Hi,
I have a machine with 4 160G disks in a raid-0 configuration. Now
after upgrading the hardware, all of a sudden raidstart can't find the
raid-superblocks anymore. Invalid magic.
I'm suspecting that it might be that the superblock was overwritten
with data or something like that. Does anybody know of a bug like
this?
We're running kernel-2.4.16 with andre's IDE patches for the large
disks.
I'll see if I can find the "magic" anywhere on the disk....
Roger.
--
** [email protected] ** http://www.BitWizard.nl/ ** +31-15-2137555 **
*-- BitWizard writes Linux device drivers for any device you may have! --*
* There are old pilots, and there are bold pilots.
* There are also old, bald pilots.
Rogier Wolff wrote:
>
> Hi,
>
> I have a machine with 4 160G disks in a raid-0 configuration. Now
> after upgrading the hardware, all of a sudden raidstart can't find the
> raid-superblocks anymore. Invalid magic.
A shot in the dark: any chance the new hardware sees a different
geometry so the RAID superblock location (which is at the end of
the disk) has shifted?
--
Eyal Lebedinsky ([email protected]) <http://samba.org/eyal/>
Rogier Wolff <[email protected]> :
[...]
> I have a machine with 4 160G disks in a raid-0 configuration. Now
> after upgrading the hardware, all of a sudden raidstart can't find the
> raid-superblocks anymore. Invalid magic.
>
> I'm suspecting that it might be that the superblock was overwritten
> with data or something like that. Does anybody know of a bug like
> this?
Cut some configuration options behind "Advanced partition selection" ?
I experienced the same symptoms 2 days ago (IDE+1 raid autodetect partition
+raid1+2.4.18-pre7-"akpm"). No superblock corruption here but a side effect
of a rescue installation in swap space.
--
Ueimor
Rogier Wolff wrote:
>
> Hi,
>
> I have a machine with 4 160G disks in a raid-0 configuration. Now
> after upgrading the hardware, all of a sudden raidstart can't find the
> raid-superblocks anymore. Invalid magic.
>
> I'm suspecting that it might be that the superblock was overwritten
> with data or something like that. Does anybody know of a bug like
> this?
>
> We're running kernel-2.4.16 with andre's IDE patches for the large
> disks.
>
> I'll see if I can find the "magic" anywhere on the disk....
Couldn't find them .
Moved the disks to the old machine, got my RAID back. But I still
can't find the magic numbers for the rais-sb anywhere.
Could it be that 48bit addressing doesn't work for PIO mode?
Either KERNEL bug, drive bug or by spec?
I found that the drives were horribly slow when I tried searching
them. So I turned DMA on only after having failed to mount the raid,
and started to "search" through the whole disk.
Can someone seek to the 8k before end of a raid-partition, and dump
the remaining part of the disk, so that I can be sure to know what it
looks like? That should include the raid-sb, right?
Roger.
--
** [email protected] ** http://www.BitWizard.nl/ ** +31-15-2137555 **
*-- BitWizard writes Linux device drivers for any device you may have! --*
* There are old pilots, and there are bold pilots.
* There are also old, bald pilots.
Rogier Wolff wrote:
> I have a machine with 4 160G disks in a raid-0 configuration. Now
> after upgrading the hardware, all of a sudden raidstart can't find the
> raid-superblocks anymore. Invalid magic.
I forgot to send out the "resolution" of this.
Turns out it has to do withg 48bit addressing and the Promise
controller that I wanted them on.
They work great now on the controller on the motherboard. Thanks
to all who helped!
Roger.
--
** [email protected] ** http://www.BitWizard.nl/ ** +31-15-2137555 **
*-- BitWizard writes Linux device drivers for any device you may have! --*
* There are old pilots, and there are bold pilots.
* There are also old, bald pilots.