Any way to find any module instances that are Not Collecting?
Summary: Kel L is seeking advice on how to efficiently determine if any modules on their server are set to 'Not Collecting' data, as they found one such instance mistakenly configured. They discovered that checking each module individually using Powershell is time-consuming due to the large number of devices and modules. Kel L is asking if there is a simpler or quicker method to generate a report identifying modules that are not collecting data.
Hi,
I stumbled on a server that had its Collecting toggled off for it Memory Usage. It should have had alerting turned off, but should have still been set to collecting. That got me wondering if I have any other devices where someone had turned COllecting off instead of only turned off Alerting.
I looked around and didn't find any way to do this. I checked with Support and was told I could try an API call. I used the Powershell Modules and found that the only way to this is to get a list of all my devices (a few thousand of them). Then pull a list of all the modules that are associated with each device. Then loop through the list and check the StopMonitoring property for each module and see if it's set to True. Each of these calls takes 1-2 seconds. Each device generally has 80-100 module that Applies To finds. I have thousands devices. Add that up, and my script will take a week or more to run and figure this out.
Does anyone know of a better way to find this information?
I was hoping for a simple Not Collecting report or something that would be quick and easy since I'd think everyone would want to know if they had things Not Collecting any data.
Thanks.
Jan Garaj
·Aug 8, 2026 at 11:02 AM · EditedDon't run calls sequentially if you have thousands of resources to process. Set up a worker pool to execute LM API requests in parallel (100+ concurrent workers), but make sure to include a rate limiter / request throttle. API rate limits (RPM/TPM) will throttle you real fast if you hit them aggressively without backoff logic.
Feed your prompt + the LM API’s Swagger/OpenAPI definition into your favorite AI coding assistant (Cursor, Copilot, ChatGPT, etc.) and let it write the worker queue code in your favorite programming language for you.
Test on 10–20 resources first to find any issues before running on the full set of resources in your LM instance.