Hi there!
I have been reading this forum a long time, but now i have a problem where I haven’t found a solution. So maybe some of you have an idea.
I have set up a new system with opensuse 11.1.
Layout was:
sda1+sdb1 as raid 1 (md0) in /boot
sda3+sdb3 as raid 1 in swap
sda5 as lvm for pv_root as /
sda6 as lvm for pv_home in /home
This worked fine. Grub loads correctly and i can boot grub from both drives - sda1 and sdb1.
Then i created a raid 1 (/dev/md2) with sdb5 and one missing device and added it to pv_root.
Then i used pvmove to move the data from sda5 to md2. This worked and i could reboot.
I could remove sda5 from the lvm but after that i could not reboot the system.
Grub still works but then there come hundreds of messages like that:
“Waiting for device /dev/pv_root_lv_root to appear. Volume group not found. Fallback to sh?”
When i type y those messages repeat until a mini-bash appears.
I can’t do anything with that bash (or shell?).
When i reboot with the Live Cd, all those lvm are still there (and can mount them and everything is correct on them).
So i thought that maybe the kernel is not correct. I had this sone times before with other systems and there an mkinitrd was the solution.
Now I mount lv_root to /mnt/root and md0 to /mnt/root/boot. Then chroot /mnt/root and then mkinitrd.
But then there comes the error
“node name not found.
Root device (/dev/dm-1) not found.”
And now i don’t know what to do…
I have searched the web for a while but found nothing helpful.
When you say what config file i should add, i can reboot to the Live System and post something from there. The “real” systems logs (from that minibash) i have to make photos of and type them by hand - and i can only see the (approximate) 20 last lines - but that seems not very usable…
Thanks alot for reading so far and i would be glad if somebody had an idea.
Now I mount lv_root to /mnt/root and md0 to /mnt/root/boot. Then chroot /mnt/root and then mkinitrd.
But then there comes the error
“node name not found.
Root device (/dev/dm-1) not found.”
Are you remounting /dev before you chroot?
mount -o bind /dev /mnt/root/dev
Are you sure mkinitrd is the correct solution? It seems more likely to me that grub is expecting / to be on the old logical volume. What is your root= in /boot/grub/menu.lst?
I am NOT sure, that mkinitrd is the correct solution. But i didn’t had an other ideas
My root in /boot/grub/menu.lst:
root=/dev/pv_root/lv_root
(When i reboot with the live cd, /dev/pv_root/lv_root contains the root file system)
You said “the OLD logical volume”, but the logical volume is still the same - it only contains an other harddisk? Do i have to tell grub, that the lv contains now an other hdd (concrete: a raid device)?
md2 is a pv, but see below:
(I wanted to post these outputs in my opening post, but there i had no way to get them without typing them by hand from a photo. Now found a way to get the outputs from the Live System):
# pvdisplay
File descriptor 4 left open
File descriptor 9 left open
File descriptor 10 left open
File descriptor 11 left open
File descriptor 15 left open
--- Physical volume ---
PV Name /dev/md3
VG Name pv_home
PV Size 675,01 GB / not usable 2,62 MB
Allocatable yes
PE Size (KByte) 4096
Total PE 172802
Free PE 8964
Allocated PE 163838
PV UUID 3Dtdbs-oV0E-a6Y8-H7Kc-Q5ue-mFSy-WQPE8j
--- Physical volume ---
PV Name /dev/md2
VG Name pv_root
PV Size 19,43 GB / not usable 704,00 KB
Allocatable yes
PE Size (KByte) 4096
Total PE 4975
Free PE 2416
Allocated PE 2559
PV UUID BZRMQQ-xpu3-V6Ip-nwQa-DToy-Vgnu-pNp4F1
"/dev/sda5" is a new physical volume of "10,00 GB"
--- NEW Physical volume ---
PV Name /dev/sda5
VG Name
PV Size 10,00 GB
Allocatable NO
PE Size (KByte) 0
Total PE 0
Free PE 0
Allocated PE 0
PV UUID 92zxlN-dxZM-NBbc-Zhc0-FDbs-3ze9-8URcVn
"/dev/sda6" is a new physical volume of "639,99 GB"
--- NEW Physical volume ---
PV Name /dev/sda6
VG Name
PV Size 639,99 GB
Allocatable NO
PE Size (KByte) 0
Total PE 0
Free PE 0
Allocated PE 0
PV UUID crlH17-VBl6-e6xJ-3364-Tq30-xjdc-dAlZj0
# lvdisplay
File descriptor 4 left open
File descriptor 9 left open
File descriptor 10 left open
File descriptor 11 left open
File descriptor 15 left open
--- Logical volume ---
LV Name /dev/pv_home/lv_home
VG Name pv_home
LV UUID Ed8kbX-g37K-oTiq-fAcM-kZtm-2ugo-dFcwU2
LV Write Access read/write
LV Status available
# open 1
LV Size 639,99 GB
Current LE 163838
Segments 1
Allocation inherit
Read ahead sectors auto
- currently set to 256
Block device 253:0
--- Logical volume ---
LV Name /dev/pv_root/lv_root
VG Name pv_root
LV UUID CNyuK1-DslE-WqUO-Wb9j-VXVv-6QFa-dc9eyA
LV Write Access read/write
LV Status available
# open 1
LV Size 10,00 GB
Current LE 2559
Segments 1
Allocation inherit
Read ahead sectors auto
- currently set to 256
Block device 253:1
# vgdisplay
File descriptor 4 left open
File descriptor 9 left open
File descriptor 10 left open
File descriptor 11 left open
File descriptor 15 left open
--- Volume group ---
VG Name pv_home
System ID
Format lvm2
Metadata Areas 1
Metadata Sequence No 7
VG Access read/write
VG Status resizable
MAX LV 0
Cur LV 1
Open LV 1
Max PV 0
Cur PV 1
Act PV 1
VG Size 675,01 GB
PE Size 4,00 MB
Total PE 172802
Alloc PE / Size 163838 / 639,99 GB
Free PE / Size 8964 / 35,02 GB
VG UUID 976Nev-dPpE-5wIa-Qpi3-1zbN-aKyH-yojpyt
--- Volume group ---
VG Name pv_root
System ID
Format lvm2
Metadata Areas 1
Metadata Sequence No 7
VG Access read/write
VG Status resizable
MAX LV 0
Cur LV 1
Open LV 1
Max PV 0
Cur PV 1
Act PV 1
VG Size 19,43 GB
PE Size 4,00 MB
Total PE 4975
Alloc PE / Size 2559 / 10,00 GB
Free PE / Size 2416 / 9,44 GB
VG UUID 1DSddC-mk7c-hnid-wOMV-e353-x9pE-T6rVlc
# cat boot/grub/menu.lst
# Modified by YaST2. Last modification on Sa Jan 10 20:24:09 CET 2009
default 0
timeout 8
gfxmenu (hd0,0)/message
##YaST - activate
###Don't change this comment - YaST2 identifier: Original name: linux###
title openSUSE 11.1
root (hd0,0)
kernel /vmlinuz-2.6.27.7-9-default root=/dev/pv_root/lv_root resume=/dev/disk/by-id/ata-SAMSUNG_HD753LJ_S13UJ1KPC19772-part3 splash=silent showopts vga=0x348
initrd /initrd-2.6.27.7-9-default
###Don't change this comment - YaST2 identifier: Original name: failsafe###
title Failsafe -- openSUSE 11.1
root (hd0,0)
kernel /vmlinuz-2.6.27.7-9-default root=/dev/pv_root/lv_root showopts ide=nodma apm=off noresume edd=off powersaved=off nohz=off highres=off processor.max_cstate=1 x11failsafe vga=0x348
initrd /initrd-2.6.27.7-9-default
I have tried that and went a bit further than my own try:
# chroot ../root
linux:/> mkinitrd -f lvm2
Kernel image: /boot/vmlinuz-2.6.27.7-9-default
Initrd image: /boot/initrd-2.6.27.7-9-default
Root device: /dev/pv_root/lv_root (mounted on / as ext3)
Kernel Modules: hwmon thermal_sys processor thermal fan jbd mbcache ext3 dm-mod edd dm-snapshot dock scsi_mod libata sata_nv crc-t10dif sd_mod usbcore ohci-hcd uhci-hcd ehci-hcd ff-memless hid usbhid raid0 raid1 xor async_tx async_memcpy async_xor raid456 linear
Features: dm block usb md lvm2 resume.userspace resume.kernel
Bootsplash: openSUSE (1024x768), openSUSE (1400x1050)
29863 blocks
cat: /proc/misc: No such file or directory
grep: /proc/misc: No such file or directory
/proc/devices: fopen failed: No such file or directory
Is device-mapper driver missing from kernel?
/proc/devices: fopen failed: No such file or directory
Failed to set up list of device-mapper major numbers
For your mkinitrd error, you’ll need to remount proc as well. Honestly, most of my recent chroot experience is with RHEL rescue mode, which takes care of this step for you. In any event, the command to remount proc is:
mount -t proc /proc/ /mnt/root/proc
I’m using LVM / over raid5 and I didn’t have to make a new initrd. But I installed it this way, so I don’t know if the initrd is created during install, or if they just use a generic one that covers pretty much everything (I assume the latter).
Based on a quick look at your output, it appears that it should just work. Are you able to assign any of the new extends to your root LV?
Is the data on md2 valid? Can you mount it alongside /dev/sda5 and confirm that it is ok?
This is just a guess (until I get a chance to experiment in a virtual machine) but could it be that /dev/sda5 is still showing up as a PV be an issue (in that it’s conflicting with your new PV)? Maybe you need to pvremove it (although I’d verify that the new device has valid data first).
I just tried that (mounting proc) yesterday after my last post, because i got an error with missing proc. But i used --bind instead of -t proc. But I also got an error and booting after that did not work - still the same problem. Today morning i read your post and tried the mkinitrd again with your type of mount - it also finished with the same error BUT: After a reboot everything works fine now lol! ?! I have done (i think so) nothing different than yesterday.
Yes - that was all ok. From the live system everything acted as it should (raid, lvm). So obviously it was not a problem with the lvm or the raid itself, because now it works without changes on raid or lvm. Maybe mkinitrd did change something before it exited with an error? I don’t know. But many thanks for you tipps and statements!
And have a nice evening
Till next time
hobbes777
and i would be glad if somebody had an idea.
…