This is my first post on this forum, figured I'd leap right in ....
I tend to agree with the p.o.v that if you have no reason to rebuild it, why risk it? If its running slow - have you checked that all disks in RAID 5 array are online and functioning properly. Have you checked event viewer for faults? Run performance monitor to detrmine if something is hogging resources.... etc.
Some guides for each point:
a) 37 hours is doable, just don't leave it till then to plan everything. Take screen dumps of specific application configs where applicable. Save them anywhere but on the server in question . Also leave time to backup AND verify. Takes twice as long but worth it.
b) If you have a separate server, promote it to DC no harm in having a secondary DC anyway - guess it shouldn't matter at this stage about RAID5 on second DC especially if you've got good full backups.
c) Take most applications off the DC keep it for name resolution & DHCP (recommended for security and performance reasons) ... and if you move apps - remember to update workstations .... avoids the 'just-got-back-from-holiday-and-IT-broke-the-network' grumps
Planning is key, make sure you have:
1) An implementation plan - how you are going to do this
2) A contingency & backout plan - In the event of data loss or disaster .... know what your critical apps are and make sure you know how to sort them, if not - who to contact for support.
3) A communication plan - although nobody is expected to be there, just let everyone know that things will be offline to avoid too many system changes
Plan, plan and plan ....
