[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#358153: testing (etch): cannot boot on software raid



On Wed, 22 Mar 2006, Vegard Svanberg wrote:

> * maximilian attems <maks@sternwelten.at> [2006-03-21 14:37]:
> 
> > > The raid1 kernel module appears to be in the initrd image, but is not
> > > loaded. Loading it manually does not create /dev/md{012}. Grub has set
> > > root=/dev/md0.
> > 
> > can you provide further infor from the busybox shell:
> > * did the devices /dev/hdc*, /dev/hde* got created?
> 
> No. 

ok if the device don't exist, mdadm can't work.

> Additionally, dmesg displays something I'm not sure how to interpret:
> 
> ata1: SATA max UDMA/133 cmd 0x6800 ctl 0x6402 bmdma 0x5400 irq 74
> ata2: SATA max UDMA/133 cmd 0x6000 ctl 0x5802 bmdma 0x5408 irq 74
> ATA: abnormal status 0x7f on port 0x6807
> ata1: disabling port
> scsi0 : ata_piix
> ATA: abnormal status 0x7f on port 0x6007
> ata2: disabling port
> scsi1 : ata_piix

blaeh that looks like an race,
can you please post which modules are all loaded in cat /proc/modules

also more lines from dmesg would be interesting.

 
> > * if not was piix loaded? (cat /proc/modules)
> 
> Yes, loaded.
> 
> > * what was your cat /proc/cmdline
> 
> root=/dev/md0 ro

that looks fine.
 
> > * does cat /proc/mdstat show something when dropped in?
> 
> No (except from the usual "personalities: [raid1], unused devices:
> <none>").
> 
> > try to run the md boot script by hand:
> > . ./conf/initramfs.conf
> > . ./scripts/functions
> > . ./scripts/local
> > ./scripts/local-top/md
> 
> No errors (and no output).
> 
> > # check if /dev/md0 is there
> 
> Not there.
> 
> > mountroot
> 
> "/dev/md0 does not exist, dropping to a shell!"

yeah expected due aboves.

-- 
maks



Reply to: