WinRM polling - bug in SolarWinds 2025.1

It seems like even if you have default polling to WMI, SolarWinds is still acting like WinRM is the primary methord.

We are still getting alerts saying the primary polling method of WinRM has failed and changed to WMI, previous versions this would imply 'use WinRM' tickbox was left enabled, disabling this would revert to WMI connections.

Only seeing this since upgrading to 2025.1 versions. did notice in the change logs it would now be the default, but looking at nodes, they are all showing 'connection mode: WmiOnly'?

Find more posts tagged with

2025.1

Accepted answers

All comments

mark.gover

I had a strange issue like this on our environment, whilst Solarwinds support couldn't really help me and wanted to fix "performance issues", I discovered using usernames like username@domain instead of domain\username seemed to solve my issue. Create a new credential with this format and test it on a node that is having the WMI problem, if the test passes then clear the alert and see if it comes back. Give this a go and let me know if it works for you too.

darrencook

Looking into this more, it appears its coming from application monitors rather than the nodes. i have went into one template to change defaults to WMI, see if it stops alerts for that version of Windows Server, will report back.

SolarWinds: Why has this been applied globally as a default, rather than a system setting to choose to apply? seems poorly implemented to mass update settings without asking if this is wanted or needed.

mkarak

Hi @darrencook , Thanks for reporting this issue. I can confirm that this is a bug in the current version (2025.1). The value of the Connection Mode dropdown from the Edit Node page (e.g., WMIOnly) is currently only respected for Node polling (e.g., CPU Load) but is ignored for polling of Applications/Components on the given node. This will be fixed in the upcoming Service Release. In the meantime, rather than updating each template, you can disable WinRM for application monitoring using this centralized setting: Disable WinRM polling on the SolarWinds Platform server

crr

Was about to post about some issues we've observed since upgrading to 2025.1 -- but saw your comment here.

We've observed that some WinRM monitored resources such as server reboots, disk capacity, and a few other things are failing to be polled properly. I can test fire the corresponding alerts so it's not an issue with the alerting engine at least. Do you have any more details on the bug you've mentioned? ETA on the fix or any other details? Thanks!