Specifications

14
• In iSCSI environments ESX hosts at the DR site must have established iSCSI sessions to the DR site
vFiler prior to executing SRM DR test or failover. The NetApp adapter does not start the iSCSI service
in the vFiler or establish an iSCSI connection to the iSCSI target.
• In iSCSI environments the DR site vFiler must have an igroup of type "vmware" prior to executing SRM
DR test or failover. The NetApp adapter does not create the “vmware” type igroup in the DR site vFiler.
• In a NFS environment VMkernel port network connectivity to the DR vFiler NFS IP addresses must
already exist prior to executing SRM DR test or failover. Ensure that network communication is
possible between the VMkernel ports on the ESX hosts and the network interfaces on the NetApp
controller, which provide NFS access. For example, ensure appropriate VLANs have been configured
if using VLAN tagging.
• A single VM must not have data on multiple vFilers. The storage layout rules described in section 4
apply equally to both physical NetApp SRM arrays and vFiler SRM arrays.
5.7 GENERAL ENVIRONMENT REQUIREMENTS
1. vCenter installed at primary and DR sites. VMware SRM requires two independent ESX environments,
each managed by its own vCenter server.
2. vSphere client available at primary and DR sites.
3. Site Recovery Manager installed at primary and DR sites.
4. NetApp site recovery adapter installed on the SRM servers at the primary and DR sites.
Note: See the preceding table regarding version dependencies between SRM and the SRA.
5. NetApp FlexVol volumes and FlexClone.
6. LUNs or NFS exports configured as datastores and connected at the primary site.
7. A small LUN or NFS export configured as a datastore at the DR site for storage for placeholder virtual
machines.
8. For FC or iSCSI implementations NetApp igroups must pre-exist on the recovery array (physical array
or vFiler) and contain the necessary ESX host initiator names. SRM will automatically map LUNs to
igroups during failover or DR testing, but igroups are not automatically created. The igroups at the DR
site may contain local LUNs for locally running VMs, but LUNs replicated from the primary site should
not be preadded to the igroups at the DR site, SRM adds these LUNs automatically during DR test or
failover.
9. For NAS implementations the following configuration rules must be followed:
a. Exports must be in the /etc/exports file. Manual exports made with the CLI without the –p option will
not be discovered by SRM.
b. Exports must use values in the RW (read/write) security field in the /etc/exports file. Exports using
the default setting of rw to all hosts will not be discovered by SRM.
Example of a /etc/export line that would be discovered by SRM:
/vol/srm5 -rw=192.168.2.0/24,root=192.168.2.0/24
Example of a /etc/export line that would not be discovered by SRM:
/vol/srm5 -rw,anon=0
10. Datastores must contain virtual machines, or virtual disks owned by virtual machines, in the vCenter
inventory to be discovered by SRM.
11. The environment used in this technical report was deployed using the best practices documented in TR-
3428: NetApp and VMware Virtual Infrastructure 3 Storage Best Practices:
www.netapp.com/us/library/technical-reports/tr-3428.html or TR-3749: NetApp and VMware
vSphere Storage Best Practices: www.netapp.com/us/library/technical-reports/tr-3749.html.
12. SnapMirror must be configured and replicating the volumes containing datastores to be protected. SRM
does not perform SnapMirror updates or baseline transfers. SnapMirror transfers must be managed and