Standby Continuous Replication (SCR)

Great blog on 'You Had Me At EHLO' released yesterday on SCR from Scott Schnoll - 'Standby Continuous Replication in Exchange Server 2007 Service Pack 1'.

Highlights for me include the fact that you can have multiple targets from a single source - unlike LCR and CCR (although it is recommended that you limit this to 4 targets per source); and the fact that you can delay replay activity so that there is time to prevent a logical corruption reaching the SCR target.  Presumably therefore you can set multiple targets with different replay delays?  ..might give administrators some interesting new options.

One of the major differences with SCR is the fact that the target can be in a different AD site which can be either a standalone mailbox server or another CCR node.  The thing to note here is that recovering to the standalone mailbox server must be using database portability and so can be a production Exchange Server for other storage groups (although none of these other storage groups should be LCR enabled).  When recovering to a CCR cluster node the CCR cluster is effectively just a standby cluster.  The MNS cluster can be setup and ready to go although there cannot be a cms configured.  This will be created during the recovery process.  I like the idea that you can recover a CCR cluster to a new CCR cluster in another data centre and AD site.  So following the loss of a data centre you can quickly recover full functionality and 'local' resilience.

One thing that worries me slightly is that if the SCR target is in another site then what happens to the transport dumpster as this is invoked according to site membership in a CCR configuration?  If not then this may have implications for how some companies approach high availability in their chosen design.  Look forward to getting some more detailed information on SCR...  Hopefully Scott will write more on his blog too...

Will write more myself on SCR and High Availability soon...