Bug#369965: [Yaird-devel] Bug#369965: Fixed.

[maximilian attems]

On Thu, Jun 22, 2006 at 11:26:38AM +0200, Jonas Smedegaard wrote:
> On Thu, 15 Jun 2006 01:50:33 +0530 Rohan Dhruva wrote:
> 
> > Well, very sorry, it was not a yaird problem at all. Somehow my
> > /dev/null in the static /dev had been replaced with a file, and hence
> > the errors. The problem did not occur with the initrd created by
> > initramfs-tools since it starts udev from within the initrd. Would the
> > be feasible for yaird ? Anyway, I am closing the bug, since yaird
> > works perfectly as it is expected to :) Sorry for the 'spam' ;)
> 
> Thanks for following up on this.
> 
> No, it really does not make sense for yaird to use udev: The very
> approach of yaird is to be smart at ramdisk build time, and exact at
> runtime - which is more or less the opposite of initramfs-tools+udev.

ok so lets go now for the other side of the view:
introduce errors on initramfs generation due to relying on exported
/sys and the strange parsing you do.
no error recovery on boot.

why you include dash on initramfs and don't use it?
unless some strange debug is set afair.

-- 
maks