LinuxQuestions.org

LinuxQuestions.org (/questions/)
-   Linux - Hardware (http://www.linuxquestions.org/questions/linux-hardware-18/)
-   -   invalid metadata checksum error, (I think) (http://www.linuxquestions.org/questions/linux-hardware-18/invalid-metadata-checksum-error-i-think-741819/)

zogthegreat 07-21-2009 07:54 PM

invalid metadata checksum error, (I think)
 
Hi everyone,

I am have a problem with a back-up machine that I had built to hold my data while I rebuilt my server. The hardware is as follows:

motherboard: ECS k7S5A w/ AMD Athlon 1800 cpu
memory: 1 gig PC133 SDRAM
HD: 1 WD 120 gig IDE + 1 Seagate 250 gig IDE in a JBOD
NIC: Realtech gigbit
OS: CentOS 5.2

The system setup and ran fine for a few months. Had a power failure in my neighborhood and when I rebooted the machine, the system loads until

unmounting old /dev
unmounting old /proc
unmounting old /sys
exec of init (/sbin/init) fails!!!: No such file or directory
Kernel panic - not syncing: Attempted to kill init!

From there my system locks and the lights on the keyboard start flashing.

I tried to use a CentOS dvd in "linux rescue" mode. The system displays a message in the background "invalid metadata checksum in area 1 /dev/hda" then loads the screen that tells me to mount the system image with the "chroot /mnt/sysimage". When I try to do this, I get the following error, "chroot cannot run command bin bash' no such file or directory".

Does anyone have any suggestions?

Thanks

zog

mostlyharmless 07-22-2009 02:23 PM

Sounds like your JBOD is busted to me. I presume it is firmware set up by the motherboard. You might want to check that the firmware/BIOS is still set to JBOD, as sometimes a power out can reset the BIOS to default, which is not JBOD. If you're real lucky, it might have diagnostics of your JBOD array.

If that doesn't do it, you'll need to mount the rescue CD with possibly dmraid to mount your array (though I don't know if it does JBOD) and possibly testdisk to see if you can rescue your data. I'm not familiar with the Centos rescue CD; you might want to try something independent and fully featured; e.g. Knoppix. Could be one of the drives or the mobo/controller has a problem (or not); you'll have to investigate....

That's where I'd start; I hope that helps.

zogthegreat 07-22-2009 05:18 PM

hi mostlyharmless,

I solved the problem with the following:

I downloaded "System Rescue CD" from http://www.sysresccd.org. Although I used this cd, I probably could have used Knoppix.

I started with:

root@sysresccd /root % lvm pvscan

PV /dev/sdb1 VG VolGroup00 lvm2 [111.78 GB / 0 free]
PV /dev/sda2 VG VolGroup00 lvm2 [232.78 GB / 0 free]

I than ran the following:

root@sysresccd /root % fsck -f /dev/VolGroup00/LogVol01

I was required to constantly hit the y key in order to continue. I could/should have used the -P argument, as it was, I had to set a weight on the "y" key. After 7 hours, I got the following output:

/dev/VolGroup00/LogVol00: ***** FILE SYSTEM WAS MODIFIED *****
/dev/VolGroup00/LogVol00: 89641/90079232 files (4.1% non-contiguous), 85329392/90062848 blocks

I then did this:

root@sysresccd# lvm lvscan

ACTIVE '/dev/VolGroup00/LogVol00' [343.56 GB] inherit
ACTIVE '/dev/VolGroup00/LogVol01' [1.00 GB] inherit

root@sysresccd /root % mount /dev/VolGroup00/LogVol00 /mnt/part1
root@sysresccd /root % mount /dev/VolGroup00/LogVol01 /mnt/part2
root@sysresccd /root % cd /mnt/part1
root@sysresccd /mnt/part1 % ls
bin dev home lost+found misc net proc sbin srv tmp var
boot etc lib media mnt opt root selinux sys usr

All of my data was there and I am able to copy it over to my new server using scp.

Hope this helps someone else.

zog


All times are GMT -5. The time now is 02:55 PM.