My partition table looks like this... Disk geometry for /dev/sda: 0.000-239372.437 megabytes Disk label type: mac Minor Start End Filesystem Name Flags 1 0.000 0.031 Apple 2 0.031 1.031 hfs untitled boot 3 1.031 220001.031 ext3 untitled 4 220001.031 222049.031 hfs swap swap 5 222049.031 239372.437 ext3 untitled sda5 was originally free space, used as the root partition for the rawhide installation I just attempted; hda3 was not part of that system. The resulting /etc/yaboot.conf written by the installer contained this line: boot=/dev/sda2 /dev/sda4 This causes ybin/mkofboot to be slightly unhappy, and also the boot-device environment variable seems to have been set to hd:4, instead of hd:2.
Do you have the install logs handy? "boot=/dev/sda2 /dev/sda4" looks like booty thinks it's setting up a raid1 device for /boot, which is obviously really strange given what you've said here.
They didn't seem wonderfully enlightening, which is why I didn't attach them.
Created attachment 122372 [details] install.log
Created attachment 122373 [details] install.log.syslog
Created attachment 122374 [details] anaconda-ks.cfg
No kidding, they don't seem enlightening at all.
Seems to be a parted bug. It's detecting my swap partition as 'hfs', as shown in the original report. If I swapoff and then zero the partition, then mkswap and swapon again, parted no longer reports it as hfs; it reports it as swap.
Hm, no. Another rawhide install gave me a yaboot.conf saying 'boot=/dev/sda2 /dev/sda4' and turned my swap partition into HFS again.
Still happening in FC5t3.
Fixed in booty 0.78-1 Python keeps target list after for loop so we were getting duplicate entries. This reworks the logic so we avoid that. http://docs.python.org/ref/for.html