Ubuntu 10.04 32-bit server only sees 8TB of 10TB LUN

ThatGraemeGuy asked:

I’ve created a 10TB LUN (7x 2TB disks in RAID6) on a storage array and presented it to our backup server, running Ubuntu 10.04 32-bit. The OS only sees 8000GB as per fdisk:

# fdisk -l /dev/sde

WARNING: GPT (GUID Partition Table) detected on '/dev/sde'! The util fdisk
doesn't support GPT. Use GNU Parted.

Disk /dev/sde: 8000.0 GB, 7999999442944 bytes
255 heads, 63 sectors/track, 972611 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disk identifier: 0x00000000

Disk /dev/sde doesn't contain a valid partition table

This isn’t a filesystem issue (I plan to use XFS FWIW), because I’m seeing the incorrect capacity before I’ve even partitioned this device.

I’ve searched Google high and low and I can’t seem to find an answer on where this limit is. Is it because of the 32-bit kernel?

update: using parted doesn’t make a difference.

# parted /dev/sde
GNU Parted 2.2
Using /dev/sde
Welcome to GNU Parted! Type 'help' to view a list of commands.
(parted) mklabel gpt
(parted) print
Model: YA-16SAE Backup_Vol_001 (scsi)
Disk /dev/sde: 8000GB
Sector size (logical/physical): 512B/512B
Partition Table: gpt

Number  Start  End  Size  File system  Name  Flags

(parted) quit                                                             
Information: You may need to update /etc/fstab.

My answer:

First, you shouldn’t be using fdisk. It can’t create partitions over 2TB anyway.

Check to see whether it’s a problem with fdisk by using parted:

parted /dev/sde print

If things look good, put a GPT label and a partition on it:

parted /dev/sde mklabel gpt
parted /dev/sde mkpart primary xfs 1 -1

If things don’t look good (which you said they didn’t) then it’s time to double check that you actually put all seven disks in the array, rather than just six of them.

View the full question and any other answers on Server Fault.

Creative Commons License
This work is licensed under a Creative Commons Attribution-ShareAlike 3.0 Unported License.