VDI is no different from mission-critical applications in the enterprise: When a disaster or critical failure occurs,...
your users expect their virtual desktops to be available as soon as humanly possible.
As a matter of fact, depending on whether or not your virtual desktop infrastructure (VDI) configuration supports disconnected desktops, most VDI users will be unable to perform their normal work duties until their desktops are operational. This is a business continuity problem masquerading as an IT problem. Remember, getting your VDI environment back up and running doesn't necessarily mean that your users can resume their day-to-day business activities that keep revenue coming in, keep customers fully supported, and keep the wheels of the business turning.
What constitutes a disaster?
What if the interruption to VDI is due to an ordinary mechanical failure or other interruption instead of a natural disaster? From the users' perspective, it doesn't matter what is the root cause of a VDI outage. If they can't perform their normal duties, you have a disaster on your hands, whether it's precipitated by a malware or virus infection in the environment, a motherboard failure in a critical VDI server, a power outage in the data center, or a tornado damaging the data center that supports your users.
Step 1: Assess the damage
Before you can start your VDI recovery efforts, you must know what exactly is broken. Your monitoring tools will likely be throwing alerts and errors left and right. The infrastructure team that supports your VDI environment will be very busy trying to identify and isolate the affected components. This damage assessment marks the beginning of your recovery efforts by identifying known down, unresponsive or unreachable servers and network devices.
Step 2: Implement your VDI recovery plan
If you don't have a VDI disaster recovery (DR) plan in place, or at least have a dedicated VDI section in your company's DR/business continuity plan, you have a very rough road ahead. Nothing has more of a positive effect on your recovery efforts than an up-to-date, well-tested DR plan.
Assuming you have one, dust off your plan and start your VDI recovery activities. You might have to restore virtual machine (VM) snapshots from the cloud, you might have physical hardware that must be repaired or replaced, or you might need to access application data from a replicated database. Using your DR plan as a guide, start the process of recovering your VDI environment, engaging with the appropriate IT support teams for each affected component.
Step 3: Continually assess the recovery process
In the event of a disaster or critical VDI failure, do not trust your monitoring tools to tell you when your environment is back up and operational. You must confirm proper VDI operation with a representative group of normal, non-IT VDI users. This is the only true test that your nightmare is over.
Step 4: Document your VDI recovery activities and outcomes
Be sure to thoroughly document each step and activity that leads to VDI recovery. This is the hardest of the five steps to complete because there will likely be many proverbial fires to address during a real DR event, and documenting what you are doing to recover specific VDI components as you do them is challenging. However, without this documentation, you condemn yourself to making the same mistakes next time.
Step 5: Update your DR plan with lessons learned
More on VDI disaster recovery
Backing up VDI for proper disaster recovery
Planning for VMware View disaster recovery
Once the DR event has been successfully resolved and all VDI operations are back to normal, your job is still not complete. It is imperative for the success of future DR and business continuity efforts that your support staff document reflects what worked, what didn't and why. Disaster recovery best practices dictate that a "lessons learned" document be developed as soon as possible after an event, and this should be used to update the existing DR plan. The only thing worse than making a mistake while striving to recover from a disaster or unplanned outage is to make the same mistake in subsequent disaster recovery efforts.
Make sure that VDI is part of your overarching DR/business continuity plan and keep that plan up-to-date by taking advantage of your company's change control process to note any and all architectural changes to the VDI environment. When and if the time ever comes to recover from a disaster, you are poised for success with a solid, well-tested DR plan and the discipline to follow that plan when everything else around you is in chaos.