Disaster Recovery / Moving to New Hardware

[hikenboots]

Sorry to post another thread on a much discussed topic ; I didn't want to hijack anyone else's thread.

I'm pressed to get Scalix CE into production on short notice but don't want to until I understand how to restore scalix from my current backup.

My backup method:
I'm using Fedora Core 4 with Software RAID 1 (mdadm) am creating a tar of my filesystem as follows:
1. Stop my databases, stop Scalix server, etc. for a consistent backup image.
2. Remove partition sdb2 of my mirrored array (/dev/md0) and remount it as /backup into the filesystem. (there's a boot record on this drive so it boots fine if drive 1 were to fail during the backup process)
3. Restart my databases, start Scalix server, etc. to get users working again.
4. Tar /restore to tape, unmount /restore and re-add sdb2 to /dev/md0

How would I go about restoring the installation to new hardware? I'm thinking that I can bring up a FC4 install on new hardware, match hostnames, IP addresses, DNS records, install Scalix then stop the services and restore /var/opt/scalix over my new install. Is this an acceptable paradigm? Which directories would need to be restored? I'll be giving it a test run this week but there's no sense reinventing the wheel; any advice would be appreciated.

[ScalixSupport]

The main thing to get right is ensuring that the permissions are correct so that the uid and gid are scalix:scalix where appropriate.

If you use tar, it will restore the permissions provided that you have the scalix uid and gid in /etc/passwd and /etc/group.

Cheers

Dave