After 2024.1 -> 2025.0 (+ patch), all my hosts show their metrics in terms of "connections"

[Royce_Simpson]

The following screenshot shows one of my host components, displaying all metrics with the same description, and all alerts are worded in terms of "connections".  All my host components show these metrics in the same way.  Other components, such as databases, ArcGIS Server, etc, all show the correct metric descriptions.  Is there a way to fix this?

[DerekLaw]

Hi @Royce_Simpson,

This is really weird behavior that I haven't seen before. Have you tried clearing the web browser cache and restarting the AM Agent service? To hopefully "reset" Monitor.

Hope this helps

[MichaelOlkin1]

After updating from 2024.1 to 2025.0, I am seeing exactly the same behavior.

Derek - I followed your advice to clear my browser cache & to restart the AM Agent service on one of the hosts, but I'm seeing the same result with this host. Screenshot included here.

[DavidColey]

Could be something with the monitors' postgres sql.  Maybe try re-registering or removing and re-adding a 'Host' machine, see if that helps?

[Royce_Simpson]

Tried that.  Same results.

[DavidColey]

shoulda figured you did....

[DerekLaw]

Hi @Royce_Simpson and @MichL , 

Apologies for the late reply.

Can you both please open a case with Esri Tech Support so they can help take a closer look?

[anasga_81182]

Hi @Royce_Simpson ,
I have also upgraded to the latest version and i am seeing exactly the same. The only workaround i found just now is to unregister the agent and register it again which implies that we need to reconfigure all the components again. I have also noticed that cpu utilized metric for one of my machines was reporting wrong cpu utilization values. after unregistring and reregister the agent the metrics got right units and the cpu is getting correct values.
@DerekLaw , this sounds to be a very serious bug
//Anas

[DerekLaw]

Hi @anasga_81182,

> I have also upgraded to the latest version and i am seeing exactly the same. 

Can you please open an Esri Tech Support case so they can collect information/background on your deployment? We're looking for as much info as possible and steps to reproduce this behavior. 

Getting valid repro steps is the only way we can investigate and resolve. 

> ...  this sounds to be a very serious bug 

Until we can reproduce the behavior, we're unable to confirm this. 

Thanks

[DerekLaw · Accepted solution]

All,

We were able to reproduce and observe this issue at Esri. This is bug with the internal AM database reference for the metrics, which got messed up when upgrading to Monitor 2025.0.

We do have a fix and this will be included as part of the ArcGIS Monitor 2025.1 release, coming in November. 

Hope this helps,