We are using MECM version 2107
1. When deploying an application to a device collection, looking at the deployment status the device remains in the “Unknown” status (Client check passed/Active).
-----Looking at the device status, everything seems to be ok, client check result is passed – policy request and hardware / software scan are recent.
-----Forcing an evaluation of application deployments, both from the console or from the client itself, and looking at the AppDiscovery log, the log is not moving / updating or it is updating listing only a small part of the applications that actually the device has in Software Center.
-----The only way to fix this situation is to push an agent reinstallation from the console.
2. Another scenario is with a client that has a Configuration Baseline deployed to it but, after evaluation, it is shown as in Error in the deployment status.
-----The specific error is “rule is in conflict with other rules”.
In this case, the logs are moving and checking them we can see that the rule contained in the baseline (only one deployed) is in conflict with … itself (?) .
-----Also in this case, looking at the device status, everything seems to be ok, client check result is passed – policy request and hardware / software scan are recent and also in this case the only way to fix this situation is to push an agent reinstallation from the console.
-----In fact, after the agent reinstallation, at next baseline evaluation, client will result Compliant without doing / touching anything else.
3. Another situation we faced is with a client having updates deployed to it that are also shown as needed in the reports. In Software Center nothing is shown and in ccmcache nothing is received.
-----From the Support Center Remote Client Tool we can see that this client should have deployment targeted.
-----Also in this third case, looking at the device status, everything seems to be ok, client check result is passed – policy request and hardware / software scan are recent and also in this case the only way to fix this situation is to push an agent reinstallation from the console.
these three anomalies to date, are not always present simultaneously on the same computer. a computer could have even one of these anomalies.
Why are these situations happening even if the devices seems to be healthy from the console?
Is there a way to identify all devices that are in this status? Is the agent reinstallation the only way to fix this?
Is there someone else who is facing these kind of issues?
Thank you for your support.
Regards