So, I'm doing my due diligence and upgrading a test site to Configuration Manager 2012 R2 SP1.
I'm performing all scenarios so I can reduce risk when I upgrade production in the coming days.
The upgrade was succesfull and everything working fine, but I wanted to test my site recovery to ensure I could get back if need be. In production I have opted to backup the database only and so this is how I have restored test:
1. Revert test box (primary site server hosting every major role including DC) to prior to SP1
2. Uninstall Config manager 2012 R2
3. Manually restore DB backup created before upgrade
4. Run the 2012 R2 setup media and choose site restore
5. Choose the option which has the database already restored manually.
Almost everything working bar Hardware Inventory.
I know that the clients (2 clients, one upgraded to new SP1 client, one on CU4) inventory delta is going to be out of wack. In my MP_Hinv.log I see:
Hinv: Timestamp check for class definition policy: DB cached copy = "blah blah blah"
Hinv Retry: cached DB policy timestamp older than report policy timestamp; retry report will be created
Hinv Retry: Created a retry file ....
My questions are:
Site restore not smart enough to handle out of wack inventories?
How should I handle this? A script to delete the inventory version from my clients forcing a full inventory?
Or delete the client object? Probably not the most desirable alternatives.
PS, anyone know if it is okay to run an SP1 client on a CU4 site?
PPS, The client not upgraded still has the scheduled task to perform the client upgrade. Does this task need to be manually removed?