White Papers
Table Of Contents
- 1 Introduction
- 2 VMware SRM terminology
- 3 Overview and prerequisites
- 4 Configuring array based replication
- 5 Installation and configuration of VMware SRM
- 6 SRM protection groups
- 7 Recovery plans
- 8 Testing
- 9 Recovery
- 10 Failback
- 11 Considerations for guest iSCSI connected volumes
- 12 Summary
- A Technical support and resources

38 Disaster Recovery with Dell PS Series SANs and VMware vSphere Site Recovery Manager | TR1073
11 Considerations for guest iSCSI connected volumes
There are many benefits to utilizing the native iSCSI initiator from inside the VM to connect to the storage
array. These can include, but are not limited to, data isolation, VSS integration, physical to virtual clustering
and snapshots.
One of the difficulties of combining this with SRM is that during a test the VMs that are brought up on the DR
site are isolated in a test network bubble. This is for the safety of the production VM and production data.
Because VMware encapsulates the VM into a file, there is no difference between booting up a VM on the
protected site and bringing it up on the recovery site. This could lead to potential issues not only with
duplicate name and IP addresses, but the server will try to connect to the same iSCSI volumes on the
protected SAN that the production VM has access to. Because of this, advanced techniques and additional
steps need to be taken when utilizing guest attached volumes. The same process can be used but in a
manual fashion. When bringing up a VM that has guest attached volumes, a clone can be created of the
replica just like SRM does. Bringing this clone online and attaching it to a separate test server can validate the
data on the replica while SRM validates the VM is configured properly. During a failover there is no isolated
network, so promoting the replicas of the guest attached volumes and then attaching them to the VM will
work.