This guest post is by Bryan Krausen who blogs at ITDiversified, where you can find his back catalogue of posts. Find out more about the guest blogger program here.
As mentioned in this article, there are many variables to take into consideration when deploying and configuring VMware Site Recovery Manager. When failing (or testing) your virtual machines from site to site, decisions must be made on what cluster will they start on, what priority will they be failed over, and what, if any, scripts will be executed after fail over has occurred. An equally important step is ensuring that the VMs can communicate on the correct network at the surviving site. VMware SRM has multiple ways of handling customization of a VM’s IP; one that is “automated” using IP Customization Rules and another that is completely manual, set configurations per VM. Both are explained and shown below.
Note: Site Recovery Manager supports customization of the same guest operation systems which vSphere 5.5U2 supports. Check the document here for a support matrix.
To set the stage, our deployment looks something like this. We’ve created a “Failover VLAN” at each site for VMs to utilize when failing over to the opposite site and will be using IP Customization Rules to specify what IP address VMs are assigned at the opposite site.
IP Customization Rules: IP Customization rules are configured at the Networking Mappings tab and is the easiest way to configure guest customization for your VMs. This is especially true if you’ve chosen to utilize a dedicated subnet and you don’t have many (or any) custom requirements such as DMZs.
An IP Customization Rule applies to VMs failing over from protected site to recovery site and, using our example above, will reconfigure NICs from an IP address on vlan 200 at the protected site to an IP on vlan 210 at the recovery site. SRM derives the IP address by preserving the host bits from the original IP address and appends them to the recovery site subnet.
For example, a VM with an IP of 10.20.200.55 would become 10.10.210.55 at the recovery site. Orange indicates the subnet at the respective sites, where the bolded number represents to number derived from the original VM and then applied at the recovery site.
To create a rule, click your desired site, add the proper Network Mapping and Add a new rule as shown below:
For the new rule, configure the subnet that exists in the Protected Site, in our case it’ll be a subnet in Louisville (10.10.200.0/24). On the right, configure the subnet which VMs will use at the Recovery Site, a subnet in Chicago (10.20.210.0/24). Make sure to add additional information which will be configured on failed VMs, such as the gateway and local DNS servers. (Remember that the protected site’s DNS could be down in the event of a failure).
Last but not least, you need to check the IP Customization Mode that each VM is configured for. The default setting is Auto, which says that the VM will be customized by using the IP Customization Rule. To find this setting, select your Recovery Plan, select the Related Objects tab, and click on Virtual Machines. Right click the VM, choose Configure Recovery, and check the settings associated with it.
Manual Customization: This option obviously is more time consuming but would be your option if you were failing over into an existing subnet where servers may already reside. As show in the screenshot above, you can customize each VM with the exact settings that you need rather than rely on a broader customization rule.
This one is pretty self explanatory but here’s a screenshot of configuring the IP at the protected site. One of the coolest features is the “Retrieve” button which will grab the IP address using VMware Tools on the VM. Pretty slick.