Firstly, a disclaimer: I’m not really au fait with VSAN, I’m just meddling.
If I’ve got 2 DCs with 2 x 40gb fiber between them (diverse paths) with sub 1ms latency, can I orchestrate VSAN to have redundancy between both DCs if we need to power off a DC?
I’m aware that stretched cluster is a thing, and I’ve used it before, but I was wondering if I had 3 nodes per DC, each all flash with 2 x disk groups per node – can I do something funky with the fault domains to allow the powering off of a DC for maintenance?
Thanks in advance!
View Reddit by candidhat – View Source