I guess I'm not clear on what you mean by moving it from dev to production. If you are happy with have it setup, and the server(s) you have it on can handle the production load, is there a reason you can't just point your users to the dev URL?
As far as disaster recovery, I can tell you how I do it. Daily snapshots of the SD VM along with daily backups of it's DB. In the even I need to roll back, I merely need to load up a snapshot from the day before and restore the DB to the same time. Yes, I'll lose a day's worth of work, but I'm okay with that. Most of our tickets are generated via the email monitor, so I'd also need to go into the mailbox and mark anything from the last day to unread so that they can picked up by the monitor again. If really necessary, I'd query the DB to get any tickets changed in the last day so at least I could give that list to the technicians. But I suppose it really comes down to how much work you are okay with losing. If the answer is none, I have no idea how you'd accomplish that.