Exchange Server

Managing the Replication Source for Exchange Server 2010 Mailbox Database Reseed

Avatar photo
Post author:Written By 4 Comments

In some Exchange Server 2010 environments the administrator may encounter a situation in which mailbox databases need to be reseeded from a single server in one site to multiple servers in another site.

Consider a scenario where the mailbox database is active in a site with a single mailbox server, and has passive copies in a site with two mailbox servers. The two mailbox servers hosting passive copies need reseeding due to some fault that has occurred.

Exchange Server 2010 multi-site DAG
Exchange Server 2010 multi-site DAG

If the reseed was commenced for both servers it would effectively mean the data was replicated across the WAN twice, which may not be ideal depending on the available bandwidth.

Exchange 2010 database reseed from active copy
Exchange 2010 database reseed from active copy

This is because by default the database copy update operation uses the active copy of the database as the source. This would be the case if you ran the Update-MailboxDatabaseCopy cmdlet with only the mandatory parameters.

However if you use the console wizard to perform the operation you’ll notice an option to choose the source for the reseed.

Choosing a database reseed source in Exchange 2010
Choosing a database reseed source in Exchange 2010

This allows you to perform the reseed more efficiently in terms of network bandwidth by reseeding one server first, and then using that passive copy as the source for the second reseed operation.

Exchange 2010 DAG reseed from passive copy
Exchange 2010 DAG reseed from passive copy

This can also be performed using Update-MailboxDatabaseCopy with the -SourceServer parameter, for example:

Update-MailboxDatabaseCopy MB-HO-01br-ex2010-mb -SourceServer ho-ex2010-mb2

An alternative would be to reseed one server first, then move the active mailbox database copy to that server before commencing the second database reseed in that site.

Tags: , ,

About the Author

Paul Cunningham

Paul is a former Microsoft MVP for Office Apps and Services. He works as a consultant, writer, and trainer specializing in Office 365 and Exchange Server. Paul no longer writes for Practical365.com.

Comments

  1. Adam Thorburn 8 Jul 2014 Reply

    Hi Paul, Thanks for the response.

    Sorry, I missed the word “without” in the line below. Hopefully you understood where I was going with the topic.

    While the first database is in a index crawling state can I start a re seed on the second database without affecting the index rebuild on the first database ?

  2. Adam Thorburn 8 Jul 2014 Reply

    Hi Paul. I have a 2 Node DAG with 4 mailbox databases. I need to perform a database re seed on the passive databases due to a storage system update. When I perform the re seed on the first database and it completes there is some time for the index to complete the rebuild. While the first database is in a index crawling state can I start a re seed on the second database with affecting the index rebuild on the first database ?

Leave a Reply