Hi, and welcome!
Vetting is still a thing if that’s what you mean.
It usually takes a few weeks for a node to get vetted. There is no official way to know when it’s done, but @brightsilence’s excellent tool can be used to show vetting progress (among other things):
Kind of. Using ETH is fine but costs more to StorjLabs. For SNOs, the main difference is that there is a minimum amount threshold to be reached when getting paid in ETH (which changes every month, depending on ETH’s fees), which doesn’t apply to zkSync. Also so far, getting paid via zkSync grants you an extra 10% payment.
Nodes don’t do anything in particular to keep HDDs active. That said, even though a node’s disk can spin down from time to time when the node is brand new, the more data it gets, the more the disk gets used, so it quickly stays ON 24/7.
Both approach sound valid to me.
What’s important to note is that all nodes behind a /24 subnet are treated as a big single node by the Storj network.
Which means that if you have several nodes on the same IP, or on several IPs that share the same /24 subnet, they will all share ingress.
Hard to answer that, this forum is full of info, good practices, experience and tools ^^ Go explore 
For me, the first thing to set up is a tool monitoring your node’s port, so you’re notified when it gets offline. A popular tool for that is uptimerobot.