Trying to mount 13 Tb disk on RedHat system.
Margaret Doll
Margaret_Doll at brown.edu
Fri Aug 14 13:02:13 UTC 2009
Progress so far
Used parted
to set the label to GPT
to create one 12 Tb partition
to set the filetype to ext3
dumpe2fs of /dev/scd1 looks good
pvcreate /dev/sdc1
pvdisplay /dev/sdc1
shows a 11.82 Tb volume
vgcreate /dev/vg0 /dev/sdc1
vgdisplay
shows a 11.82 Tb volume group
lvcreate -L 11.82T -n lv0 /dev/vg0
lvdisplay
shows a 11.82 Tb logical volume group
Problem comes with creating a filesystem on the logical volume
mkfs.ext3 /dev/vg0/lv0
mke2fs 1.39 (29-May-2006)
mkfs.ext3: Filesystem too large. No more than 2**31-1 blocks
(8TB using a blocksize of 4k) are currently supported.
Is 8 Tb the largest filesystem you can mount on a RedHat system.
Our system is 2.6.18-128.4.1.el5xen... 86_64 x86_64 x86_64 GNU/Linux
Thanks so far for everyone's help.
On Aug 14, 2009, at 8:22 AM, Margaret Doll wrote:
>
> On Aug 14, 2009, at 4:50 AM, Nigel Wade wrote:
>
>> Margaret Doll wrote:
>>> We got to create the 13 Tb partition on the aux disk on the
>>> RedHat system by using parted, but then mkfs.ext3 doesn't work on
>>> any partition larger than 2 Tb.
>>
>> It should do. I have 6 and 7.8TB filesystems. What version of RH
>> are you running? I suppose there may be limitations on some of the
>> kernels, I am running 64bit RHEL 4. However, I think there is
>> limitation of 8TB for ext2/3 on systems which have a max. block/
>> page size of 4kb. You need 8kb blocks/pages to get to 16TB, or a
>> different filesystem type and it's unlikely your system can handle
>> that. Do you know what the max. page size if for your system?
>
> How would I find this?
>>
>>> If we split the aux disk into 2 Tb partitions, I understand from http://www.linuxnix.com/2009/04/logical-volume-manager-lvm-in-redhat.html
>>> that we use fdisk to change the partition type to 83 Linux LVM.
>>> Unfortunately fdisk will only see the first 2 Tb partition, so we
>>> can't create a LVM
>>> of the partitions.
>>
>> If you want to go this way create individual LUNs on your RAID.
>> There is nothing to be gained by partitioning the RAID. You can
>> join the LUNs using LVM. That would enable you to create an LVM of
>> 13TB, but you've gained nothing over using the bare device. You
>> don't need a partition table on a device, you can create a
>> filesystem on the entire device, /dev/sdc for example, rather than /
>> dev/sdc1 etc.
>
> Using parted, I think we did create and ext3 filesystem on the
> entire disk.
>>
>>> We went back to parted and created one 13 Tb partition. Then
>>> inside parted, we used
>>> mkfs 0 ext3
>>> The program said that ext3 was not supported in this version of
>>> parted, but ext2 was. I oked, ext2.
>>> Inside parted
>>> (parted) print
>>> Model: IFT A16F-G2430 (scsi)
>>> Disk /dev/sdc: 13.0TB
>>> Sector size (logical/physical): 512B/512B
>>> Partition Table: gpt
>>> Number Start End Size File system Name Flags
>>> 1 17.4kB 13.0TB 13.0TB ext3 primary
>>> the partition is labeled as a ext3 file system.
>>> Our 13 Tb partition was added to /etc/fstab as a ext3 filesystem
>>> and mounted on the system.
>>> "df -h" though lists it as a 1.8 Tb system.
>>> /dev/sdc1 1.8T 196M 1.7T 1% /m3team
>>
>>
>> Did you actually create a filesystem on that 13TB partition? Or did
>> you just use the filesystem which you had already created
>> previously when the partition was only 2TB? I'd be interested to
>> know what mkfs said when you asked it to create a 13TB filesystem,
>> for example what block size did it set.
>>
>> What does 'dumpe2fs /dev/sdc1' show?
> dumpe2fs /dev/sdc1 > dump.rpt
>
> Where dump.rpt shows
>
> Filesystem volume name: <none>
> Last mounted on: <not available>
> Filesystem UUID: 2e6d48bb-5cde-4be0-af94-edf95ad64801
> Filesystem magic number: 0xEF53
> Filesystem revision #: 1 (dynamic)
> Filesystem features: has_journal resize_inode dir_index
> filetype needs_recovery sparse_super large_file
> Default mount options: (none)
> Filesystem state: clean
> Errors behavior: Continue
> Filesystem OS type: Linux
> Inode count: 244154368
> Block count: 488281245
> Reserved block count: 24414062
> Free blocks: 480569334
> Free inodes: 244154357
> First block: 0
> Block size: 4096
> Fragment size: 4096
> Reserved GDT blocks: 907
> Blocks per group: 32768
> Fragments per group: 32768
> Inodes per group: 16384
> Inode blocks per group: 512
> Filesystem created: Thu Aug 13 15:26:09 2009
> Last mount time: Thu Aug 13 15:59:28 2009
> Last write time: Thu Aug 13 15:59:28 2009
> Mount count: 1
> Maximum mount count: 33
> Last checked: Thu Aug 13 15:26:09 2009
> Check interval: 15552000 (6 months)
> Next check after: Tue Feb 9 14:26:09 2010
> Reserved blocks uid: 0 (user root)
> Reserved blocks gid: 0 (group root)
> First inode: 11
> Inode size: 128
> Journal inode: 8
> Default directory hash: tea
> Directory Hash Seed: 149a19fb-3302-49ff-8314-3a02abd591d1
> Journal backup: inode blocks
> Journal size: 128M
>
>
> Group 0: (Blocks 0-32767)
> Primary superblock at 0, Group descriptors at 1-117
> ...
> ...
>
> Group 14900: (Blocks 488243200-488275967)
> Block bitmap at 488243200 (+0), Inode bitmap at 488243201 (+1)
> Inode table at 488243202-488243713 (+2)
> 32254 free blocks, 16384 free inodes, 0 directories
> Free blocks: 488243714-488275967
> Free inodes: 244121601-244137984
> Group 14901: (Blocks 488275968-488281244)
> Block bitmap at 488275968 (+0), Inode bitmap at 488275969 (+1)
> Inode table at 488275970-488276481 (+2)
> 4763 free blocks, 16384 free inodes, 0 directories
> Free blocks: 488276482-488281244
> Free inodes: 244137985-244154368
>
>
>>
>> --
>> Nigel Wade, System Administrator, Space Plasma Physics Group,
>> University of Leicester, Leicester, LE1 7RH, UK
>> E-mail : nmw at ion.le.ac.uk
>> Phone : +44 (0)116 2523548, Fax : +44 (0)116 2523555
>>
>> --
>> redhat-list mailing list
>> unsubscribe mailto:redhat-list-request at redhat.com?subject=unsubscribe
>> https://www.redhat.com/mailman/listinfo/redhat-list
>
> --
> redhat-list mailing list
> unsubscribe mailto:redhat-list-request at redhat.com?subject=unsubscribe
> https://www.redhat.com/mailman/listinfo/redhat-list
More information about the redhat-list
mailing list