BP question: how do you handle patch tuesday for persistent LC desktops?

[LucD]

Just curious to know if there are any best practices guidelines for handling patch tuesday on persistent linked-clone desktops.

A new snapshot after patching the master ?

Is it better to use quickprep instead of sysprep ? In other words, the same GUID on all stations, or a new GUID after each recompose.

Refresh or recompose ?

Any risks loosing or corrupting the user settings ?

Do you force users off, to avoid stations with outdated patches ? How many days after patch tuesday do you give the users before forcing them off ?

[mittim12]

1:   For linked clones you always have to create a new snapshot to push out updtes to the master.

2:  Quickprep is much faster than sysprep and typically sysprep is used when required by a specific application

3:  recompose would be needed as refresh only sets the VM back to previous snapshot

4:  Unless you have roaming profiles or a user data disk you will probably lose settings when doing a recompose or refresh.

5:   I guess it depends on the severity of the patchs but it's up to the administrator.