Didn't get answer

[Andrew Farris]

On Mon, 2004-09-27 at 19:03 -0400, Alan Cox wrote:
> On Tue, Sep 28, 2004 at 12:19:26AM +0100, Andrius wrote:
> > I didn't get answer to my question. Is it problem unclear or unknown? 
> In installation process of Fedora Core 3 test 2 for AMD64 I got such bug 
> message  "Bug: Assertion (heads > 0) at disk_dos.c:471 in function
> probe_partition_for_geom() failed." (after keyboard configuration). 
> If I press cancel I can go through other steps, but when I got the same 
> message when it is time to copy packages to disk and installation stops.
> > -- 
> 
> Can yo describe your disk setup ?

For background I'll explain mine because I also saw this message on my
P4 machine.  It only happened when I had used fdisk to modify the
partition table after Anaconda was running (via vt2).  The message was
displayed when Disk Druid configuration, not before (at keyboard conf).

I manually deleted and created new partitions inside an existing
extended partition which previously housed FC2 (in rawhide form), using
fdisk just prior to Disk Druid in Anaconda.  FC2 fdisk had created the
previous partitions inside the extended, however Windows XP SP1 had
created the extended partition itself.  Now the new partitions created
with FC3t2 fdisk caused the assertion failure.

I could not proceed with the installation without this error until I
removed the fdisk created partitions used manual Disk Druid instead (a
reboot was also necessary for geometry to be re-read).

In the below setup, /dev/hdb5 was not deleted in the reformat and
reinstall to FC3t2 (which is why it is now displayed out of disk order).
The extended partition was left as is during the whole process.

-#-> fdisk /dev/hdb -l

Disk /dev/hdb: 40.0 GB, 40020664320 bytes
255 heads, 63 sectors/track, 4865 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes

   Device Boot      Start         End      Blocks   Id  System
/dev/hdb1               1         263     2112516    b  W95 FAT32
/dev/hdb2             264        3657    27262305    f  W95 Ext'd (LBA)
/dev/hdb3            3658        4865     9703260    c  W95 FAT32 (LBA)
/dev/hdb5            3214        3657     3566398+  83  Linux
/dev/hdb6   *         264         271       64197   83  Linux
/dev/hdb7             272         666     3172806   83  Linux
/dev/hdb8             667        1074     3277228+  83  Linux
/dev/hdb9            1075        1208     1076323+  82  Linux swap
/dev/hdb10           1209        3213    16105131   83  Linux

Partition table entries are not in disk order

Apparently, the fdisk and disk druid results varied enough to cause this
error.. or perhaps it was due to using fdisk while anaconda may have
already checked disk geometry for disk druid?  This has not been an
issue previously.