External array showing as /dev/sda

Eberhard Moenkeberg emoenke at gwdg.de
Sun Mar 21 16:12:33 CDT 2010


On Sun, 21 Mar 2010, David Hubbard wrote:

> From: Behalf Of J. Epperson
>> I'm somehow missing how getting the non-installable smaller
>> GPT VD to be /dev/sda will change that scenario. The other
>> responder echoed one of my initial thoughts when he suggested
>> turning off the external array.  That should do it.
> If I could get the internal raid controller to be
> /dev/sda, then the RHEL/centos installer will not
> care about the fact that the external array is
> too big and would require GPT to boot off of, then
> the installer would let me proceed.   It was only
> an issue with it being /dev/sda since that made the
> installer think there was no way to writen an MBR
> and boot off of it.
> But, unplugging external did lead me the right direcation.
> What I've had to do is this:
> 1) Internal array I had desired to be single RAID 50
> across 8 drives.  Thanks to Dell's choice of LSI
> for their current raid controllers and LSI missing
> the feature that most others seem to have in being
> able to present parts of one array as multiple
> logical drives, I ended up having to waste the first
> two drives to make a RAID 1 mirror smaller than 2 TB
> and then only six remaining drives in the RAID 50.
> 2) Unplugged the external array and installed centos
> using normal non-GPT boot to the raid 1 virtual
> drive.  It installed to /dev/sda.
> 3) After install, edit /boot/grub/device.map and
> changed it to show:
> (hd0) /dev/sdb
> Then:
> grub
> grub> device (hd0) /dev/sda
> device (hd0) /dev/sda
> grub> root (hd0,0)
> root (hd0,0)
> Filesystem type is ext2fs, partition type 0x83
> grub> setup (hd0)
> setup (hd0)
> Checking if "/boot/grub/stage1" exists... no
> Checking if "/grub/stage1" exists... yes
> Checking if "/grub/stage2" exists... yes
> Checking if "/grub/e2fs_stage1_5" exists... yes
> Running "embed /grub/e2fs_stage1_5 (hd0)"...  15 sectors are embedded.
> succeeded
> Running "install /grub/stage1 (hd0) (hd0)1+15 p (hd0,0)/grub/stage2
> /grub/grub.conf"... succeeded
> Done.
> 4) Reboot, connect external array while server is
> rebooting, comes back up and boots off of internal
> array from bios, grub is happy because now it is
> set up for /dev/sdb.
> Only downside to this situation is if something were
> to fail and take the external array down the server
> won't boot since internal will go back to being
> /dev/sda.  But if the external array is down then
> we've got issues anyway. :-)

The real matter is the sequence of drivers within the initrd file.

I can't tell about RH, but with SUSE you have /etc/sysconfig/kernel with a 
line like

   INITRD_MODULES="amd74xx megaraid_mbox processor thermal fan jbd ext3 \
   edd aic7xxx qla2300 tg3"

where you can change the sequence for your needs.
After changing, the initrd file needs a rebuild.

Viele Gruesse
Eberhard Moenkeberg (emoenke at gwdg.de, em at kki.org)

Eberhard Moenkeberg
Arbeitsgruppe IT-Infrastruktur
E-Mail: emoenke at gwdg.de      Tel.: +49 (0)551 201-1551
Gesellschaft fuer wissenschaftliche Datenverarbeitung mbH Goettingen (GWDG)
Am Fassberg 11, 37077 Goettingen
URL:    http://www.gwdg.de             E-Mail: gwdg at gwdg.de
Tel.:   +49 (0)551 201-1510            Fax:    +49 (0)551 201-2150
Geschaeftsfuehrer:           Prof. Dr. Bernhard Neumair
Aufsichtsratsvorsitzender: Dipl.-Kfm. Markus Hoppe
Sitz der Gesellschaft:     Goettingen
Registergericht:           Goettingen  Handelsregister-Nr. B 598

More information about the Linux-PowerEdge mailing list