Forum Discussion

Naveenkumar's avatar
Naveenkumar
Icon for Neophyte rankNeophyte
2 months ago

Issue in auto refreshing EC2 properties

I’m encountering an issue with my Windows ASG group EC2 instances. Whenever a new instance is added, certain properties from a dynamic group—most notably “wmi.user” and “wmi.pass”—are applied. However, by the time the instance is registered in LogicMonitor, WMI isn’t immediately available because some automations are still configuring the WMI credentials on the host.

A few minutes later, WMI starts working on the host, and I can successfully test it using wbemtest from the local collector. However, the LogicMonitor portal still shows that WMI is not responding. Interestingly, when I use the collector’s debug console and explicitly specify the WMI credentials, it pulls the information successfully. But if I don’t specify the credentials manually, it fails to work.

The only way to resolve this is by manually running “refresh properties,” after which WMI starts working without making any changes.

I’m trying to figure out if there’s a way to automatically force a properties refresh every 15 minutes to ensure everything works as expected without manual intervention.

  • Hi Mike,

    1. Even after 12 hours the alert on Win_WMI_Access_Denied_ErrorCodes check on port 135 is not cleared after it started working.
    2. Using Wbemtest from collectors to make sure wmi can be accessible from the collector.
    3. Also using the debug console with explicitly mentioning the wmi username and password can get the metrics but if I run the debug command without mentioning creds it gives access denied error.
    4. After that I'm clicking on refresh properties button as mentioned in the screenshot.
    5. If I reload the device page all the missing datasources are applied, Alert is cleared, metrics started to poll.  

     

    • Mike_Moniz's avatar
      Mike_Moniz
      Icon for Professor rankProfessor

      Hmm, I didn't even realize there was a "Refresh Properties" option, never used it before :) I generally use AutoDiscover to force update when I'm in a hurry. Sound like it would be equivant in this case.

      Might need to ask support but my guess is that LM does it first discovery, fails to detect anything and you may need to wait 24+ hours for it to run again, without forcing it manually. As a workaround I would suggest attempting to delay having LM detect the new EC2 instance until it's fully ready for monitoring. For example you can setup a Tag in EC2 that is something like monitoring=true only after your configurations are complete, and setup LM to only auto-add those devices that have that tag set.

  • LogicMonitor portal still shows that WMI is not responding

    How are you determining WMI is not responding? Is it from the Win_WMI_Access_Denied_ErrorCodes check on port 135? That check runs every 10 minutes checking 1 extra time, so alerts on it should clear after 20 minutes of WMI working. But the "135" entry in the resource tree will only update once a day so it may take a day before the entry itself goes away. Forcing an AutoDiscover will speed this up. But this is just a visual thing and it's the alert itself that helps notify you that WMI is not working.

    Or are you looking at another way to determine WMI is not working?

    Also what exactly are you doing to "refresh properties"? Re-saving the device settings? Running AutoDiscover?