Test backups

[clincha]

When I have a storage solution in place

• I want to run a simulation of one node crashing so that I can make sure the applications are still available
• I want to delete the entire cluster so that I can make sure I can recover from an entire cluster failure quickly
• I want to delete the persistent volume from a node so that I can make sure I can recover a single persistent volume
• I want to recover to a previous snapshot so that I can revert mistakes