Fail over/Recovery roadblock

[Aaron E.]

We are upgrading to S4 I'm just waiting on the servers being ordered 
today and there are few items on the list prior to production which 
includes failover/recover. In my mind unless I have a proper procedure I 
cannot go live so I'm looking at recovery procedures in case of DC1 
failure in a two DC set-up. Was hoping for some insight / procedures 
that people are using in a production environment. I'll list my thought 
process and hope others can chime in on anything I'm missing or haven't 
thought of.

#1. I know that AD is replicated minus group policy so I will rsync the 
policies, no problem there..

#2. I understand the process of seizing roles so there shouldn't be a 
problem there. (Ill test when I get the full logistics down)

#3. Finally DNS, I was first thinking I could rsync the proper files 
over then I remembered a conversation with Ricky N. about inodes and dns 
files.. So I am at a loss and was hoping for some direction on procedure 
for dns recovery/ backup solution in case of failure on DC1.

#4. Missing anything?


And now Deep thoughts by Jack Handy!!! (if anyone members old snl)!
I was thinking about corosync / pacemaker and rsyncing the whole samba 
configuration in a failover scenario so I would only have one DC active 
at a time with exact copies, this would provide me with the same ip on 
all servers if needed., heck maybe even have 4 servers in this 
configuration just in case.... Thoughts?