I'm just installing RH7 on an older P166 box with two IDE disks. One
2GB Quantum Fireball ST2.1A and a 6GB IBM disk. In the text installer's
partition list, both disks appear with the same parameters, it seems.
I didn't think much of it when partitioning the 2GB disk and so
partitioned it witr a 6GB ext2 partition. While formatting the kernel
went into a loop of errors. After rebooting, the kernel still couldn't
access the disk on account of the MBR being bogus, it seems.
After realizing what was wrong I only put a 2GB partition on the disk
(now showing 4GB free) and the installation seems to progress without
I don't know if this could be a jumper problem of sorts, and I don't really
feel like playing with it. Life's too short. But I figured I would file
a bug on it - if you can/care to reproduce, fine. If not, just nuke the
FYI /proc geometry reads:
The text the kernel prints out about the disk during booting looks correct.
Did you try to partition with Disk Druid or fdisk?
Could you include the contents of /proc/ide/<drivedevice>/geometry for both
I used fdisk. Below are three files: hda geometry, hdb geometry, fdisk -l
thinktwice:> cat /tmp/a
thinktwice:> cat /tmp/b
thinktwice:> cat /tmp/c
Disk /dev/hda: 255 heads, 63 sectors, 256 cylinders
Units = cylinders of 16065 * 512 bytes
Device Boot Start End Blocks Id System
/dev/hda1 * 1 249 2000061 83 Linux
/dev/hda2 250 790 4345582+ 5 Extended
/dev/hda5 250 256 56196 82 Linux swap