarwinds, and don't do the HP stuff, but for xClarity SNMP is available via the 1.3.6.1.4.1.19046.11.1.1 MIB. I have monitors for the fans in xClarity-monitored servers. Set up pollers in the Solarwinds UnDP Poller config and you can alert on the polled values. Check the documentation here.

Try this. If I was doing it I'd add the 'is not unmanaged' option in line 3 of the trigger scope. I don't have easy access to SWQL Studio right now to verify it but it looks right logically.

Your trigger needs to be an AND block of the 'node' properties - node name, polling status (though I'd change 'status' to equal 'up'), then the 'what fires the alert' block is an AND block of your component values. IF (node name is SERVER001 AND server is up) THEN IF (monitored component is Spooler AND its status is not…

I did think about changing it in the DB, but also thought that might have ramifications. Dammit. Pleased you got sorted though.

As someone who went through exactly this exercise a few weeks ago thanks to a typo on my part when I first created the groups for our new poller I can say that unfortunately you need to create new Windows Groups with the new name once you've set it in AD. What would be awesome would be if you could duplicate&edit groups…

First thing I'd do is open a powershell session and 'test-netconnection -p 17778 <poller>' to see if the server can contact the poller. All the logs for the agent are in C:\ProgramData\SolarWinds\Logs\Agent, so if there's any errors to be found they'll be in there.

Yep - AppInsight for SQL requirements and permissions

We have this issue on 2023.3 too, and I *think* it's because SW will attempt a logon with the machine account before trying the account it was configured for when you ran the Config Wizard. Like most folk (I think) we use a single windows account for SQL access from all pollers and web servers, but we still get 18456…

Thanks for the reply. We were waiting to actually get a RRAS server to explore but that's not moving as quickly as we thought. Meanwhile, someone came up with this list of monitorable attributes which I suspect is pulled straight from the Microsoft Documentation. Currently I only have an alert for the RRAS service itself.…

The thing that tripped me up back in the beginning was that for OAUTH2 you need the tenant (directory) ID to get an access token. It also requires which backend process you're using for the data, eg a plain Microsoft login for a service endpoint or a Microsoft Graph. 90% of ours are plain logins against the endpoint we…

This is quite a vague question because we don't know what setups you have for authorisation in the API call, or what you're looking at monitoring. I use APIs for Azure polling and this might be different to what your aim is. I'll say one thing though, the biggest pain and the most poorly documented is OAUTH2 authorisation…

What's the end point here? Are these new servers that you need to onboard for monitoring?

Ah, this is the Macros I've been learning about this week. They're context based and I haven't worked out how the context works yet, but I assume it's a subset of the query in the dashboard SWQL. Your usage of [dbo] there makes me think you're doing a SQL query and not a SWQL one? Or just listing the field names in the…

Your SWQL query is wrong since you won't have a table called OSG_Company in the database. If you post the code we'll be able to track the problem down.

So you want the alert to fire at say, 1pm, but you want that alert to immediately go away and not fire again until 1pm the next day even if the trigger condition matches?

In the alert definition, go to the 'reset condition' tab and select 'reset this alert automatically after', then give it the number of hours needed, probably 23 in your case. This will keep the alert active and won't retrigger until that time period has expired.

I can see the top line in that script that says 'Arguments example', so is the script in your component monitor expecting the arguments and starts with $logfile_path = $args[0] $regex = $args[1] instead of what you're showing us from the ISE where those variables are populated manually? Can you send a screenshot of the…

No problem :) 

Go to customersupport.solarwinds.com and log in, then go to this menu: The Environment Reviews page explains the process, but handily gives the wrong location for the Diagnostics collection tool, on my Poller(s) it's in Start -> Solarwinds Platform -> Solarwinds Diagnostics. Run that and click 'start and save diagnostics'…