I gave a presentation at my local VMUG meeting recently on the new features of the Virtual Distributed Switch (VDS) that comes with vSphere 5.1. I thought I’d share some of those in this blog.
There are some really neat networking things that came out with this new release that I’ve been researching and spending some time on in the lab.
I already mentioned my favorite enhancement, roll back. Essentially what this does is constantly check to see if the management network is still connected. Keep in mind, the management network is what allows you to centrally manage your vSphere hosts and allows for things like High Availability configuration. In this new version, if an administrator happens to make a change that causes the management network to become disconnected, that change is automatically rolled back and an alarm goes off saying this task could not be completed. You can see an example of this in the YouTube video I created here. This even works if you’re trying to migrate your management network from your Virtual Standard Switch (VSS) to the VDS, which I’ve personally had trouble with in the past. Many admins have opted to just leave their management network on the VSS and keep all their Virtual Machine networks on the VDS, but this takes away from ease of management. The new roll back feature along with other features, such as backup and network health check can allow admins to feel more at ease putting their management networks on the VDS.
Backup and recovery is something you’ve been able to do with physical switches forever. However, there hasn’t been a good way to really backup the configuration of the VDS until now. As I stated above, this is another reason admins should feel more comfortable using the VDS exclusively. You can now backup your switches and port groups and use them to restore later or even use them as network templates to import into a new vSphere environment. If you have a backup you can restore after an incorrect change was made, but more importantly, if you experience any database corruption or data loss, you don’t need to worry about recreating your switches from scratch. Just restore the configuration and its back! VMware has also made it so that you can restore the VDS from the DCUI which previously was only an option for the VSS.
As I mentioned above, the new network health check feature is also pretty handy. It monitors VLAN, MTU and Network Adapter Teaming configurations. For example, if you defined a port group as using VLAN 5, but the physical port it’s connected to doesn’t trunk VLAN 5, the VDS will throw up an alert to let you know you’ve incorrectly configured something. It does the same for MTU mismatches and NIC Adapter Teaming misconfigurations. This feature does require that you have at least two physical uplinks for the VLAN and MTU monitoring and two active uplinks along with two hosts for the NIC Adapter Teaming checks.
There are several other features that have been integrated into the new version. You can read about them more in depth here (PDF). I also wanted to mention that these new features only come with Enterprise+ licensing as that is the only level that you can get the Virtual Distributed Switch with. These enhancements can only be utilized by the new web client as well. These new features are no longer being added to the old VI Client (fat client). There is a bit of a learning curve on that, so I would suggest putting this in a lab and practice using it before you upgrade your environment. One last caveat, none of these features will be available on the old standard switch. It will continue to work, but will most likely not receive any new features.