kernel panic after vg00 not found on boot

[CReichert]

Hello all,

I have c7000 blade enclosures, bl460/680 blades, cisco 9124 fibre switches, and an emc clariion cx3 disk array.

We have 75 blades functioning in this environment.

I tried to move a linux boot lun from one bl460 in enclosure x to enclosure y. To accomplish this, I used EMC's Navisphere to make the lun mapping change. i removed the original wwn and added the new wwn. Zoning was already in place and the new server does see the proper lun.

On starting the bl460 in enclosure y, Red Hat boots, but ends up panic'ing when it can't find vg00. Here's the relevant text:

.
Decompressing Linux...done.
Booting the kernel.
Red Hat nash version 4.2.1.13 starting
Reading all physical volumes. This may take a while...
No volume groups found
Volume group "VolGroup00" not found
ERROR: /bin/lvm exited abnormally! (pid 588)
mount: error 6 mounting ext3
mount: error 2 mounting none
switchroot: mount failed: 22
umount /initrd/dev failed: 2
Kernel panic - not syncing: Attempted to kill init!

I move the lun back to the original server and it works fine.

The only other change is that the lun is now presented down a different path from the clariion (spa2 instead of spa5). Seems like an emc issue though i'm not sure how to describe the issue so that i don't get the traditional/expected run around from support.

Anyone have an idea of why linux would boot off it's disk then not find it's disk? Same OS, identical bl460 configuration, identical enclosure hardware.

Thanks - any insight would be fantastic!

Chris

[Zeev Schultz]

Well,usually during first stage of Linux system boot, it loads /sbin/initrd* image and run set of commands - one of these is "lvm vgscan -ignorelockingfailures".

Anyway,what precedes to "lvm vgscan" and "lvm vgchange -ay" are udev and dev-mapper commands.

Question is if you have disks WWID information under /dev/mapper/... - i.e disks devices are built using WWID information.

Another question would be if your physical volumes are properly presented. This can be checked from problematic server using install media - you'll get to your LUNS during filesystems setup stage. So I'd put an install media and try to check if LUNs are seen.

[Andrew Cowan]

This will happen if you have moved/copied the Volume Group from another disk e.g. built it on hd0 and then moved it to hd1.

Y0u get a bit further if you edit your Grub Loader file and specify the correct root and kernel location e.g. hd(1,0)