SCIENTIFIC-LINUX-USERS Archives

August 2011

SCIENTIFIC-LINUX-USERS@LISTSERV.FNAL.GOV

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Reply To:
Date:
Tue, 2 Aug 2011 12:14:35 -0400
Content-Type:
text/plain
Parts/Attachments:
text/plain (66 lines)
On Thu, Jul 28, 2011 at 9:32 AM, Troy Dawson <[log in to unmask]> wrote:
> On 07/27/2011 05:23 PM, Rich wrote:
>>
>> Hi all,
>> I've got questions about two different failure modes I experienced
>> while trying to kickstart SL6.1 on some hardware to play with.
>>
>> First, when I kickstarted it on this hardware, it immediately
>> hardlocked (display black, keyboard unresponsive with no lights on or
>> blinking, no network activity), I believe while trying to bring up the
>> network (I can't tell; it doesn't have network up prior to dying, so
>> remote syslog isn't working, and then it's utterly responseless
>> afterward). I booted it with noprobe, and it came up, but as soon as I
>> tried to load the forcedeth module for its onboard NIC, /sbin/loader
>> experienced SIGSEGV. This happens reproducibly.
>>
>> Second, when I tried installing 6.1 with the 6.0 vmlinuz/initrd.img
>> (where the above problem didn't occur at all), if I try to install in
>> text or graphical mode, it complains "Please add the following
>> sections and try again" "Bootloader configuration", where I know I
>> have bootloader specified correctly and not inside a pre/post/packages
>> section (I use "bootloader --location=mbr --driveorder=sda"; sans
>> quotes, obviously) - this looked similar to RH bug #506428, so I
>> figured I'd give cmdline mode a try.
>> If I use "cmdline" install mode, I fail with "In interactive step
>> timezone, can't continue" - which looks like RH bug 462595 or 672095.
>>
>> Thoughts, anyone?
>>
>> - Rich
>
> Hi Rich,
> I just want to narrow down some things.
>
> For your kickstarts, you are using a pxeboot kickstart?

Correct.

> OK, the real question is, how are you booting and starting your kickstart?

root@gwln1:/tftpboot/pxelinux.cfg>cat netboot-sl61.x86_64.private
default linux
label linux
  kernel vmlinuz-sl60.x86_64
  append load_ramdisk=1 initrd=initrd-sl60.x86_64.img ip=dhcp network
ks=nfs:[private IP of KS server]:/export/d0/kickstart/sl61/phaconfigs/
ksdevice=eth0 nokill loglevel=debug syslog=[private IP of KS server]
cmdline

> Are you able to kickstart, using the same kickstart, with SL 6.0?
> (with the one obvious change that the kickstart is pointing to 6.1 versus
> 6.0)

Are you asking if I can do what the above pxelinux config says,
wherein I use the SL6 pxeboot initrd/vmlinuz and point it at SL61, or
if I can kickstart SL60 with the identical kickstart file with
s/sl61/sl60/g?

The former results in the behavior I described in my original email;
the latter I have not tried, and will doso now.

I can provide the KS file(s) I'm using - would it be preferable for me
to post a URL to them, or attach them all?

- Rich

ATOM RSS1 RSS2