On Wed, 13 May 2009, Tezyn Drasdin wrote:

> Group,
> 
> I am trying to set up a PXE boot install setup for SL 53 using NFS.  I have a working PXE server, that I
> am able to install Fedora, Oracle Unbreakable, Gentoo from using NFS for media access.

I install SL53 using PXE and NFS, so this should work.

> When I boot the workstation, everything appears to go fine (Welcome to Scientific Linux, Choose your
> Language, etc).  But when I get to detecting/load the media, it fails on me, giving me the error:
> 
> loader received SIGSEGV! Backtrace:
> [0x8048cf4]
> [0x4cf420]
> [0x81a9858]
> [0x805de62]
> [0x806119a]
> [0x8049326]
> [0x804a9a3]
> [0x81623c8]
> [0x8048131]
> install exited abnormally [1/1]
> sending termination signals... done
> sending kill signals... done
> disabling swap ...
> unmounting filesystem ...
>      /mnt/runtime done
>      disabling /dev/loop0
>      /mnt/source2 done
>      disabling /dev/loop1
>      /proc done
>      /dev/pts done
>      /sys done
>      /tmp/ramfs done
>      /mnt/source done
> you man safely reboot your system

Are there any other messages on other virtual consoles
- type <Ctrl><Alt><f2> , <Ctrl><Alt><f3> etc. ?

> I don't believe that this is related to the NFS connection itself, because I am able to install the
> other RHE derived systems, using this server, and if I don't point to the correct directory, it does
> successfully see that there isn't the correct media in that folder.  I believe that this means that the
> directory is being successfully mounted, and read, which leads me to believe it is an issue with the
> media itself.
> 
> I have tested this with both the dvd and the 7 disc cd media, on both physical hardware and esxi virtual
> machine, with exactly the same results.  I have checked the md5sum on all of the media against the
> listed checksums and all pass.  I have also checked the hash of the initrd.img and vmlinuz that I am
> using against those in the /isolinux and /images/pxeboot/ directories, and all 3 are the same.

Assuming that you can install from the disks, my first guess
would be that the network driver on the initrd doesn't like your
ethernet device (may be fine when quite but fail under load?).
You could try the install with a different model of ethernet card.
Once installed the updated kernel may support the orginal/onboard
ethernet.

I've watched people grovel around and update drivers inside
the initrd.img so that you don't have to do this for all your new 
machines, but can't say how they did it.

-- 
Dr. Andrew C. Aitchison		Computer Officer, DPMMS, Cambridge
[log in to unmask]	http://www.dpmms.cam.ac.uk/~werdna