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

Re: Bug#564150: Is Debian on PowerPC dead? [Re: Bug still present [Re: Bug#562575: installation-reports: Buisnesscard PowerPC installer loops with Segmentation faults]]



Hi,

I ran the debian-installer in another tty and here is the output (I had to wrote it by hand due to a dhclient error)

~ # set -x
~# . /sbin/debian-installer
+ set -e
+ export MENU=/usr/bin/main-menu
+ . /lib/debian-installer.d/S20speakup
+ lsmod
+ grep -q speakup_
+ . /lib/debian-installer.d/S30term
+ . /lib/debian-installer/detect-console
+ [ -z  ]
+ readlink /proc/self/fd/0
+ TERM_TYPE=virtual
+ TERM_FRAMEBUFFER_TRY=yes
+ export TERM_TYPE
+ . /lib/debian-installer.d/S35framebuffer-linux
+ [ -n yes ]
+ [ -e /dev/fb0 ]
+ debconf-get debian-installer/framebuffer
Segmentation fault
[snip]
+ debconf-get debian-installer/theme
Segmentation fault
+ theme=
~ #


The only powerpc computer that I have (PS3) is the one that I'm trying to install :(

maybe Rick had other machines previously installed and he could run the cdebconf powerpc version.

--
Julian.

On Sat, Jan 16, 2010 at 7:38 PM, Ian Campbell <ijc@hellion.org.uk> wrote:
On Sat, 2010-01-16 at 17:56 +0800, Julian Hernandez Gomez wrote:
> * Problem seems to be related to the changes made to cdebconf 0.146
> (debconf-get segfault)

Can you repro in a non-d-i environment with the non-udeb cdebconf? Or
maybe by unpacking the cdebconf udeb onto your system? Or by unpacking
an installer initrd and chrooting? (any of these will make debugging a
lot easier).

0.145 was r59369 in d-i svn and 0.146 was r61813 so I guess you could
have a poke/bisect through there. 59369-61813 includes lots of
non-cdebconf changes, and many of the cdebconf changes are l10n ones
which I reckon can be ignored on a first pass, leaving only around 20
actual commits, assuming 0.145 is ok.

Ian.
--
Ian Campbell

"It's the best thing since professional golfers on 'ludes."
               -- Rick Obidiah


Reply to: