On Tue, 17 Mar 1998, Jim Millick wrote:
> The motherboard is an AMI PCI Excalibur, with a primary and
> secondary PCI controller built in. It's about 4 years old,
> with the most recent BIOS installed.
...
> mkfs starts building the inode tables, and counts up
> nnn/407. At various points along the way (I've tried this
> 7x now), the nnn count-up just stops. I can't toggle to
> another virtual console, vulcan nerve pinch doesn't work,
This smells of the RZ1000 IDE controller, which did silly things like lock
up when you tried to use configurations any more complex than 'disabled'
(The CMD640 was better, it just loses data). Compile your kernel with the
CMD640 and RZ1000 bugfixes, and see if it gets better.
You might, as a last resort, try adding
ide1=serialize
to your kernel command line. Cures all wounds.
On the other hand, the fact that the BIOS can't figure this out either
might lead me to believe you have a problem other than buggy controller.
Does your CD-ROM work on the primary channel as the slave? What sort of
CDROM is it? I assume you have all your master/slave jumpers set
correctly. The CDROM must always be the slave on a channel with two
devices.
--
PLEASE read the Red Hat FAQ, Tips, Errata and the MAILING LIST ARCHIVES!
http://www.redhat.com/RedHat-FAQ /RedHat-Errata /RedHat-Tips /mailing-lists
To unsubscribe: mail [EMAIL PROTECTED] with
"unsubscribe" as the Subject.
