At 03:47 PM 01/06/2000 , Bruce Kall wrote:
The easiest thing to do is to create a small /boot partition before
cylinder 1024 ...
[fcbrants@w2 fcbrants]$ df
Filesystem 1k-blocks Used Available Use% Mounted on
/dev/sda6 4040000 1365216 2469556 36% /
/dev/sda1 15522 2699 12022 18% /boot
[fcbrants@w2 fcbrants]$ cat /etc/redhat-release
Red Hat Linux release 6.1 (Cartman)
HTH
Franko
>I just put RH5.0 on a 18GB disk, but I had to get a 2.2.x kernel on it
>to have fdisk
>work correctly (the stock RH5.0 comes with 2.0.32). I'm using 2.2.13 and
>fdisk
>works fine with my 18GB disk.
>
>Essentially, I loaded up 5.0 on the hard drive with partitions up to
>1024 cylinders.
>I then put on 2.2.13 kernel and then created another partition that went
>to the end of the
>drive and all is well.
>
>Bruce
>
>
>
>
>
>
>Ryan Caveney wrote:
> >
> > From: Rick Forrester <[EMAIL PROTECTED]>
> > >[EMAIL PROTECTED] said:
> > >> I am having major problems installing 5.2 on an 18GB EIDE disk.
> > >> I have downloaded and created the latest 5.2 install image (boot.img)
> > >> from Dec 29, 1998. The system can not seem to see beyond the 1024th
> > >> cylinder of the disk. Is this just a total lie and can I just do
> > >> this, or is this a legit limit that can not be crossed??
> > >
> > >> I've been staring at howtos and screens on this all day, and to be
> > >> honest, my head is starting to hurt over this.
> > >
> > >Paul,
> > > There are 1024 cylinder issues, but they can be worked around,
> > >normally fairly easily. What's your partition structure like?
> >
> > I just installed RH6.0 on a machine with a 2GB disk, the BIOS of
> > which for some unknown reason accepts disk geometries with up to 65535
> > cylinders but only 16 **heads** -- NT had been running happily on it with
> > 4465 cyl * 15 heads, but Disk Druid set the geometry for me to 523 cyl
> * 128
> > heads (which I failed initially to notice), so of course it failed utterly
> > to even fully launch LILO on boot attempts.
> > fdisk, on the other hand, worked like a charm, though I did have to
> > use "expert mode" and the verify command gave strange and incorrect
> > results -- although the partitions had been created properly, expert-level
> > verify was unable to report a 3D address above 1023 cyl so it claimed most
> > of my partitions overlapped or failed to fit inside the extended one.
> > Nonetheless, I rebooted and installed as I had intended, with the disk
> > geometry properly detected and no further problems.
> > My suggestion, therefore, is that the "workaround" you should apply
> > is: use fdisk to set your geometry and partition tables, ignoring
> > verification warnings. :} Of course, please do insure that the
> partition(s)
> > where LILO and the kernel will live fit completely under the 1024-cylinder
> > ceiling, just in case.
> >
> > Ryan Caveney
> >
> > --
> > To unsubscribe: mail [EMAIL PROTECTED] with "unsubscribe"
> > as the Subject.
>
>--
>====================================================================
>Bruce Kall
>Mayo Foundation
>Rochester, MN
>
>e-mail: [EMAIL PROTECTED]
>phone: (507) 255-4768
>
>====================================================================
>
>
>--
>To unsubscribe: mail [EMAIL PROTECTED] with "unsubscribe"
>as the Subject.
Frank C. Brants
Netopia, Inc.
Systems Engineer
franko -at- netopia.com
Fax 972-855-6610
Office / Cell 214-769-0354
Text Page http://primeco.textmsg.com/
--
To unsubscribe: mail [EMAIL PROTECTED] with "unsubscribe"
as the Subject.
