What’s a Recovery Database and Why Do I Need One?

There will come a time in every Exchange admin’s career where s/he will need to use those backup tapes that they so diligently create weekly. Whether to recover a failed server, bring back a deleted mailbox, or just restore a deleted message that some users remembered that they needed, unless you have Exchange archiving for all your message content, you will quickly need to become familiar with a recovery database. In this article we will go over a few situations where you will need a recovery database, and we shall see what you can (and cannot) do with them.

Recovery databases are required to restore Exchange data from a backup, and if you don’t have a recovery database already configured, then setting up a recovery database is the first step of any restore operation. You can create one (and only one) recovery database on any Exchange Server that holds the mailbox role. Situations where you will need a recovery database include:

  1. When you need to perform dial tone recovery on a failed server,
  2. When you need to recover a deleted mailbox from a tape backup,
  3. When you need to recover one or more items from a backup that a user has since deleted.

Any mailbox server can host a recovery database, and that database does not count against the maximum number of databases a server with an Exchange 2010 Standard edition license can host. It will, of course, need as much space as the data you wish to restore to it, but you can create a recovery database and keep it at the ready should you ever need to restore data.

To create a recovery database, open an Exchange Management Shell, and run the following command, substituting the appropriate paths for your environment:

new-mailbox database –recovery –name databasename –server servername –edbfilepath path –logfolderpath path [enter]

The –recovery switch identifies this as a recovery database. There are several distinctions between regular mailbox databases and recovery databases that you should consider:

  1. A recovery database does not count against the maximum number of databases your license allows (5 for standard).
  2. You can only use a recovery database to restore data. You cannot use it as the source for a backup.
  3. Client connections are blocked.
  4. You cannot create a DAG or make other copies of a recovery database.
  5. Normal maintenance operations are not performed on recovery databases.
  6. Recovery operations restore data, but not ACLs.
  7. You only use recovery databases for mailbox data. You cannot use a recovery database for any public folder related operations.

With a recovery database in place and ready to use, you can perform a mailbox restore request using the Exchange Management Shell. Here’s an example command, where we will restore data to the mailbox used for Exchange archiving:

new-mailboxrestorerequest –sourcedatabase name –sourcestoremailbox “display name”
–targetmailbox emailaddress –targetisarchive [enter]

With the content restored to the Exchange archive folder, the data is immediately available to the user, but won’t force a sync of all that data to the user’s OST. They can drag and drop content to their primary mailbox, which will then sync to their OST only the content that they need.


This guest post was provided by Casper Manes on behalf of GFI Software Ltd. GFI is a leading software developer that provides a single source for network administrators to address their network security, content security and messaging needs. Read more on how to improve your Exchange archiving.

All product and company names herein may be trademarks of their respective owners.