[3514] in SIPB_Linux_Development
Re: Some concerns with the SIPB redhat installer
daemon@ATHENA.MIT.EDU (Greg Hudson)
Mon Jun 18 09:54:22 2001
Message-Id: <200106181354.JAA22847@egyptian-gods.MIT.EDU>
To: John Hawkinson <jhawk@MIT.EDU>
cc: linux-dev@MIT.EDU
In-Reply-To: Your message of "Sun, 17 Jun 2001 23:31:30 EDT."
<200106180331.XAA25875@multics.mit.edu>
Date: Mon, 18 Jun 2001 09:54:09 -0400
From: Greg Hudson <ghudson@MIT.EDU>
> Is it not possible to just hack the installer to verify that the
> linux partition is early enough and punt otherwise, so you don't
> leave the user with a disk that won't boot usefully?
I have a machine next to me which has a Linux partition past the 8GB
mark (starts about 50GB in, if I remember right), and it boots just
fine. Ironically, I had to fight with the Red Hat 7.1 installer to
make this happen, because the installer incorrectly decided that my
machine's BIOS didn't have the necessary feature. I don't know how it
decided that. (Specifically, Disk Druid simply refused to create a
boot partition past the 8GB mark, so I had to run fdisk by hand, and
then it gave me a warning at some point but let me proceed.)
So, two observations: (a) the naive fix for this problem is
unnecessarily limiting for some users, and (b) SIPB will get a less
naive fix for this problem simply by moving to Red Hat 7.1, although
that fix seems to still be too conservative.
I agree that simply yielding an unbootable machine, with no
diagnostic, is pretty sad.
I think the Red Hat 7.1 installer also has some automatic partitioning
support (when it doesn't fail like I mentioned above), but I don't
remember clearly. It would almost certainly have to be hacked up to
make an AFS cache partition to be useful for the SIPB installer, of
course.
