Dc restore / hyperv replica

Home Forums Server Operating Systems Windows Server 2012 / 2012 R2 Dc restore / hyperv replica

This topic contains 1 reply, has 2 voices, and was last updated by  Anonymous 4 months, 4 weeks ago.

Viewing 2 posts - 1 through 2 (of 2 total)
  • Author
    Posts

  • MarkJones
    Participant
    #167615

    Hi Everyone.

    Single forest single domain environment with multiple sites with additional DCs, would it be possible to use a hyperv replica of a DC to restore it (Authoritative or Non Authoritative)? Is it the same situation as a DC backup restore?

    Regards
    Mark


    Anonymous
    #372109

    A replica is just what it says, a copy. What replication does is copy the state of the VHDX file from source to destination, not what’s in RAM of the VM in question. The replica VM is registered to the destination hypervisor, same as the original is. Because the virtual hard disk file of the replica is identical to the original (name, IP address, SID, etc.) the replica VM cannot be running on the same network as the original. The failover scenario is that the original dies, so you select the replica to become the operational VM of the same name, and start that one. There’s a time lag of roughly 5 minutes between when the primary VM virtual hard disk is written vs when the same write takes place on the replica, but there has to be some period where the replication happens.

    I have 2 failover clusters in the same domain with hi-speed links between. If the primary cluster becomes non-operational for some reason, I can fail over all the VMs to the replica site and my domain comes back up. The service outage duration will be down to how long it takes you to bring your domain back up. The replica becomes the primary at this point. The simplest recovery at this point is to throw away the original at source (VM definition and all copied VHDK files), then start replication from what became the primary. Once that replication catches up, you can failover to the specimen VM back in the first cluster, make it the primary again, and reverse the replication back to how it was to start with before the failure.

Viewing 2 posts - 1 through 2 (of 2 total)

You must be logged in to reply to this topic.