All of lore.kernel.org
 help / color / mirror / Atom feed
From: Guillaume Rousse <Guillaume.Rousse@inria.fr>
To: Tim Deegan <Tim.Deegan@xensource.com>
Cc: xen-devel@lists.xensource.com
Subject: Re: troubles using network boot with HVM guest
Date: Wed, 06 Jun 2007 16:34:11 +0200	[thread overview]
Message-ID: <4666C5E3.209@inria.fr> (raw)
In-Reply-To: <20070606103400.GA21901@york.uk.xensource.com>

[-- Attachment #1: Type: text/plain, Size: 1806 bytes --]

Tim Deegan a écrit :
> At 14:58 +0200 on 04 Jun (1180969137), Guillaume Rousse wrote:
>> As a potential issue, dhcp reply comes from another network, through a
>> proxy. I once had troubles with windows client ignoring those answers if
>> their origin was not strictly set to the local network broadcast adress.
> 
> From a quick rummage around in the packet trace, it looks like the DHCP
> siaddr ("next server to contact") field in the reply packet is set to
> zero, and etherboot explicitly discards dhcp replies where that field is
> zero.  (Disclaimer: this is based on about ten minutes of poking at the
> etherboot source and RFC2131 -- I am by no means a DCHP expert).
> 
> Can you check the DHCP server settings, and whether your proxy is doing
> something wierd to the packet? 
I did check with a real host, it worked perfectly. But you find the
issue: the server was only answering PXE requests, whereas the emulated
network card send an etherboot one.Just adding etherboot configuration
on the server was enough to provide a functional answer.

I was able to make the virtual domain download a grub stage2 built for
rtl839. As I was not able to rebuild grub myself (compilation errors), I
used this one
http://cvs.mandriva.com/cgi-bin/viewvc.cgi/gi/tools/i386/netboot/rtl8139/?pathrev=MDK-2006_0-update

Once loaded, the virtual host send a new DHCP request, and then ignore
the answer once again, despite the fact than the next-server field is
correct. According to my knowledge, it should finish the DHCP
transaction, then requires /boot/grub/menu.list from the tftp server.

I'm joining the new traffic capture again. Any hint welcome, I'm a bit
lost in netboot subtilities :)
-- 
Guillaume Rousse
Moyens Informatiques - INRIA Futurs
Tel: 01 69 35 69 62


[-- Attachment #2: bootp.cap --]
[-- Type: application/octet-stream, Size: 4578 bytes --]

[-- Attachment #3: Type: text/plain, Size: 138 bytes --]

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel

  reply	other threads:[~2007-06-06 14:34 UTC|newest]

Thread overview: 11+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2007-06-04 12:27 troubles using network boot with HVM guest Guillaume Rousse
2007-06-04 12:51 ` Tim Deegan
2007-06-04 12:58   ` Guillaume Rousse
2007-06-06 10:34     ` Tim Deegan
2007-06-06 14:34       ` Guillaume Rousse [this message]
2007-06-06 15:07         ` Tim Deegan
2007-06-06 15:27           ` Guillaume Rousse
2007-06-04 15:20 ` Anthony Liguori
2007-06-04 16:12   ` Guillaume Rousse
2007-06-04 18:14     ` Anthony Liguori
2007-06-04 19:36       ` Guillaume Rousse

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=4666C5E3.209@inria.fr \
    --to=guillaume.rousse@inria.fr \
    --cc=Tim.Deegan@xensource.com \
    --cc=xen-devel@lists.xensource.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.