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

Bug#814288: Generates corrupt (?) initrd images



Control: tag -1 moreinfo unreproducible

On Tue, 2016-02-09 at 17:51 -0500, Christopher David Howie wrote:
> Package: initramfs-tools
> Version: 0.120
> 
> I have a rather odd case where update-initramfs sometimes generates good
> initrds and sometimes it does not.  I'm not sure if the bug lies here or
> in the kernel, but after sinking hours into investigating this issues
> I'm very confident there is a bug somewhere.  I figure I will start with
> this package and see where we get.
> 
> Kernel info, for the record:
> 
> -----
> Linux debian-workstation 3.16.0-4-amd64 #1 SMP Debian
> 3.16.7-ckt20-1+deb8u3 (2016-01-17) x86_64 GNU/Linux
> -----
> 
> Every initrd generated on the target machine with "update-initramfs -k
> all -u" causes a kernel panic on the next boot:

Which boot loader are you using?

[...]
> I captured an example of a working and a broken initrd.  If I extract
> them (gunzip < $file | cpio -i -dv) and diff the two trees, only one
> file reports as being different: /etc/boottime.kmap.gz .  If I gunzip
> this file in both trees, diff reports no differences.
[...]

diffoscope (https://try.diffoscope.org/) agrees with that.

This works for me:

    kvm -m 1G -kernel /boot/vmlinuz-3.16.0-4-amd64 -initrd initrd.img-3.16.0-4-amd64.broken

Of course the initramfs can't find your root device, but it eventually
drops to a rescue shell.  Which I can't use because... is that 
Dvorak layout?

Ben.

-- 
Ben Hutchings
Beware of bugs in the above code;
I have only proved it correct, not tried it. - Donald Knuth

Attachment: signature.asc
Description: This is a digitally signed message part


Reply to: