[CentOS] large SCSI RAID, replacing server

[James A. Peltier]

----- Original Message -----
| I have a system running CentOS 6.3, with a SCSI attached RAID:
| 
| 
| http://www.raidweb.com/index.php/2012-10-24-12-40-09/janus-ii-scsi/2012-10-24-12-40-59.html
| 
| 
| For disaster recovery purposes, I want to build up a spare system
| which could take the place of the server hosting the RAID above.
| 
| But here's what I see:
| 
| # fdisk -l /dev/sdc
| 
| WARNING: GPT (GUID Partition Table) detected on '/dev/sdc'! The util
| fdisk doesn't support GPT. Use GNU Parted.
| 
| 
| Disk /dev/sdc: 44004.7 GB, 44004691814400 bytes
| 255 heads, 63 sectors/track, 5349932 cylinders
| Units = cylinders of 16065 * 512 = 8225280 bytes
| Sector size (logical/physical): 512 bytes / 512 bytes
| I/O size (minimum/optimal): 524288 bytes / 524288 bytes
| Disk identifier: 0x00000000
| 
|    Device Boot      Start         End      Blocks   Id  System
| /dev/sdc1               1      267350  2147483647+  ee  GPT
| Partition 1 does not start on physical sector boundary.
| #
| 
| 
| But here's the partitions I have:
| 
| # df -k |grep sdc
| /dev/sdc1            15379809852 8627488256 6596071608  57% /space01
| /dev/sdc2            6248052728 905001184 5279574984  15% /space02
| /dev/sdc5            8175038780 2418326064 5673659088  30% /space03
| /dev/sdc4            6248052728 1444121916 4740454252  24% /space04
| /dev/sdc3            6248052728 1886640284 4297935884  31% /space05
| #
| 
|  
| How can I build up a new system to be ready for this existing RAID?
|    Or will the latest/greatest CentOS just know what to do, and allow
| me to simply copy the /etc/fstab over and respect it?

Personally, I wouldn't (and don't) partition the disk(s) at all.  I instead use LVM to manage the storage.  This allows far greater flexibility and could allow you to provision the disks better.  I'd recommend you use LVM on your new machine to create the volumes and use file system volume labels instead of physical drive locations such as /dev/sdc1.  Volume labels are not system specific, although if you have conflicting labels it could be problematic.

pvcreate /dev/sdc
vgcreate DATA
lvcreate -L 15379809852 -n space01 DATA
lvcreate -L 6248052728 -n space02 DATA
...

mkfs.xfs -L space01 /dev/DATA/space01
mkfs.xfs -L space02 /dev/DATA/space02
...

then in /etc/fstab

LABEL=space01 /space01 xfs defaults 0 0 
LABEL=space02 /space02 xfs defaults 0 0 
...

This of course all depends on what it is that you're trying to accomplish but I would certainly recommend moving away from using partitions.  If you're just rsync'ing the data you don't have to worry about the sizes being the same, you just have to worry about the sizes being sufficiently large to store the amount of data needed.

Currently I can see that you have well over provisioned the amount of space required and being able to manage growth of say /space01 which seems to be much larger, is somewhat fixed in your configuration, how would you handle the growth or shrinking of other partitions to make room.  With LVM you would provision space with some growth of what's there now and then grow file systems as needed.

As a side note, you may want to investigate Gluster or DRDB to actually replicate the data across the nodes giving you a more "true" replication and fail-over configuration.

-- 
James A. Peltier
Manager, IT Services - Research Computing Group
Simon Fraser University - Burnaby Campus
Phone   : 778-782-6573
Fax     : 778-782-3045
E-Mail  : jpeltier at sfu.ca
Website : http://www.sfu.ca/itservices

“A successful person is one who can lay a solid foundation from the bricks others have thrown at them.” -David Brinkley via Luke Shaw