The advice, which is specifically for virtual machines using Azure, shows that sometimes the solution to a catastrophic failure is turn it off and on again. And again.
Have you tried turning it off and on and off and on and off and on and off and on and off and on and off and on and off and on and off and on and off and on and off and on and off and on and off and on and off and on and off and on and off and on again?
Have you tried turning it off and on and off and on and off and on and off and on and off and on and off and on and off and on and off and on and off and on and off and on and off and on and off and on and off and on and off and on and off and on again?