Bug#253069: debian-installer TC1 successful installation on Apple Powerbook G4
On Mon, Jun 07, 2004 at 12:53:55AM +0100, James Troup wrote:
> Package: installation-reports
>
> INSTALL REPORT
>
> Debian-installer-version: TC1
> uname -a: Linux elisha 2.4.25-powerpc #1 mer avr 14 15:38:38 CEST 2004 ppc GNU/Linux
> Date: 2004-05-06 10PM GMT
> Method: Installed off netinst CD
> [http://cdimage.debian.org/pub/cdimage-testing/sarge_d-i/powerpc/tc1/sarge-powerpc-netinst.iso]
>
> Machine: Apple Powerbook G4
> Processor: 1.5Ghz G4
> Memory: 512MB
> Root Device: 80GB IDE (TOSHIBA MK8026GAX)
> Root Size/partition table: ...
> %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
> /dev/hda
> # type name length base ( size ) system
> /dev/hda1 Apple_partition_map Apple 63 @ 1 ( 31.5k) Partition map
> /dev/hda2 Apple_Bootstrap untitled 19532 @ 64 ( 9.5M) NewWorld bootblock
> /dev/hda3 Apple_Bootstrap untitled 155281251 @ 19596 ( 74.0G) NewWorld bootblock
Wah. What happened there to give you two Apple_Bootstrap partitions? Is
that just what 'fdisk -l' says after installation?
Ah; perhaps you had / marked as bootable, due to a now-fixed bug in
partman-auto. That would also have made the yaboot-installer question
more confusing since it would've "helpfully" offered to overwrite / with
the bootstrap partition.
> (o) I hope the auto-partitioning logic varies between platforms
> because 10Mb is far too small for /boot on hppa[1] at least and
> possibly others?
Yeah. In fact, the 10MB partition you've got isn't /boot, but the
bootstrap partition, which is a powerpc-specific thing being the
partition where the yaboot bootloader gets installed. It's not actually
mounted anywhere.
Was there something that implied it was /boot?
> (o) yaboot asked me what partition to use. It'd be nice if it could
> have been told that by the auto-partitioning stuff?
It would, but parted doesn't give us accurate enough answers for that.
:-( The current implementation is basically a sufficiently-working
kludge which is unlikely to break but is a bit ugly; I'd love to replace
it with something better once something better has been devised (namely
fixing parted to have a better means of probing for Apple_Bootstrap
partitions, or else having an hfsutils-udeb).
Maybe excluding any partitions which are explicitly mounted would help.
> Conclusion: Great work, you guys. :-)
Cool. Thanks for testing!
--
Colin Watson [cjwatson@flatline.org.uk]
Reply to: