vSphere Replication 5.8 and Site Recovery Manager 5.5

vCloud Suite 5.8 had just been released with new product releases like vCenter 5.5 Update2, ESXi 5.5 Update2, SRM 5.8, vSphere Replication 5.8 to name a few. For a complete list, you can refer to this link.

Itching to try out the new vSphere Replication 5.8 in conjunction with SRM 5.5, I went to the download site to get it. But looking at the dowload page for VR 5.8, it looks like vSphere Replication is not supported in SRM 5.5.

And verifying the VMware Product Interoperability Matrix, it is confirmed that only SRM 5.8 is supported.

But for standalone vSphere Replication without SRM, VR 5.8 is supported down to ESXi 5.0

 

 

Deploying an Additional vSphere Replication Server

Depending on a customer’s requirement, there may be a need to deploy additional vSphere Replication servers. A few use cases are for availability and load balancing. Another one I can think of is if you want to deploy a vSphere Replication Server on a remote site that is managed by a vCenter Server on the main site. With vSphere Replication 5.5, you can add up to 9 VRS for a total of 10 including the first VR appliance which holds both the VRMS and VRS functionality. For a complete detail on vSphere Replication/SRM limits, see KB Article 2034768

In this article, I’ll show how easy it is to deploy an additional vSphere Replication Appliance.

1. In the web client, go to Manage–> vSphere Replication–> Replication Servers. Click on the OVF deployment icon to deploy the VRS ovf.

2. The file to select is vSphere_Replication_Addon_OVF10.ovf. This OVF is only 512MB in RAM, all other resources (CPU, disk) is the same as the vSphere Replication appliance OVF.

3. Review the OVF details and click Next.

4. Select the name of the VRS and folder to place it to, then click Next

5. Select the cluster as a resource to run the VRS. Click Next

6. Select the datastore where the VRS will be located. If needed, you can also change the virtual disk format. Click Next

7. Select the portgroup that will be used by the VRS. Under IP Allocation I selected Static-Manual and configured the DNS, Netmask, and Gateway settings. Click Next

8. Provide a password for the VRS root account as well as the IP address of the VRS. Click Next

9. Click Finish to start the OVF deployment.

Once the VRS is powered on and initialized, we now have to register this new VRS as an additional VRS. To do so, perform the following:

1.  In the web client, go to Manage–> vSphere Replication–> Replication Servers. Click the middle icon (Register a virtual machine as vSphere Replication Server) and select the newly deployed VRS.

2. Once registered, you will now see the new VRS under the vSphere Replication tab

You can now select this VRS or use Auto-assign when protecting a VM with vSphere Replication:

If you decided you don’t need the additional VRS, you need to unregister the VRS before removing/deleting it.

That’s it. Never gets easier than that..

VM Replication status is “Not Active” after upgrade to vSphere 5.5/VR 5.5

Weird issue I encountered after I deployed vSphere Replication appliance on my upgraded vSphere 5.5 lab. I was able to configure vSphere replication for a VM but for strange reasons, the VM replication shows as Not Active. This worked fine when I was still using vSphere 5.1/VR 5.1. The hostd.log file shows the following error:

The configuration for vSphere Replication all looks good, all are green so I can’t figure out what’s wrong here.

Luckily, I found a  VM community post which is similar to my issue found here

I remember that in 5.1, I configured a vSphere Replication vmkernel port tick box in web client. This vmkernel port tick box is now removed in vSphere 5.5. As far as I know, that tick box is useless in vSphere 5.1 coz even though its checked, VR traffic will always default to the management vmkernel. See Sunny Dua’s article on properly configuring a different management network for vSphere replication

Going back to my issue, I followed the suggestion to remove the lines similar to the following in esx.conf of the ESXi host:

/net/vmkernelnic/child[0001]/tags/4 = “true”

The number [0001] may change depending on the environment. On mine its [0002]. After making the change and rebooting the ESXi hosts, vSphere Replication started working fine again.

“Failed to Initialize” message on vSphere Replication 5.5 Appliance

I’m currently spending most of my lab time honing my SRM skills preparing for my next project. The past few days I have upgraded my lab to SRM 5.5.1 using a NetApp storage simulator and NetApp SRA 2.1. All works good so far with array based replication.

So I decided to try using vSphere Replication. However I am encountering issues after deploying the vSphere Replication appliance. It will deploy fine but its not taking the IP address that I set (reverts to DHCP). When I login to the management webpage https://ip-address:5480, it gives me the message “Failed to Initialize“. I tried redeploying the appliance a couple of times but still encounters the same issue.

Then I realized I have been deploying the appliance using the vSphere Client. So I tried deploying it using the Web Client, and after deployment and poweron of the appliance, I don’t see the error anymore and I was able to login and configure the appliance.

I’m not sure if this is new to vSphere Replication 5.5 but anyways, will always try to remind myself to try and move away to the vSphere Client. But since SRM5.5 is still managed using the vSphere Client, then that will have to wait 🙂