Description of problem: using --append in a kickstart file will remove console=ttyS0,115200 Version-Release number of selected component (if applicable): anaconda 12.15 (the one we have in F12 Alpha) How reproducible: always Steps to Reproduce: 1. install 2 machines as follows machine A: + boot into kickstart with console=ttyS0,115200 + use a kickstart file containing: bootloader --location=mbr machine B: + boot into kickstart with console=ttyS0,115200 + use a kickstart file containing: bootloader --location=mbr --append="banana" 2. note, both kickstarts have a one liner in %post that erases 'rhgb quiet' from grub.conf as I need to be able to see what the machines do at boot. perl -p -i -e 's/ rhgb quiet//g' /boot/grub/grub.conf 3. let kickstart install finish 4. boot the machines into the freshly installed OS Actual results: bootloader --location=mbr ends up being: Kernel command line: ro root=/dev/mapper/vg_oma01-lv_root console=ttyS0,115200 LANG=en_US.UTF-8 SYSFONT=latarcyrheb-sun16 KEYTABLE=us bootloader --location=mbr --append="banana" ends up being: Kernel command line: ro root=/dev/mapper/vg_oma03-lv_root banana LANG=en_US.UTF-8 SYSFONT=latarcyrheb-sun16 KEYTABLE=us Expected results: if with --append we keep LANG, SYSFONT and KEYTABLE, I would expect it to also keep my console= This is especially annoying on machines that have only serial output Additional info: rawhide today has anaconda 12.25 but that is not installable for me, AttributeError: 'module' object has no attribute 'F12_AutoPart', do tell if you want this tested against a later anaconda or if you need me to write a new rawhide ks file just for this test (seems the autopart is not being digested, so worst case I manually craft partition info for the test)
Does updates=http://clumens.fedorapeople.org/524004.img fix this problem? You will definitely need to test this against rawhide.
anaconda 12.27 bails out with PartedException: Unable to determine the dm type of /dev/dm-0. Guess my rawhide mirror is not current enough, will try again Thu or Fri
No, there's something strange going on with anaconda/parted right now and detecting existing LVs. You can work around this problem by clearing out the disk, but that may not be possible for you.
(In reply to comment #1) > Does updates=http://clumens.fedorapeople.org/524004.img fix this problem? Yes it does. Thank you. Do tell once the fix went into anaconda and I can re-test if you want.
Thanks for testing. This will be fixed in the next build of anaconda.
confirm, anaconda-12.32-1 fixes this bug without needing to use the updates.img from Comment #1 Thanks Chris