VMware SRM: VMs recover with Offline RMD Volumes.

Jeff Miller sent over this handy write up involving VMware’s Site Recovery Manager and Windows 2008 R2. 

Windows Server 2008 (enterprise and datacenter) has introduced a new default disk policy that causes a long delay to boot a server while using SRM to bring servers online.  The SAN policy determines whether a newly discovered disk is brought online or remains offline and whether it is made read/write or remains read-only.  The default setting of forcing all SAN disks to remain offline the first time a disk is discovered can cause applications like Exchange and SQL to take a very long time to fail, which in turn will cause the server a very long time to get to a login prompt.

In order to reduce the time it takes to bring servers online, change this setting from the default of offline to online on your servers.  RecoverPoint will replicate these changes to the DR site and now SRM is able to bring the server online quicker and get applications online quicker.  Prior to this small change, the servers would take a very long time to come online because services such as Exchange and SQL would take a very long time to timeout before we were able to login to the server, online the disk and reboot again.

Here are the steps to check to see what your current setting is:

open up a command prompt

type diskpart

type san

if it is currently says SAN Policy : Offline Shared

then you will need to type the following to resolve this

SAN POLICY=OnlineAll

You may also want to add this into your unattended build or into your VMware templates to ensure that you don’t run into this again in the future.

Editors Note: If you are running SRM, you’ll want to take Jeff’s advice and pop this policy into your templates for good measure.