When doing a full install of RH 6.1.92, it erroneously told me I had insufficient disk space. My guess is that it's using an unsigned int for size of partition, and since I've got a /usr partition of 2.4gb, it thinks I've got negative disk space in /usr. The RH6.1 install worked fine, and I was able to upgrade from that to 6.1.92 with no trouble. I tried tracking the problem down in source code, but don't understand python well enough to be able to find it. Sorry. My disklabel looks like: root@tuck:/home/fletcher# fdisk /dev/hda Command (m for help): p Disk /dev/hda: 255 heads, 63 sectors, 784 cylinders Units = cylinders of 16065 * 512 bytes Device Boot Start End Blocks Id System /dev/hda1 1 33 265041 83 Linux /dev/hda2 34 784 6032407+ 5 Extended /dev/hda5 34 78 361431 82 Linux swap /dev/hda6 79 111 265041 83 Linux /dev/hda7 112 412 2417751 83 Linux /dev/hda8 413 478 530113+ 83 Linux /dev/hda9 479 751 2192841 83 Linux /dev/hda10 752 784 265041 83 Linux Command (m for help): q root@tuck:/home/fletcher# df Filesystem 1k-blocks Used Available Use% Mounted on /dev/hda1 256667 34312 209103 14% / /dev/hda6 256667 43 243372 0% /tmp /dev/hda7 2379700 1538108 720708 68% /usr /dev/hda8 521748 19620 475624 4% /var /dev/hda9 2158384 473060 1575684 23% /usr/local /dev/hda10 256667 2580 240835 1% /home root@tuck:/home/fletcher#
I presume you meant "signed int" when you said "unsigned int" ???
Yes, I meant 'signed'. Sorry 'bout that. I'll start looking into the source code for this more deeply tomorrow (2/16/00) evening when I get the chance. Hopefully python won't confuse me as much as I've already confused myself. :)
There is a pretty good chance this is what is happening. We found a similiar bug in the RAID code which resulted in much of the same behavior. I will also be playing around in the lab with this setup, so hopefully one of the two of us will be able to shed light on this matter.
Attempted to reproduce with specified partition table and it works in latest beta of installer.
So, which "latest beta of installer" did you try, and where can I get it to try it out and ensure that this bug's been fixed? I did spend about 3 hours looking through the Anaconda source, and got thoroughly lost.