When a new vRanger Virtual Appliance (VA) is deployed, and the VA(s) are used for vRanger Replication functionality, the secondary "scratch" disk on the VAs should be preserved for use by the new version, since it contains the .vzmap block-map files for the replicant VM, and other important files. If a new, blank scratch disk is deployed, vRanger will have to recreate all this information, which can be very time-consuming, and not desirable.
JUST TO BE SAFE, if possible, copy the scratch disk to another location if you have room to do it (Not a requirement just a precaution).
1. Remove (detach) the scratch disk from the old VA via the vSphere client ("Edit Settings" page for the VA VM), but do NOT delete it from disk.Make a note of its path.
2. Delete the old VA from vRanger, but DO NOT check the box to remove from disk (this will remove the VA from the database, but preserve the actual files on the datastore ... This will allow you to deploy the new one.)
3. Deploy the new VA (either manually or through the vRanger Deployment Wizard), and when you add the second disk (check the box to use the VA for replication and create a DIFFERENT scratch disk at a different path of the SAME SIZE as the original (old scratch disk). This is important, and it's only temporary.)
4. Once the new VA is deployed with its "dummy" scratch disk, via the vSphere client, delete the NEW scratch disk from disk, and simply add a second disk to the new VA by pointing the disk to the EXISTING .vmdk of the OLD scratch disk.