proxmox crashed during restart

[machines]

hello there,

actually i have problem which is described several times in the forum.

we just restarted our fresh installation of proxmox ve 1.9-7 using an adapter 5405z
and for some reason the file system crashed.

fsck.ext3 says > no such such file .... ./dev/pve/root

the superblock could not be read ... you might try running ... e2fsck -b 8139 <device>

is there a way to get my vm-system saved before i do this ? i also tried to boot the
system using the cd and the debug-mode, but i cannot ssh into the machine.

thanks for your FAST answers. ;-)

greetings marco.

 

[machines]

hello again,

due to the fact that we need the system to be running at our customer, we decided to make
a fresh installation of proxmox and create a new vm-machine.

now i get the following message at the console:

CE: hpet5 increasing min_delta_ns to 22500 nsec
CE: hpet2 increasing min_delta_ns to 15000 nsec

is that perhaps the problem that crashed my system this morning?
can someone explain that to me?

the system is a fujitsu TX150S7 with two 1 TB HDD on an adaptec 5405Z,
pve-manager /1.9/6542,
Linux 2.6.32-6-pve#1 SMP

thanks for your help.

greetings marco.

 

[udo]

hello there,

actually i have problem which is described several times in the forum.

we just restarted our fresh installation of proxmox ve 1.9-7 using an adapter 5405z
and for some reason the file system crashed.

fsck.ext3 says > no such such file .... ./dev/pve/root

the superblock could not be read ... you might try running ... e2fsck -b 8139 <device>

is there a way to get my vm-system saved before i do this ? i also tried to boot the
system using the cd and the debug-mode, but i cannot ssh into the machine.

thanks for your FAST answers. ;-)

greetings marco.

Hi,
not an fast answer... but the error don't show that the filesystem crashed. The logical volume pve-root is'nt available in this time.
Linux boot the kernel in /boot and change the root to (in this case) /dev/pve/root - to do that the driver for that storage must be loaded at this time (inside the kernel or in the initrd).

In you case, it's looks like that the kernel don't find the disk, or something wrong with the lvm.

Udo

 

[machines]

Hello,

I just have to pick up this thread again because my customer was in the mood to switch the server the hard way off ...

Now i started fsck as described in this thread:
http://forum.proxmox.com/threads/3043-root-password-for-maintenance-or-Control-D-to-continue

Running fsck /dev/pve/root & fsck /dev/pve/data corrects some errors but restarting the system still forces the fsck to run again,
which then dies with code 16 (exit status 16).

Is there perhaps an idea for me how I can get the system working again without a new installation ?

Thanks for your answers.

Greetings Marco.