You can restore the indexes from last good backup then do an update on them to populate the missing items from the index

Once your Centera's are good and your databases are good then rebuilding the index is an easy task but the preferred is the restore then update.

This does work I have done it sucessfully

I saw a couple references to "Updating" but couldn't find a method.  Is it like the Repair option that only comes up if there's a repair log for the user?

I see it now.  Joys of using this product since 4.0, they move shit' around all the time...  However, that's find for a couple users, any known method for Updating, lets say, 5,000 users on a server?  or is it CTS for who ever has to click on every Vault Archive?

For anyone else like me who's never found it before, it's on the Index Volumes tab of the Exchange Mailbox Archive, and right click on the line item.

I wouldn't recommend doing an Update on 5000 users at the same time, bear in mind that Storage is invoked to crawl the .DVS (and associated files if EV8+) and re-index the metadata etc.

As Scanner001 mentioned, updates are the preferable method as it's less work for the processes to do (but that does depend on how old your backups are!)

So that would be a No then... 

a) of course the storage crawler is going to have to pull that data back

b) there are only so many threads that would be running

c) only so many indexing threads that would be running

d) they would queue, and not actually index all 5000 at once.

At the end of the day I don't want to have to advise a client to manually click on 5000 Archives to update them, that's insane.  I'm not worried about performance in this clients case, we've been able to archive 150,000 items per hour per server, I'm sure it can hand the crawling and reindexing of archived data, which is a lot less work the archiving it in the first place...

you dont necessarily even have to force the update to occur, any time a user searches their archive, an update is performed, if you archive an item to an online index, an update will be performed.

Most likely what will happen in a DR scenario if you are *only* restoring the indexes (and not messing about with the databases) is that the indexes will immediately go to failed state, after which EV will automatically bring online and update each failed index

The real issues would be restoring indexes without the databases when a roll over has occured or new archives created, that would cause some extra headaches

Was wondering where you went off to.  It must be nice to have access to the "inner" Vault.  I don't suppose this is documented anywhere externally accessible to present to the client is there?

take a look at this mate, see if it helps any
https://www-secure.symantec.com/connect/articles/how-long-will-my-index-take-rebuild