[Outages-discussion] Azure Postmortem

[Seth Mattinen]

On 9/12/18 10:46 AM, Matt Hoppes wrote:
> How long does it take to shutdown a server?  It shouldn't take more than 
> a minute or two I wouldn't think - and it seems it would be better to 
> just power off than risk hardware damage if the temperature got too 
> extreme.
> 
> I don't understand that part.


Let's say if a physical host or pod needs to shut down, and normal 
operating procedure is to take all the VMs running there and migrate 
them somewhere else. No problem, until hundreds or thousands try to shut 
down simultaneously. Then there's either nowhere for them to migrate to 
which causes a deadlock, or network capacity instantly hits a wall 
causing the migrations to drag out beyond whatever time is left.