Topic: Superblocks gone poof...

[nickw]

I have 7.5.1 installed on a fujitsu rx100 s5, 2 two 250gb sata discs, everything done pretty much following the defaults for the install, and everything worked fine until this last weekend, when the power went away, and then as far as I can tell, came back again a couple times after the UPS was drained..

Anyhow the system came up this morning, and while mounting lvm says "kernel panic: discs not syncing, trying to kill init".

After booting from a Live CD (the 7.5.1 rescue option doesn't find any system at all installed) I've been able to check md0 and md1, but not md2...

It seems the first sda2/sdb2 superblock got hosed (on both disks) and there aren't any backups in any of the normal places.

I've tried fsck -b 8193 (and more than a few others) but it doesn't seem to make any difference..

Does anyone know if Centos/contribs, uses some strange options for its ext3 installs?

BTW, for some reason, testdisk sees the partitions as deleted..

Nick

[Daniel B.]

SME Uses LVM by default, not ext3 directly on the RAID device. So, you first need to check if LVM is stil available:

Code: [Select]

vgscan
vgchange -ay main

Then, if something is found, you should be able to check the FS on top of it

Code: [Select]

e2fsck /dev/main/root


[nickw]

I saw two resources under the main group, root and swap, I'll see if it lets me scan either one..