SRM Tutorial Part 7: Configuration #1

After we setup and installed successful the SRM environment in the last parts it´s now time to get to the configuration. Due to the fact the whole configuration including all the steps is too much content for a single post I spread it into two parts.


SRM Tutorial Part 1: Lab setup
SRM Tutorial Part 2: Components & design
SRM Tutorial Part 3: SRM installation
SRM Tutorial Part 4: NetApp Ontap Simulator – Setup & configuration
SRM Tutorial Part 5: Configure NetApp SnapMirror
SRM Tutorial Part 6: SRA installation
SRM Tutorial Part 7: Configuration #1
SRM Tutorial Part 8: Configuration #2
SRM Tutorial Part 9: Advanced configuration & troubleshooting

Configure remote connection:

Start the vSphere Client and connect to the vCenter of the primary site. After you started the SRM GUI you see now just the single SRM instance. So the first step is to create a connection to the SRM instance on the secondary site.
Click on the link Configure Connection on the right to start the wizard.

Next you must enter the vCenter FQDN of the remote site.

If you aren’t using official certificates you will get a warning message, which you can ignore.

The last step of the wizard asks you for the vCenter credentials of the remote site.

After the connection got established successful you will be asked again for the login of the remote site. This will pop up by the way every time you start the SRM plug-in (even if you use the linked mode and have the same login data on both sites), which becomes sometimes a little bit irritating.

Now you will see both sites on the left in the same interface. The “local” caption behind one of the sites shows on which site you are logged in. Also you see that the status changed to connected.

Add storage arrays:

Before you can start to add the storage arrays to SRM, you need to verify that both SRA adapters work properly. As both sites are now connected the previous warning (shown in Part 6: SRA Installation) should now disappear. Maybe you need to click again on the “Rescan SRAs” link.

On the Summary tab on the Array Managers menu you find the option Add Array Manager. This step is required for both sides, but you can now stay in the GUI from the primary site and just select the secondary site on the left.

First you need to select the matching SRA for the array you want to add. In this tutorial we just installed a single SRA, which means you will just have a single option here. In an environment with many arrays from different vendors you will have multiple SRAs installed you can select here. Also you need to define a display name for the array (you can change this anytime).

Next you need to add the IP addresses of the array and the user credentials for the root login. In this lab I´m just using a single interface of the NetApp simulator that’s why the system and the NFS IP address is the same. This will properly not the case in a productive environment.

Hopefully the wizard response with a successfully reply. Otherwise take a look on the troubleshooting section in the last part of this tutorial, because this is a point where many people stuck with some errors.

On the left you should see now both arrays below the related site.

Finally you need to configure the relationship between the two recently added arrays. Click on the Array Pairs tab, where you will find an Enable button.

On the Devices tab you will now see the two SnapMirror replications we setup during the last parts of this tutorial. The arrow shows the direction of replication, which would change after a failover and a reprotect (more about this later).

Resource-, Folder- and Network-Mappings:

Go back to Sites on the left, where you may already noticed the additional mapping tabs since we setup the connection between the primary and the secondary site. But before we can go to into the configuration of these I will show you at first my vSphere environment.

At the primary datacenter I have a couple of VMs, which I want to protect with SRM. Next to these I also have some local VMs, which are not on a replicated datastore and which I don´t want to protect with SRM. Please note the use of resource pools in this example. It´s not a requirement to use resource pools with SRM, but it makes the resource mapping in the next steps more clear (and will also often apply in a real world condition).

The folder structure of the VM and Template view looks similar to my resource pools. Of course you can have any structure you need and without any relationship to the Host and Cluster view.

First we map the resources to the secondary site. Here you need to build a relationship between clusters and or resource pools to the same on the secondary site. In my lab I only want to protect the VMs below the resource pool “SRM Protected VMs”. If you are not using resource pools you would need to configure the mapping at the cluster level.
Select the object, right-click and select Configure Mapping.

In the pop-up you see the structure of your secondary site. Select the resource pool, where SRM should put the VMs into if a failover occur.

After the configuration SRM displays the relationships. Of course you can map multiple objects.

The approach for the folder mapping is exactly the same.

Also the configuration steps for the network mapping is the same. In my lab this mapping is quite simple. I just have a single VM network so nothing special at all. May in a productive environment you don´t have a stretched VLAN over both sites. In this case you may need to put the network connection of your VMs into a totally different VLAN or multiple networks form site A can end up in a single network of site B. Please note that also distributed switches are supported on this point.

Placeholder datastore:

First I want to give you a basic understanding what placeholders are used for in SRM. If you protect a VM form the primary site (what we will do in the next part) SRM creates a placeholder VMX file (and some other small files) on a so called placeholder datastore at the secondary site. The placeholder VMX file is already registered in the vCenter of the secondary site and also shown in the inventory, but are handled as special SRM objects until a failover occurs. Most of this file is a 1:1 copy of the original VMX file but also includes the modification we configured during our last steps (the whole mapping process). You will see more about this in the next part of this tutorial. May you wounder why we also need to created this placeholder datastore on the primary site. SRM 5 introduced the new reprotect and failback feature. After a switchover to the secondary site took place you can reprotect the environment, when the primary site is available again. In this case SRM will flip over the whole protection the other way around, which means that in this scenario the placeholder datastore is also required at the primary site.

At this point now the environment is ready to protect VMs with SRM. How to build Protection Groups and Recovery Plans I´m going to show in the next part.