Anaconda gets interrupted by LVM activation - BZ 232862 - anyone else seeing this?

[G.Wolfe Woodbury]

The bug report may well prove that I don't know as much as I thought I did,
but is anyone else seeing rawhide (f7test) failures with non-default disk
layouts?

Bugzilla 232862 details much of my problem, but I present a summary here.

My testbed machine that it is failing on is a MSI motherboard, VIA chipset, and
a Celeron @600MHz (an older machine, yes).

I get consistent LVM activation failures during the installation no matter
what I try to do with the disk layouts.  I seem to be able to get an installation
process going when I accept the "default" layout (and ssometimes even then!)

I thought for a while (briefly) that I had found a syntax error in anaconda's
call to vgchange, but that is obviously not the problem.

There is the change over from the old ATA method to the new, and this started
happening at that transition.  However, I have wiped the partition table/MBR
and started from scratch a couple of times, still getting the errors.  I have
even tride hand configuring with no LVM partitions/volumes at all and still get
vgchange failures.

I've done memtest86+, run some heavy math stuff and otherwise verified that
under FC6 (with updates) that the hardware is still basically good, but this
machine just does not like FC7 series installs.

-- 
G.Wolfe Woodbury
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
Url : http://lists.fedoraproject.org/pipermail/test/attachments/20070418/310cd351/attachment.bin