Recovering VMware Management Stack in Under 20 Minutes
When was the last time your organization was able to successfully test and validate recoverability of all management components of an enterprise-size VMware infrastructure under 20 minutes? Maybe never? If so why?
The reality is this is a difficult task to achieve, especially when having to adhere to strict recovery time objectives. The difficulty of the task may be due to complexity, time consumption, operations, dependencies, and so much more.
There are lots of critical reasons as to why this is a challenge.
VMware recommends that organizations run and maintain all infrastructure management components in a separate and dedicated cluster. This “management cluster” covers services or features for business applications (management stack) such as orchestration, auditing, authentication, automation, service catalogs, etc. This recommendation helps prevent contention between business and infrastructure related applications, and it is also beneficial for business continuity planning.
Now, in the event of a disaster or failure, the primary focus is always on the line of business applications. I’m not advocating against this, but I have a word of caution about prioritizing the management cluster. It will be much tougher to bring back those line of business applications – and get the business back on track – without managing infrastructure components; these provide all of the various infrastructure and management services to the lines of business applications.
Bottom line: The business applications may not be able to function until the management stack is online.
Test and Development
We all know there is a demand for the testing and development of applications to be as close and as accurate as possible to the application and data found in the production environments. This can be challenging to achieve at times for many reasons. It is simple enough to restore a VM or an application to the infrastructure. However, in many cases, applications depend on core infrastructure services that won’t be available in the test and development environments.
This raises another tough challenge in the data center that is also worth thinking about. The good news is there is a solution to this problem.
Don’t just take my word for it. Take a look at the demonstration below based on a solution we have developed for VMware Cloud Foundation. The demonstration highlights the recoverability of the VMware Cloud Foundation management stack in under 20 minutes. The solution and approach can be applied to both business continuity as well as test and development scenarios.
Pay attention as this is a recoverability scenario and not necessarily availability. The demonstration showcases the complete loss and deletion of all of the VMs that are stored and running in the vSAN Datastore.
Also note that this solution can be applied to management infrastructures that aren’t necessarily based on VMware Cloud Foundation. This has been tested and validated to work with any VMware vSphere related infrastructure.