2

One of my Debian servers (kernel 2.6.30-AMD64) refuses to start the software RAID array that houses the root partition in initramfs. It dumps me with a busybox console. When I follow the necessary steps to continue booting it works fine (start the array with mdadm -A and then have LVM scan the volumes with pvscan and then vgchange -ay).

I've tried starting with boot options rootdelay=10 to no avail.

Also I've updated the initramfs and unpacked it to inspect whether it really tries to assemble the raid array (it does).

Output before dumping to console :

mount: mounting none on /dev failed: No such device
W: devtmpfs not available, falling back to tpmfs for /dev

and then some lvm messages saying it can't find the volumes holding the root partitions.

Does anybody have a clue how I could fix this?

Improve this question
3
  • have you changed the partition IDs to RAID autodetect? (set 1 raid on in parted)
    – Hubert Kario
    Oct 18, 2010 at 21:36
  • yes, well, I didn't change them, they were already set on creation. Under 'Flags' in parted it says raid on all partitions that need to be assembled.
    – Jasper
    Oct 19, 2010 at 11:11
  • Do you have your RAID described in /etc/mdadm.conf?
    – poige
    Feb 28, 2011 at 18:41

3 Answers 3

Reset to default
1

You may need to create /etc/mdadm.conf and possibly rebuild initramfs. Check this answer for details.

Improve this answer
1
  • yes, I have the /etc/mdadm/mdadm.conf (this being Debian) and it's correct and I did update the initramfs and checked it and it's correct. I can also see that the initramfs 'knows' it's correct, because I can assemble with mdadm --assemble --scan (in initramfs shell).
    – Jasper
    Nov 25, 2011 at 14:43
1

For starting server by hand, please try:

(initramfs) mdadm /dev/md0 -A /dev/sda1 /dev/sdb1
[...]
(initramfs) mdadm /dev/md1 -A /dev/sda2 /dev/sdb2
[...]
(initramfs) mdadm /dev/md2 -A /dev/sda3 /dev/sdb3
[...]

and so on, all of your md partitions.

If you have LVM, go on with:

(initramfs) vgchange -ay
  3 logical volume(s) in volume group "pve" now active

then simply type:

(initramfs) exit

and if all is fine, your system will boot.

Improve this answer
0

I have similar problem, with duplicate name on mdadm.conf

I fix on the following way, thanks to user130370 solution. [initramfs]mdadm --examine --scan > /etc/mdadm/mdadm.conf [initramfs]mdadm -q --assemble --scan --run [initramfs]exit

rebot automatically

from the system as first

#update-initramfs -u

Improve this answer

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .