Bug#343042: [Yaird-devel] Bug#343042: #343042: linux-image-2.6.14-2-686: Boot aborts with message '/bin/cat

2005-12-14 Thread Erik van Konijnenburg
On Wed, Dec 14, 2005 at 10:42:58AM +, Richard Antony Burton wrote:
 Package: yaird
 Version: 0.0.12-1
 Followup-For: Bug #343042
 
 FYI the above workaround does not work for machines with SATA drives. I 
 couldn't
 find a combination of modules that would get it to boot.
 
 Switching back to previous yaird (0.0.11-12) fixes the problem on both PATA 
 SATA systems.

Hmm, you're the first to report problems with SATA, and considering
what was changed between 0.0.11 and 0.0.12, I would not expect any
problems in this area.

To help debugging, could you post your version of /etc/yaird/Default.cfg,
plus the output of yaird -v for the working version and the broken version?

Thanks,
Erik


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]



Bug#343042: [Yaird-devel] Bug#343042: #343042: linux-image-2.6.14-2-686: Boot aborts with message '/bin/cat

2005-12-12 Thread Jonas Smedegaard
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

On Mon, 12 Dec 2005 16:24:26 +
Richard Antony Burton [EMAIL PROTECTED] wrote:

 Considering the potential for harm I wonder if there could be
 something done to mitigate this happening again in future. Avoiding
 releasing yaird at the same time as a minor kernel update would have
 helped, because the bug would have only been hit by a small number
 before it could be fixed. However perhaps the more important issue is
 that of releasing minor kernel updates that overwrite the existing
 kernel, this is obviously vunerable to more problems than just a
 badly timed yaird break. That's obviously an issue for the kernel
 maintainers.

One appreach that you could do locally is to copy your /boot/initrd.*
to /boot/initrd.*.good or whatever. Then if problems occur (and you are
using GRUB) you can edit the GRUB menu item to load the alternative
image.

Such approach is tightly related to the capabilities of your
bootloader, while the ramdisk image is generated from your kernel
package postinst. Enhancing this for the distribution (rather than
each admin doing local hacks depending on their choice of bootloader
+kernel+arch) probably requires something like
http://wiki.debian.org/FlexibleKernelHandling


 p.s. before anyone tells me, i have been reminded of why it's a good
 idea to keep a known working kernel on the box ;-)

Ahh - too late I guess (or look the other way when those other two
replies of mine hits your inbox) :-P


 - Jonas

- -- 
* Jonas Smedegaard - idealist og Internet-arkitekt
* Tlf.: +45 40843136  Website: http://dr.jones.dk/

 - Enden er nær: http://www.shibumi.org/eoti.htm
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.2 (GNU/Linux)

iD8DBQFDnaw5n7DbMsAkQLgRAuIsAJ91bT/Wqc+Ft59qp2y6nA5MrYtYowCfZfVR
8HWNh+MrNWtzfRKIz25f/UI=
=iWOO
-END PGP SIGNATURE-