Red Hat Bugzilla – Bug 211423
invalid partition table: x86_64 boot failure on mainboard SATA raid
Last modified: 2007-11-30 17:11:46 EST
Description of problem:
Kernel stops during boot with a panic.
I get error message: Invalid partition table on /dev/mapper/sil_afbhabcfchdip1
Version-Release number of selected component (if applicable):
(2.6.17-1.2187_FC5 works fine)
I had used my mainboard's BIOS to create a hardward RAID 0 over two Serial ATA
160 GB disk drives. When I installed FC4 on that machine, I installed it to that
When I updated to FC5 6 months ago, I had to change my /etc/fstab entries to use
that cryptic /dev/mapper/sil_afbhabcfchdip1 name to make the system boot.
Peter, mkinitrd bug? or device-mapper?
Still broken with kernel-2.6.18-1.2239.fc5
I plan to shorty upgrade this machine to FC6. Hopefully that'll work.
Please let me know if you want me to do anything to help you diagnose this
issue, before I upgrade.
Still broken with latest fc5 kernel update 2257
FWIW, the error message also says "invalid signature 0".
Do you plan to work on this bug soon and need my help in testing?
Or should I go ahead and try to upgrade to FC6?
Maybe the upgrade won't work anyway, if the bug is still present in FC6...
I would like to upgrade the machine to FC 6 this weekend, sorry, I'm not able to
postpone it more.
I tried to do a DVD-based update on that machine using the Unity Respin FC 6 iso
The update failed. It complained about an invalid partition table on that disk.
Well, at least the FC6 kernel still used the same hardware device in
I wrote 200 megs of zeros to the device and did a fresh install, to trigger the
installer to init the partition table. But after install, GRUB did not load.
I finally decided to remove the hardware RAID configuration. I unplugged the
SATA disks from the RAID-onboard-plugs, and plugged them to the standard SATA
plugs, and configured a software RAID.
As I am no longer able to provide information and what is causing this strange
behaviour on my machine, I'm closing this bug.