Hello
Can some one please clarify the following for me as I have read different information on the same question and am therefore not sure of the correct answer.
Windows 2003 R2 DFS
I am going to add a new member to the an existing DFS share (i.e. new namespace server and new replica member)
Therefore I take a NT Backup of the data on the existing member, I then restore this data to the relevant folder on what is to become the new member (i.e. pre-seeding), the idea being not all the data has to be synced across the WAN.
Now I have read the following two accounts of what happens when I add the new member
1: Any preexisting data in the folder (i.e. the data I pre-seeded earlier) will be moved the the PreExisting folder (thereby leaving the shared folder temporarily empty)
2: DFS will then compare the hashes of the files on the existing member to the files hashed of the files in the preexisting folder and if they are the same it will move the files back to the normal DFS shared folder (i.e. where it was originally moved them out from) possibly via the staging folder
The above makes sense to me (not sure if the data goes back via the staging folder though?)
I have also read the following
1b: As above
2b: DFS never touches these files again (files in the preexistence folder) i.e. leaves them in preexisting and does not move them back. Rather DFS will get a copy of all files from the existing member as it is considered "Authoritative" (i.e. primary member) until the full initial sync has been completed from the primary member to this new member (at which point the existing member is not longer authoritative for this particular relationship)
if 2b above is correct? this would make a nonsense of the per-seeding work,
Can someone please tell me which (if any) of the above behaviors are correct please, when adding a new member with pre-seeded data from a current member
Thanks all
JoB333
x1!