Q. How does Standby Continuous Replication (SCR) work?

A. SCR is a replication feature available in Exchange 2007 SP1 and above. SCR works similarly to Cluster Continuous Replication (CCR). Closed log files are pulled by the SCR server and played into the SCR copy of the mail database. The key difference between CCR and SCR is that SCR doesn't involve the servers being in a cluster. This means failover to SCR is a manual procedure and requires you to change objects in Active Directory (AD) to point to the new mailbox server. SCR is commonly used for replication to a secondary location, because the SCR box doesn't have to be in the same AD site as the mail server it's replicating from. An illustration of SCR backup is shown here.

A storage group protected by SCR can contain only one database. The storage group can have an unlimited number of SCR replicas, but the best practice is to have no more than four. The source for SCR can be a non-clustered mailbox server, a CCR clustered mailbox server, or an SCC clustered mailbox server. An SCR server can be a target for multiple mailbox servers.

Another difference between CCR and SCR is that there is a delay in the playback of replicated logs into the local copy of the mail database. The delay is for a number of reasons, including handling lossy failovers if the SCR source is a CCR cluster and preventing logical corruptions from being played into the SCR. Logs are delayed by default for 50 transaction logs and 24 hours, whichever comes second, and the database on the SCR won't be seeded until you hit the 50 transaction log and 24 hour mark. You can modify the 24-hour delay with the ReplayLagTime attribute but you can't change the 50 transaction log lag.

See the video "Exchange 2007 High Availability" at ITTV.net for more information.

Related Reading:

Videos:

Audio:

Check out hundreds more useful Q&As like this in John Savill's FAQ for Windows. Also, watch instructional videos made by John at ITTV.net.