Re: [opensuse] Software RAID with existing drive
On 07/09/2007 05:47 AM, Brandon Carl wrote:
Well, I've gotten really far, I think. I mounted both raid partitions and copied over everything under "/" except for /sys, /mnt, and /proc
/dev/md10 (RAID1: /dev/hdb1 & missing) will mount as / /dev/md11 (RAID1: /dev/hdb3 & missing) will mount as /home
I have grub installed in the MBR of both drives, so I can boot from either. My menu.lst is setup as so:
title Boot from Hard Drive A root (hd0,0) kernel /boot/vmlinuz root=/dev/hda1 noapic resume=/dev/hda5 splash=silent showopts initrd /boot/initrd
title Boot from RAID root (hd1,0) kernel /boot/vmlinuz root=/dev/md10 noapic resume=/dev/hdb5 splash=silent showopts initrd /boot/initrd If you raided swap, resume would be /dev/md2, or whatever you swap raid was. Just read below, so swap, or the resume line, may be ok.
Ok.
That way I can boot from either the first drive, or the one-drive RAID on the second drive.
I have run into a problem, however. When I attempt to choose "Boot from RAID" it gets to the boot commands, but it stalls at "Waiting for device /dev/md10 to appear: ................................. not found -- Exiting to /bin/sh".
Why is it now md10? It should be md0, md1, md2. What does /etc/mdadm.conf show you?
I changed it to md10 because somehow I accidentally deleted /dev/md0, /dev/md1, and /dev/md2, probably in an attempt to unmount them, lol. I couldn't find a way to create them again, so I just decided to use md10 and md11. I can't imagine this would affect anything, however. /etc/mdadm.conf does not exist on either drive.
After this I reboot my computer and choose the "Boot from Hard Drive A" and then when I try to mount the /dev/md10 as /mnt it says: "cannot read superblock". I get around this by going into the Yast partitioner and deleting the /dev/md10 and /dev/md11 and by doing "mdadm -S /dev/md10;mdadm -S /dev/md11" as super user. I then do "mdadm -C /dev/md10 -l raid1 -n 2 /dev/hdb1 missing; mdadm -C /dev/md11 -l raid1 -n 2 /dev/hdb3 missing" to recreate the two raid partitions. After that it is back to normal. Why md10? If it is the first md, it should be 0.
See above.
Here is the contents of my fstab under the /dev/md10, in case it's helpful: /dev/md10 / reiserfs acl,user_xattr,usrquota,grpquota 1 1 /dev/md11 /home reiserfs acl,user_xattr,usrquota,grpquota 1 2 /dev/hdb5 swap swap defaults 0 0 proc /proc proc defaults 0 0 proc /proc proc defaults 0 0 sysfs /sys sysfs noauto 0 0 debugfs /sys/kernel/debug debugfs noauto 0 0 usbfs /proc/bus/usb usbfs noauto 0 0 devpts /dev/pts devpts mode=0620,gid=5 0 0 /dev/fd0 /media/floppy auto noauto,user,sync 0 0
Is that a typo, or are you missing a comma between reiserfs and acl? Do you really have proc twice? You did make sure there was a /proc and /sys folder for mounting their respective filesystems.
I copied directly from the file, but there is a few spaces in between the reiserfs and the acl,user... The formatting got messed up, sorry.
I have decided against mirroring the swap space, and just using both the /dev/hda5 and /dev/hdb5 as swap space, so it is doubled.
OK
Is this a wise choice?
So that is where I am stuck. I cannot figure how to get past the "waiting for device /dev/md10 to appear......". I have tried "mkinitrd" and "cd /mnt; chroot /mnt; mkinitrd" to try and fix it, but to no avail.
Anyone have any ideas? Thanks for your help thus far!
Check /etc/mdadm.conf to make sure you really have such raid devices.
The file does not exist.
-- Joe Morris Registered Linux user 231871 running openSUSE 10.2 x86_64
-Brandon Carl -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
participants (1)
-
Brandon Carl