Red Hat Bugzilla – Bug 450479
Anaconda unable to recognise grub, "replace" crashes
Last modified: 2008-09-23 10:49:55 EDT
Description of problem:
Anaconda was (apparently) unable to find/recognize my grub bootloader and/or
/boot partition. As a result, it treated it as if the system had a third-party
boot loader and refused to update the config. I had an option to replace the
boot loader or skip config altogether. I selected replace, causing a traceback
and crash, which is attached.
Version-Release number of selected component (if applicable):
Anaconda 184.108.40.206, according to the traceback.
It happened every time until I told anaconda to skip instead (at least 2 or 3
Steps to Reproduce:
1. Somehow get into a state where anaconda doesn't recognize a perfectly valid
2. Upgrade the installation.
3. See that anaconda doesn't recognize /boot, but you can replace or skip
4. Select replace, click next
5. Watch the installer crash and offer to save the traceback/debug information.
3. Anaconda gives you the option to update, replace, or skip
4. Selecting any of the provided options doesn't crash, but does what it says it
Barring that, anaconda should never crash, ever (especially not for options it
gives me). I'm just glad it hadn't modified my disk yet.
This occurred when I updated my gateway machine (which has my DNS, & DHCP
server, hence the manual/IP setup) from Fedora 7 to Fedora 9. The partitioning
of the drive is from the default (or near-default) partitioning of an earlier
install, so I don't know of anything that would cause the installer to act like
$ df -h
Filesystem Size Used Avail Use% Mounted on
72G 9.6G 59G 15% /
/dev/sda1 99M 19M 76M 20% /boot
tmpfs 124M 0 124M 0% /dev/shm
# fdisk -l
Disk /dev/sda: 80.0 GB, 80000000000 bytes
255 heads, 63 sectors/track, 9726 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Disk identifier: 0xd0f4738c
Device Boot Start End Blocks Id System
/dev/sda1 * 1 13 104391 83 Linux
/dev/sda2 14 9726 78019672+ 8e Linux LVM
Hope this helps.
Created attachment 308665 [details]
Traceback generated by anaconda
I do believe I've fixed up the problem with not identifying the existing bootloader configuration for F10 Beta.