Re: [Soekris] Soekris PXE boot

2008-02-21 Thread Bill Maas
Hi Brian,

On Thu, 2008-02-21 at 13:43 +, Brian Candler wrote:
> On Thu, Feb 21, 2008 at 10:50:00AM +0100, Bill Maas wrote:
> > Hi Voipcrazy,
> > 
> > http://www.modemsite.com/56k/x2-hyperterm.asp
> > 
> > Search "characters twice".
> > 
> > It's called a manual;).
> 
> It's nothing to do with hyperterminal or half-duplex settings.

I'm very sorry..

> > > But what I get on my soekris console is that 
> > > 
> > > CCoouulldd  nnoo
> > > tt  ffiinndd  kk
> > > eerrnneell  iimm
> > > aaggee::  lliinn
> > > uuxx
> > > 
> > > bbtt::
> 
> (1) The character duplication is because pxelinux is outputting both to the
> COM port and the VGA BIOS. The Soekris redirects the VGA BIOS characters to
> the COM port. Hence you get them twice.
> 
> (2) The 16-column line wrap is a bug in the Soekris VGA BIOS emulation. You
> can patch and rebuild pxelinux.0 to avoid this, by stopping pxelinux writing
> to the VGA BIOS. Details at the end of
> 
> http://archive.netbsd.se/?ml=soekris-tech&a=2007-11&m=5685776
> 
> and the patch, with instructions on how to rebuild pxelinux.0 using nasm, is
> higher up. If I remember rightly, this patch will also prevent the character
> duplication.
> 
> Just for reference, here's a working pxelinux.cfg/default:
> 
---
> serial 0 19200 0x303

DEFAULT linux

> label linux
>   kernel openwrt-x86-2.6-vmlinuz
>   append initrd=openwrt-x86-2.6-rootfs.cpio.gz init=/etc/preinit console=tty0 
> console=ttyS0,19200n8
---

Might do the trick.. 

Bill

> And here's a working grub/menu.lst (with settings to allow the unit to boot
> even when no serial cable is connected)
> 
> serial --unit=0 --speed=19200 --word=8 --parity=no --stop=1
> terminal --timeout=0 --dumb serial
> hiddenmenu
> 
> default 0
> timeout 0
> 
> title   OpenWrt
> root(hd0,0)
> kernel  /boot/vmlinuz init=/etc/preinit console=tty0 console=ttyS0,19200n8 
> reboot=bios
> initrd  /boot/rootfs.cpio.gz
> boot
> 
> However, fixing the character output problems is mostly cosmetic. It won't
> solve the problem of the missing kernel.
> 
> B.
> 
-- 
"What's a computer?" - MES

___
Soekris-tech mailing list
Soekris-tech@lists.soekris.com
http://lists.soekris.com/mailman/listinfo/soekris-tech


Re: [Soekris] Soekris PXE boot

2008-02-21 Thread Jochen Grotepass
voip crazy schrieb:
> Thanks for your answers, but my problem persists,
>
> I think PXE boot works on my soekris, look thos tftpd log:
>
> Feb 20 16:51:17 localhost dhcpd: DHCPDISCOVER from 00:00:24:c9:5a:14 
> via eth1
> Feb 20 16:51:17 localhost dhcpd: DHCPOFFER on 192.168.30.63 
>  to 00:00:24:c9:5a:14 via eth1
> Feb 20 16:51:18 localhost dhcpd: DHCPREQUEST for 192.168.30.63 
>  (192.168.30.253 ) from 
> 00:00:24:c9:5a:14 via eth1
> Feb 20 16:51:18 localhost dhcpd: DHCPACK on 192.168.30.63 
>  to 00:00:24:c9:5a:14 via eth1
> Feb 20 16:51:18 localhost in.tftpd[3379]: RRQ from 192.168.30.63 
>  filename pxelinux.0
> Feb 20 16:51:18 localhost in.tftpd[3379]: tftp: client does not accept 
> options
> Feb 20 16:51:18 localhost in.tftpd[3380]: RRQ from 192.168.30.63 
>  filename pxelinux.0
> Feb 20 16:51:25 localhost in.tftpd[3381]: RRQ from 192.168.30.63 
>  filename pxelinux.cfg/01-00-00-24-c9-5a-14
> Feb 20 16:51:25 localhost in.tftpd[3381]: sending NAK (1, File not 
> found) to 192.168.30.63 
> Feb 20 16:51:25 localhost in.tftpd[3382]: RRQ from 192.168.30.63 
>  filename pxelinux.cfg/C0A81E3F
> Feb 20 16:51:25 localhost in.tftpd[3382]: sending NAK (1, File not 
> found) to 192.168.30.63 
> Feb 20 16:51:26 localhost in.tftpd[3383]: RRQ from 192.168.30.63 
>  filename pxelinux.cfg/C0A81E3
> Feb 20 16:51:26 localhost in.tftpd[3383]: sending NAK (1, File not 
> found) to 192.168.30.63 
> Feb 20 16:51:26 localhost in.tftpd[3384]: RRQ from 192.168.30.63 
>  filename pxelinux.cfg/C0A81E
> Feb 20 16:51:26 localhost in.tftpd[3384]: sending NAK (1, File not 
> found) to 192.168.30.63 
> Feb 20 16:51:27 localhost in.tftpd[3385]: RRQ from 192.168.30.63 
>  filename pxelinux.cfg/C0A81
> Feb 20 16:51:27 localhost in.tftpd[3385]: sending NAK (1, File not 
> found) to 192.168.30.63 
> Feb 20 16:51:28 localhost in.tftpd[3386]: RRQ from 192.168.30.63 
>  filename pxelinux.cfg/C0A8
> Feb 20 16:51:28 localhost in.tftpd[3386]: sending NAK (1, File not 
> found) to 192.168.30.63 
> Feb 20 16:51:28 localhost in.tftpd[3387]: RRQ from 192.168.30.63 
>  filename pxelinux.cfg/C0A
> Feb 20 16:51:28 localhost in.tftpd[3387]: sending NAK (1, File not 
> found) to 192.168.30.63 
> Feb 20 16:51:29 localhost in.tftpd[3388]: RRQ from 192.168.30.63 
>  filename pxelinux.cfg/C0
> Feb 20 16:51:29 localhost in.tftpd[3388]: sending NAK (1, File not 
> found) to 192.168.30.63 
> Feb 20 16:51:29 localhost in.tftpd[3389]: RRQ from 192.168.30.63 
>  filename pxelinux.cfg/C
> Feb 20 16:51:29 localhost in.tftpd[3389]: sending NAK (1, File not 
> found) to 192.168.30.63 
> Feb 20 16:51:30 localhost in.tftpd[3390]: RRQ from 192.168.30.63 
>  filename pxelinux.cfg/default
> Feb 20 16:51:48 localhost in.tftpd[3676]: RRQ from 192.168.30.63 
>  filename 
> debian-installer/i386/boot-screens/boot.txt
> Feb 20 16:51:58 localhost in.tftpd[3692]: RRQ from 192.168.30.63 
>  filename linux
> Feb 20 16:51:58 localhost in.tftpd[3692]: sending NAK (1, File not 
> found) to 192.168.30.63 
> Feb 20 16:51:58 localhost in.tftpd[3693]: RRQ from 192.168.30.63 
>  filename linux.cbt
> Feb 20 16:51:58 localhost in.tftpd[3693]: sending NAK (1, File not 
> found) to 192.168.30.63 
> Feb 20 16:51:58 localhost in.tftpd[3694]: RRQ from 192.168.30.63 
>  filename linux.0
> Feb 20 16:51:58 localhost in.tftpd[3694]: sending NAK (1, File not 
> found) to 192.168.30.63 
> Feb 20 16:51:58 localhost in.tftpd[3695]: RRQ from 192.168.30.63 
>  filename linux.com 
> Feb 20 16:51:58 localhost in.tftpd[3695]: sending NAK (1, File not 
> found) to 192.168.30.63 
> Feb 20 16:51:58 localhost in.tftpd[3696]: RRQ from 192.168.30.63 
>  filename linux.c32
> Feb 20 16:51:58 localhost in.tftpd[3696]: sending NAK (1, File not 
> found) to 192.168.30.63 
>
>
> But what I get on my soekris console is that
>
> CCoouulldd  nnoo
> tt  ffiinndd  kk
> eerrnneell  iimm
> aaggee::  lliinn
> uuxx
>
> bbtt::
>
> I type "linux" as kernel image and it does not boot
>
> The caracters are writing twice and it cannot find the kernel image 
> version
>
> My /pxelinux.cfg/default
>
> have a line like this SERIAL 0,19200 for console connection
>
> And the default image to boot is configured like that in the same file

Re: [Soekris] Soekris PXE boot

2008-02-21 Thread Brian Candler
On Wed, Feb 20, 2008 at 05:12:17PM +0100, voip crazy wrote:
>Thanks for your answers, but my problem persists,
>I think PXE boot works on my soekris, look thos tftpd log:
...
>Feb 20 16:51:58 localhost in.tftpd[3692]: RRQ from [27]192.168.30.63
>filename linux
>Feb 20 16:51:58 localhost in.tftpd[3692]: sending NAK (1, File not
>found) to [28]192.168.30.63
>Feb 20 16:51:58 localhost in.tftpd[3693]: RRQ from [29]192.168.30.63
>filename linux.cbt
>Feb 20 16:51:58 localhost in.tftpd[3693]: sending NAK (1, File not
>found) to [30]192.168.30.63
>Feb 20 16:51:58 localhost in.tftpd[3694]: RRQ from [31]192.168.30.63
>filename linux.0
>Feb 20 16:51:58 localhost in.tftpd[3694]: sending NAK (1, File not
>found) to [32]192.168.30.63
>Feb 20 16:51:58 localhost in.tftpd[3695]: RRQ from [33]192.168.30.63
>filename [34]linux.com
>Feb 20 16:51:58 localhost in.tftpd[3695]: sending NAK (1, File not
>found) to [35]192.168.30.63
>Feb 20 16:51:58 localhost in.tftpd[3696]: RRQ from [36]192.168.30.63
>filename linux.c32
>Feb 20 16:51:58 localhost in.tftpd[3696]: sending NAK (1, File not
>found) to [37]192.168.30.63

If your tftp server home directory is /var/tftpboot then you can see the
client is asking for

  /var/tftpboot/linux
  /var/tftpboot/linux.cbt
  /var/tftpboot/linux.0
  /var/tftpboot/linux.com
  /var/tftpboot/linux.c32

and none of these files exists.
___
Soekris-tech mailing list
Soekris-tech@lists.soekris.com
http://lists.soekris.com/mailman/listinfo/soekris-tech


Re: [Soekris] Soekris PXE boot

2008-02-21 Thread Brian Candler
On Thu, Feb 21, 2008 at 10:50:00AM +0100, Bill Maas wrote:
> Hi Voipcrazy,
> 
> http://www.modemsite.com/56k/x2-hyperterm.asp
> 
> Search "characters twice".
> 
> It's called a manual;).

It's nothing to do with hyperterminal or half-duplex settings.

> > But what I get on my soekris console is that 
> > 
> > CCoouulldd  nnoo
> > tt  ffiinndd  kk
> > eerrnneell  iimm
> > aaggee::  lliinn
> > uuxx
> > 
> > bbtt::

(1) The character duplication is because pxelinux is outputting both to the
COM port and the VGA BIOS. The Soekris redirects the VGA BIOS characters to
the COM port. Hence you get them twice.

(2) The 16-column line wrap is a bug in the Soekris VGA BIOS emulation. You
can patch and rebuild pxelinux.0 to avoid this, by stopping pxelinux writing
to the VGA BIOS. Details at the end of

http://archive.netbsd.se/?ml=soekris-tech&a=2007-11&m=5685776

and the patch, with instructions on how to rebuild pxelinux.0 using nasm, is
higher up. If I remember rightly, this patch will also prevent the character
duplication.

Just for reference, here's a working pxelinux.cfg/default:

serial 0 19200 0x303
label linux
  kernel openwrt-x86-2.6-vmlinuz
  append initrd=openwrt-x86-2.6-rootfs.cpio.gz init=/etc/preinit console=tty0 
console=ttyS0,19200n8

And here's a working grub/menu.lst (with settings to allow the unit to boot
even when no serial cable is connected)

serial --unit=0 --speed=19200 --word=8 --parity=no --stop=1
terminal --timeout=0 --dumb serial
hiddenmenu

default 0
timeout 0

title   OpenWrt
root(hd0,0)
kernel  /boot/vmlinuz init=/etc/preinit console=tty0 console=ttyS0,19200n8 
reboot=bios
initrd  /boot/rootfs.cpio.gz
boot

However, fixing the character output problems is mostly cosmetic. It won't
solve the problem of the missing kernel.

B.
___
Soekris-tech mailing list
Soekris-tech@lists.soekris.com
http://lists.soekris.com/mailman/listinfo/soekris-tech


Re: [Soekris] Soekris PXE boot

2008-02-21 Thread Bill Maas
Hi Voipcrazy,

http://www.modemsite.com/56k/x2-hyperterm.asp

Search "characters twice".

It's called a manual;).

Bill

On Wed, 2008-02-20 at 17:12 +0100, voip crazy wrote:
> Thanks for your answers, but my problem persists,
> 
> I think PXE boot works on my soekris, look thos tftpd log:
> 
> Feb 20 16:51:17 localhost dhcpd: DHCPDISCOVER from 00:00:24:c9:5a:14
> via eth1
> Feb 20 16:51:17 localhost dhcpd: DHCPOFFER on 192.168.30.63 to
> 00:00:24:c9:5a:14 via eth1
> Feb 20 16:51:18 localhost dhcpd: DHCPREQUEST for 192.168.30.63
> (192.168.30.253) from 00:00:24:c9:5a:14 via eth1
> Feb 20 16:51:18 localhost dhcpd: DHCPACK on 192.168.30.63 to
> 00:00:24:c9:5a:14 via eth1
> Feb 20 16:51:18 localhost in.tftpd[3379]: RRQ from 192.168.30.63
> filename pxelinux.0
> Feb 20 16:51:18 localhost in.tftpd[3379]: tftp: client does not accept
> options
> Feb 20 16:51:18 localhost in.tftpd[3380]: RRQ from 192.168.30.63
> filename pxelinux.0
> Feb 20 16:51:25 localhost in.tftpd[3381]: RRQ from 192.168.30.63
> filename pxelinux.cfg/01-00-00-24-c9-5a-14
> Feb 20 16:51:25 localhost in.tftpd[3381]: sending NAK (1, File not
> found) to 192.168.30.63
> Feb 20 16:51:25 localhost in.tftpd[3382]: RRQ from 192.168.30.63
> filename pxelinux.cfg/C0A81E3F
> Feb 20 16:51:25 localhost in.tftpd[3382]: sending NAK (1, File not
> found) to 192.168.30.63
> Feb 20 16:51:26 localhost in.tftpd[3383]: RRQ from 192.168.30.63
> filename pxelinux.cfg/C0A81E3
> Feb 20 16:51:26 localhost in.tftpd[3383]: sending NAK (1, File not
> found) to 192.168.30.63
> Feb 20 16:51:26 localhost in.tftpd[3384]: RRQ from 192.168.30.63
> filename pxelinux.cfg/C0A81E
> Feb 20 16:51:26 localhost in.tftpd[3384]: sending NAK (1, File not
> found) to 192.168.30.63
> Feb 20 16:51:27 localhost in.tftpd[3385]: RRQ from 192.168.30.63
> filename pxelinux.cfg/C0A81
> Feb 20 16:51:27 localhost in.tftpd[3385]: sending NAK (1, File not
> found) to 192.168.30.63
> Feb 20 16:51:28 localhost in.tftpd[3386]: RRQ from 192.168.30.63
> filename pxelinux.cfg/C0A8
> Feb 20 16:51:28 localhost in.tftpd[3386]: sending NAK (1, File not
> found) to 192.168.30.63
> Feb 20 16:51:28 localhost in.tftpd[3387]: RRQ from 192.168.30.63
> filename pxelinux.cfg/C0A
> Feb 20 16:51:28 localhost in.tftpd[3387]: sending NAK (1, File not
> found) to 192.168.30.63
> Feb 20 16:51:29 localhost in.tftpd[3388]: RRQ from 192.168.30.63
> filename pxelinux.cfg/C0
> Feb 20 16:51:29 localhost in.tftpd[3388]: sending NAK (1, File not
> found) to 192.168.30.63
> Feb 20 16:51:29 localhost in.tftpd[3389]: RRQ from 192.168.30.63
> filename pxelinux.cfg/C
> Feb 20 16:51:29 localhost in.tftpd[3389]: sending NAK (1, File not
> found) to 192.168.30.63
> Feb 20 16:51:30 localhost in.tftpd[3390]: RRQ from 192.168.30.63
> filename pxelinux.cfg/default
> Feb 20 16:51:48 localhost in.tftpd[3676]: RRQ from 192.168.30.63
> filename debian-installer/i386/boot-screens/boot.txt
> Feb 20 16:51:58 localhost in.tftpd[3692]: RRQ from 192.168.30.63
> filename linux
> Feb 20 16:51:58 localhost in.tftpd[3692]: sending NAK (1, File not
> found) to 192.168.30.63
> Feb 20 16:51:58 localhost in.tftpd[3693]: RRQ from 192.168.30.63
> filename linux.cbt
> Feb 20 16:51:58 localhost in.tftpd[3693]: sending NAK (1, File not
> found) to 192.168.30.63
> Feb 20 16:51:58 localhost in.tftpd[3694]: RRQ from 192.168.30.63
> filename linux.0
> Feb 20 16:51:58 localhost in.tftpd[3694]: sending NAK (1, File not
> found) to 192.168.30.63
> Feb 20 16:51:58 localhost in.tftpd[3695]: RRQ from 192.168.30.63
> filename linux.com
> Feb 20 16:51:58 localhost in.tftpd[3695]: sending NAK (1, File not
> found) to 192.168.30.63
> Feb 20 16:51:58 localhost in.tftpd[3696]: RRQ from 192.168.30.63
> filename linux.c32
> Feb 20 16:51:58 localhost in.tftpd[3696]: sending NAK (1, File not
> found) to 192.168.30.63
> 
> 
> But what I get on my soekris console is that 
> 
> CCoouulldd  nnoo
> tt  ffiinndd  kk
> eerrnneell  iimm
> aaggee::  lliinn
> uuxx
> 
> bbtt::
> 
> I type "linux" as kernel image and it does not boot
> 
> The caracters are writing twice and it cannot find the kernel image
> version
> 
> My /pxelinux.cfg/default
> 
> have a line like this SERIAL 0,19200 for console connection
> 
> And the default image to boot is configured like that in the same file
> 
> kernel linux
> append vga=normal initrd=initrd.gz
> console=ttyS0,19200n8
> ramdisk_size=14332 root=/dev/rd/0 rw  --
> 
> 
> Why is this happening?
> How sould I adjust my Hyperteminal?
> 
> Thanks 
> 
> Best Regards.
> 
> VoipCrazy
> 
> 
> 
> ___
> Soekris-tech mailing list
> Soekris-tech@lists.soekris.com
> http://lists.soekris.com/mailman/listinfo/soekris-tech
-- 
"What's a computer?" - MES

___
Soekris-tech mailing list
Soekris-tech@lists.soekris.com
http://lists.soekris.com/mailman/listinfo/soekris-tech


Re: [Soekris] Soekris PXE boot

2008-02-20 Thread voip crazy
Thanks for your answers, but my problem persists,

I think PXE boot works on my soekris, look thos tftpd log:

Feb 20 16:51:17 localhost dhcpd: DHCPDISCOVER from 00:00:24:c9:5a:14 via
eth1
Feb 20 16:51:17 localhost dhcpd: DHCPOFFER on 192.168.30.63 to
00:00:24:c9:5a:14 via eth1
Feb 20 16:51:18 localhost dhcpd: DHCPREQUEST for 192.168.30.63 (
192.168.30.253) from 00:00:24:c9:5a:14 via eth1
Feb 20 16:51:18 localhost dhcpd: DHCPACK on 192.168.30.63 to
00:00:24:c9:5a:14 via eth1
Feb 20 16:51:18 localhost in.tftpd[3379]: RRQ from 192.168.30.63 filename
pxelinux.0
Feb 20 16:51:18 localhost in.tftpd[3379]: tftp: client does not accept
options
Feb 20 16:51:18 localhost in.tftpd[3380]: RRQ from 192.168.30.63 filename
pxelinux.0
Feb 20 16:51:25 localhost in.tftpd[3381]: RRQ from 192.168.30.63 filename
pxelinux.cfg/01-00-00-24-c9-5a-14
Feb 20 16:51:25 localhost in.tftpd[3381]: sending NAK (1, File not found) to
192.168.30.63
Feb 20 16:51:25 localhost in.tftpd[3382]: RRQ from 192.168.30.63 filename
pxelinux.cfg/C0A81E3F
Feb 20 16:51:25 localhost in.tftpd[3382]: sending NAK (1, File not found) to
192.168.30.63
Feb 20 16:51:26 localhost in.tftpd[3383]: RRQ from 192.168.30.63 filename
pxelinux.cfg/C0A81E3
Feb 20 16:51:26 localhost in.tftpd[3383]: sending NAK (1, File not found) to
192.168.30.63
Feb 20 16:51:26 localhost in.tftpd[3384]: RRQ from 192.168.30.63 filename
pxelinux.cfg/C0A81E
Feb 20 16:51:26 localhost in.tftpd[3384]: sending NAK (1, File not found) to
192.168.30.63
Feb 20 16:51:27 localhost in.tftpd[3385]: RRQ from 192.168.30.63 filename
pxelinux.cfg/C0A81
Feb 20 16:51:27 localhost in.tftpd[3385]: sending NAK (1, File not found) to
192.168.30.63
Feb 20 16:51:28 localhost in.tftpd[3386]: RRQ from 192.168.30.63 filename
pxelinux.cfg/C0A8
Feb 20 16:51:28 localhost in.tftpd[3386]: sending NAK (1, File not found) to
192.168.30.63
Feb 20 16:51:28 localhost in.tftpd[3387]: RRQ from 192.168.30.63 filename
pxelinux.cfg/C0A
Feb 20 16:51:28 localhost in.tftpd[3387]: sending NAK (1, File not found) to
192.168.30.63
Feb 20 16:51:29 localhost in.tftpd[3388]: RRQ from 192.168.30.63 filename
pxelinux.cfg/C0
Feb 20 16:51:29 localhost in.tftpd[3388]: sending NAK (1, File not found) to
192.168.30.63
Feb 20 16:51:29 localhost in.tftpd[3389]: RRQ from 192.168.30.63 filename
pxelinux.cfg/C
Feb 20 16:51:29 localhost in.tftpd[3389]: sending NAK (1, File not found) to
192.168.30.63
Feb 20 16:51:30 localhost in.tftpd[3390]: RRQ from 192.168.30.63 filename
pxelinux.cfg/default
Feb 20 16:51:48 localhost in.tftpd[3676]: RRQ from 192.168.30.63 filename
debian-installer/i386/boot-screens/boot.txt
Feb 20 16:51:58 localhost in.tftpd[3692]: RRQ from 192.168.30.63 filename
linux
Feb 20 16:51:58 localhost in.tftpd[3692]: sending NAK (1, File not found) to
192.168.30.63
Feb 20 16:51:58 localhost in.tftpd[3693]: RRQ from 192.168.30.63 filename
linux.cbt
Feb 20 16:51:58 localhost in.tftpd[3693]: sending NAK (1, File not found) to
192.168.30.63
Feb 20 16:51:58 localhost in.tftpd[3694]: RRQ from 192.168.30.63 filename
linux.0
Feb 20 16:51:58 localhost in.tftpd[3694]: sending NAK (1, File not found) to
192.168.30.63
Feb 20 16:51:58 localhost in.tftpd[3695]: RRQ from 192.168.30.63 filename
linux.com
Feb 20 16:51:58 localhost in.tftpd[3695]: sending NAK (1, File not found) to
192.168.30.63
Feb 20 16:51:58 localhost in.tftpd[3696]: RRQ from 192.168.30.63 filename
linux.c32
Feb 20 16:51:58 localhost in.tftpd[3696]: sending NAK (1, File not found) to
192.168.30.63


But what I get on my soekris console is that

CCoouulldd  nnoo
tt  ffiinndd  kk
eerrnneell  iimm
aaggee::  lliinn
uuxx

bbtt::

I type "linux" as kernel image and it does not boot

The caracters are writing twice and it cannot find the kernel image version

My /pxelinux.cfg/default

have a line like this SERIAL 0,19200 for console connection

And the default image to boot is configured like that in the same file

kernel linux
append vga=normal initrd=initrd.gz console=ttyS0,19200n8
ramdisk_size=14332 root=/dev/rd/0 rw  --


Why is this happening?
How sould I adjust my Hyperteminal?

Thanks

Best Regards.

VoipCrazy
___
Soekris-tech mailing list
Soekris-tech@lists.soekris.com
http://lists.soekris.com/mailman/listinfo/soekris-tech


Re: [Soekris] Soekris PXE boot

2008-02-18 Thread Bill Maas
Errata/Supplement:

On Tue, 2008-02-19 at 08:27 +0100, Bill Maas wrote:
[...]

> http://ftp.debian.org/debian/dists/etch/main/installer-i386/current/images/netboot/
> 
> Select the right distrib dir of course.. You'll also need the matching
> pxelinux.0 sitting next to it.

Not: it's in the tarball.

[...]
> Common trouble:
> - missing hosts.allow entry (I'm not sure if a standalone tftpd
>   actually requires one, but it's always safe)
> - there is any firewall running on the TFTP server blocking TFTP
> - wrong "next-server" in dhcpd.conf
> - files under /tftpboot aren't world-readable (0644 root.root)
>   (on Debian, 'tftpd' runs as user "nobody" I think).
- symblic links from inside /tftpboot to locations outside it
 (chroot..)  ;)

Bill

> 
> That should get you going.
> 
> Bill
> 
> 
> > Thanks in advance.
> > 
> > Best Regards.
> > 
> > VoipCrazy.
> > ___
> > Soekris-tech mailing list
> > Soekris-tech@lists.soekris.com
> > http://lists.soekris.com/mailman/listinfo/soekris-tech
-- 
"What's a computer?" - MES

___
Soekris-tech mailing list
Soekris-tech@lists.soekris.com
http://lists.soekris.com/mailman/listinfo/soekris-tech


Re: [Soekris] Soekris PXE boot

2008-02-18 Thread Bill Maas
Hi Voipcrazy,

On Mon, 2008-02-18 at 18:05 +0100, voip crazy wrote:
> Dear list,
> 
> Five days ago, I started to try boot my soekris, vía PXE, without
> success, my soekris get an IP address and It stop to boot.
> I do not know why It is doing that, I search google, and find some
> links but all of them fails to boot.
> I want to install Debian on it an my soekris is 5501 model.
> 
> http://roland.entierement.nu/pages/debian-on-soekris-howto.html
> 
> Someone, could send me the TFTP files you use to boot a debian
> instalation?
> Or explain me where to get the Debian netboot.tar.gz, to boot the
> soekris 5501?

http://ftp.debian.org/debian/dists/etch/main/installer-i386/current/images/netboot/

Select the right distrib dir of course.. You'll also need the matching
pxelinux.0 sitting next to it. I've run it succesfully as follows:

# mkdir /tftpboot
# tar xzvpf netboot.tar.gz
# mv debian-installer/i386/* /tftpboot

I run tftpd-hpa as a standalone daemon:
--
#/etc/default/tftpd-hpa
#Defaults for tftpd-hpa
RUN_DAEMON="yes"
#OPTIONS="-l -s /var/lib/tftpboot"
OPTIONS="-l -s /tftpboot"
--

You may have to add an entry to /etc/hosts.allow:
--
# /etc/hosts.allow
in.tftpd: LOCAL 192.168.1. : allow
--

Or something similar...

I never managed to get it work through inetd with Ubuntu/Debian
GNU/Linux. Of course no-one will stop you from running it
from /var/lib/tftpboot, but I don't like having persistent files
under /var (while many others abhor having non-FSSTND stuff inside "/" -
whatever..).

My PXE config looks like this:
--
# /tftpboot/pxelinux.cfg/default
SERIAL 0 19200 0
DISPLAY boot.txt
DEFAULT linux
LABEL linux
kernel linux
append vga=normal initrd=initrd.gz console=ttyS0,19200n8
ramdisk_size=14332 root=/dev/rd/0 rw  --
PROMPT 1
TIMEOUT 0
--

Common trouble:
- missing hosts.allow entry (I'm not sure if a standalone tftpd
  actually requires one, but it's always safe)
- there is any firewall running on the TFTP server blocking TFTP
- wrong "next-server" in dhcpd.conf
- files under /tftpboot aren't world-readable (0644 root.root)
  (on Debian, 'tftpd' runs as user "nobody" I think).

That should get you going.

Bill


> Thanks in advance.
> 
> Best Regards.
> 
> VoipCrazy.
> ___
> Soekris-tech mailing list
> Soekris-tech@lists.soekris.com
> http://lists.soekris.com/mailman/listinfo/soekris-tech
-- 
"What's a computer?" - MES

___
Soekris-tech mailing list
Soekris-tech@lists.soekris.com
http://lists.soekris.com/mailman/listinfo/soekris-tech


Re: [Soekris] Soekris PXE boot

2008-02-18 Thread Jochen Grotepass
voip crazy schrieb:
> Dear list,
>
> Five days ago, I started to try boot my soekris, vía PXE, without 
> success, my soekris get an IP address and It stop to boot.
> I do not know why It is doing that, I search google, and find some 
> links but all of them fails to boot.
> I want to install Debian on it an my soekris is 5501 model.
>
> http://roland.entierement.nu/pages/debian-on-soekris-howto.html
>
> Someone, could send me the TFTP files you use to boot a debian 
> instalation?
> Or explain me where to get the Debian netboot.tar.gz, to boot the 
> soekris 5501?
>
> Thanks in advance.
>
> Best Regards.
>
> VoipCrazy.
>   
VoipCrazy, first of all it depends what kind of OS your dhcp-server is. 
If this is a debian Server, read the instructions from Debian carefully:
http://www.debian.org/releases/stable/i386/ch04s06.html.en
One important thing I missed at the beginning was to use the tftp-hpa 
package. The other tftp server (package tftpd) does not work at all for 
pxeboot.

Be careful when reading the settings in pxelinux.cfg/default
These must(!) be modified (if you haven't already done that) with the lines:
SERIAL 0 19200 0
in the first line and
append "console=ttyS0,19200n8" to every "append" line of this file 
(otherwise you don't see anything).

This way, debian netinstall is pretty simple to do (even if OpenBSD is 
far more simpler ;-)

I am running 4 5501-70 with Debian here and 6 with OpenBSD - went 
through all that pain.

Cheers
Jochen

> 
>
> ___
> Soekris-tech mailing list
> Soekris-tech@lists.soekris.com
> http://lists.soekris.com/mailman/listinfo/soekris-tech
>   

___
Soekris-tech mailing list
Soekris-tech@lists.soekris.com
http://lists.soekris.com/mailman/listinfo/soekris-tech


[Soekris] Soekris PXE boot

2008-02-18 Thread voip crazy
Dear list,

Five days ago, I started to try boot my soekris, vía PXE, without success,
my soekris get an IP address and It stop to boot.
I do not know why It is doing that, I search google, and find some links but
all of them fails to boot.
I want to install Debian on it an my soekris is 5501 model.

http://roland.entierement.nu/pages/debian-on-soekris-howto.html

Someone, could send me the TFTP files you use to boot a debian instalation?
Or explain me where to get the Debian netboot.tar.gz, to boot the soekris
5501?

Thanks in advance.

Best Regards.

VoipCrazy.
___
Soekris-tech mailing list
Soekris-tech@lists.soekris.com
http://lists.soekris.com/mailman/listinfo/soekris-tech


Re: [Soekris] OpenBSD Soekris <-> Soekris PXE boot

2008-02-08 Thread Bill Maas
Hi Ervin,

On Thu, 2008-02-07 at 23:37 +0100, ervin wrote:
> guidance for OpenBSD Bootstrap here -> http://www.ultradesic.com/?section=74

Which does not reveal an awful lot about specific trouble and possible
errors.. In fact, I recently tarted up my own doc, which is a bit more
verbose, especially in the error section (which is why I wrote it in the
first place):

http://stsx.xs4all.nl/www.stsx.org/openbsd/obsd-bootsoekris.html

(Added a few lines about firewalling this morning).

Thanks anyway, and Brian as well,

Bill


By the way, I think my initial error was a faulty "next-server", after
that I went into a chain of overlapping fixes/new errors. It happens..


> On 06/02/2008, Brian Candler <[EMAIL PROTECTED]> wrote:
> > On Wed, Feb 06, 2008 at 12:16:45PM +0100, Bill Maas wrote:
> > > I've been running dhcpd/tftpd on OpenBSD 3.8-4.2 on a PC without
> > > problems, for booting my Soekris boxens. However, when I run [almost]
> > > the exact same config on a net4521, and try booting another Soekris box
> > > off that one, TFTP times out. On the boot server, 'inetd -d' shows a
> > > rapid re-execution of tcpd (or of tftpd without using TCP wrappers) when
> > > the client tries to fetch the pxeboot image. The same happens by the way
> > > when I try to fetch files manually using 'tftp' from Linux or OpenBSD,
> > > but those implementations seem to have a reputation for stupidity and
> > > brokenness.
> > >
> > > The really weird thing is that the same config works on a PC as boot
> > > server, but not on a Soekris box. Anyone seen this before?
> >
> > No, but my suggestion would be to disable your OpenBSD pf temporarily and
> > see if the problem goes away.
> >
> > TFTP is a pretty ugly protocol. An exchange looks like this:
> >
> > client   UDP port XXX > UDP port 69server
> >
> >  UDP port XXX < UDP port YYY
> >
> >  UDP port XXX > UDP port YYY   ...etc
> >
> > That is, the server choses a dynamic port for sending its replies from. It's
> > not easy to firewall properly.
> >
> > Maybe the difference between your clients is that one happens to have chosen
> > XXX=69 as its source port, but the other has chosen XXX=dynamic high port
> > (which is the "normal" thing to do). tcpdump could confirm or refute that
> > hypothesis.
> >
> > Regards,
> >
> > Brian.
> > ___
> > Soekris-tech mailing list
> > Soekris-tech@lists.soekris.com
> > http://lists.soekris.com/mailman/listinfo/soekris-tech
> >
> 
> 
> 
-- 
SOD Against NLS

___
Soekris-tech mailing list
Soekris-tech@lists.soekris.com
http://lists.soekris.com/mailman/listinfo/soekris-tech


Re: [Soekris] OpenBSD Soekris <-> Soekris PXE boot

2008-02-07 Thread ervin
guidance for OpenBSD Bootstrap here -> http://www.ultradesic.com/?section=74

On 06/02/2008, Brian Candler <[EMAIL PROTECTED]> wrote:
> On Wed, Feb 06, 2008 at 12:16:45PM +0100, Bill Maas wrote:
> > I've been running dhcpd/tftpd on OpenBSD 3.8-4.2 on a PC without
> > problems, for booting my Soekris boxens. However, when I run [almost]
> > the exact same config on a net4521, and try booting another Soekris box
> > off that one, TFTP times out. On the boot server, 'inetd -d' shows a
> > rapid re-execution of tcpd (or of tftpd without using TCP wrappers) when
> > the client tries to fetch the pxeboot image. The same happens by the way
> > when I try to fetch files manually using 'tftp' from Linux or OpenBSD,
> > but those implementations seem to have a reputation for stupidity and
> > brokenness.
> >
> > The really weird thing is that the same config works on a PC as boot
> > server, but not on a Soekris box. Anyone seen this before?
>
> No, but my suggestion would be to disable your OpenBSD pf temporarily and
> see if the problem goes away.
>
> TFTP is a pretty ugly protocol. An exchange looks like this:
>
> client   UDP port XXX > UDP port 69server
>
>  UDP port XXX < UDP port YYY
>
>  UDP port XXX > UDP port YYY   ...etc
>
> That is, the server choses a dynamic port for sending its replies from. It's
> not easy to firewall properly.
>
> Maybe the difference between your clients is that one happens to have chosen
> XXX=69 as its source port, but the other has chosen XXX=dynamic high port
> (which is the "normal" thing to do). tcpdump could confirm or refute that
> hypothesis.
>
> Regards,
>
> Brian.
> ___
> Soekris-tech mailing list
> Soekris-tech@lists.soekris.com
> http://lists.soekris.com/mailman/listinfo/soekris-tech
>



-- 


mvh/best regards  ervin
___
Soekris-tech mailing list
Soekris-tech@lists.soekris.com
http://lists.soekris.com/mailman/listinfo/soekris-tech


Re: [Soekris] OpenBSD Soekris <-> Soekris PXE boot

2008-02-06 Thread Brian Candler
On Wed, Feb 06, 2008 at 12:16:45PM +0100, Bill Maas wrote:
> I've been running dhcpd/tftpd on OpenBSD 3.8-4.2 on a PC without
> problems, for booting my Soekris boxens. However, when I run [almost]
> the exact same config on a net4521, and try booting another Soekris box
> off that one, TFTP times out. On the boot server, 'inetd -d' shows a
> rapid re-execution of tcpd (or of tftpd without using TCP wrappers) when
> the client tries to fetch the pxeboot image. The same happens by the way
> when I try to fetch files manually using 'tftp' from Linux or OpenBSD,
> but those implementations seem to have a reputation for stupidity and
> brokenness.
> 
> The really weird thing is that the same config works on a PC as boot
> server, but not on a Soekris box. Anyone seen this before?

No, but my suggestion would be to disable your OpenBSD pf temporarily and
see if the problem goes away.

TFTP is a pretty ugly protocol. An exchange looks like this:

client   UDP port XXX > UDP port 69server

 UDP port XXX < UDP port YYY

 UDP port XXX > UDP port YYY   ...etc

That is, the server choses a dynamic port for sending its replies from. It's
not easy to firewall properly.

Maybe the difference between your clients is that one happens to have chosen
XXX=69 as its source port, but the other has chosen XXX=dynamic high port
(which is the "normal" thing to do). tcpdump could confirm or refute that
hypothesis.

Regards,

Brian.
___
Soekris-tech mailing list
Soekris-tech@lists.soekris.com
http://lists.soekris.com/mailman/listinfo/soekris-tech


Re: [Soekris] OpenBSD Soekris <-> Soekris PXE boot

2008-02-06 Thread Bill Maas
Hi Yannick,

it seems you're right: yet another test, with the '-l' removed again,
and:

(I)nstall, (U)pgrade or (S)hell?

I'm still not sure what went wrong in the first place, but at least it
works now.

Thanks.

Bill

On Wed, 2008-02-06 at 08:05 -0500, Yannick Gravel wrote:
> Hi Bill,
> 
>   Except for the first one, I have been doing just that with my net4501, 
> net4801, net5501.
> 
>   Do you run "CURRENT" or some 3.8-4.2 on you net4521? The "-l" option 
> for tftpd is new to 4.3, so if you are not running CURRENT that may be 
> the reason for the "rapid re-execution".
> 
>   Cheers
> 
> Y.
> 
> Bill Maas wrote:
> > Hi,
> >
> > I've been running dhcpd/tftpd on OpenBSD 3.8-4.2 on a PC without
> > problems, for booting my Soekris boxens. However, when I run [almost]
> > the exact same config on a net4521, and try booting another Soekris box
> > off that one, TFTP times out. On the boot server, 'inetd -d' shows a
> > rapid re-execution of tcpd (or of tftpd without using TCP wrappers) when
> > the client tries to fetch the pxeboot image. The same happens by the way
> > when I try to fetch files manually using 'tftp' from Linux or OpenBSD,
> > but those implementations seem to have a reputation for stupidity and
> > brokenness.
> >
> > The really weird thing is that the same config works on a PC as boot
> > server, but not on a Soekris box. Anyone seen this before?
> >
> > Files:
> >
> > /etc/dhcpd.conf:
> > ---
> > #   $OpenBSD: dhcpd.conf,v 1.1 1998/08/19 04:25:45 form Exp $
> > #
> > # DHCP server options.
> > # See dhcpd.conf(5) and dhcpd(8) for more information.
> > #
> >
> > # Network:  192.168.1.0/255.255.255.0
> > # Domain name:  my.domain
> > # Name servers: 192.168.1.3 and 192.168.1.5
> > # Default router:   192.168.1.1
> > # Addresses:192.168.1.32 - 192.168.1.127
> > #
> > #shared-network LOCAL-NET {
> > #   option  domain-name "my.domain";
> > #   option  domain-name-servers 192.168.1.3, 192.168.1.5;
> > #
> > #   subnet 192.168.1.0 netmask 255.255.255.0 {
> > #   option routers 192.168.1.1;
> > #
> > #   range 192.168.1.32 192.168.1.127;
> > #   }
> > #}
> >
> > # Subnetwork declaration
> > subnet 192.168.1.0 netmask 255.255.255.0 {
> > option domain-name "localdomain";
> > option subnet-mask 255.255.255.0;
> > option broadcast-address 192.168.1.255;
> > option domain-name-servers 194.109.6.66, 194.109.104.104;
> > option routers 192.168.1.1;
> > }
> >
> > # Fixed addresses for local hosts
> > # (this is for PXE-booting the boor server itself)
> > group {
> > next-server 192.168.1.2;
> > host net4521 {
> > hardware ethernet 00:00:24:XX:XX:XX;
> > fixed-address 192.168.1.11;
> > filename "pxeboot";
> > }
> > }
> >
> > # (this is for "my" clients)
> > group {
> > next-server 192.168.1.11;
> > host net4801 {
> > hardware ethernet 00:00:24:YY:YY:YY;
> > fixed-address 192.168.1.8;
> > filename "pxeboot";
> > }
> > host net5501 {
> > hardware ethernet 00:00:24:ZZ:ZZ:ZZ;
> > fixed-address 192.168.1.9;
> > filename "pxeboot";
> > }
> > }
> > ---
> >
> > /etc/inetd.conf
> > ---
> > [...]
> > tftpdgram   udp wait
> > root/usr/libexec/tcpd /usr/libexec/tftpd -l -s /tftpboot
> > [...]
> > ---
> >
> > /etc/hosts.allow
> > ---
> > # /etc/hosts.allow - see hosts_options(5)
> > sshd: ALL : allow
> > sendmail: LOCAL 192.168.1. .internal : allow
> > tftpd: LOCAL 192.168.1. .internal : allow
> > ftpd: LOCAL 192.168.1. .internal : allow
> > ALL: ALL : deny
> > ---
> >
> > /etc/hostname.sis0
> > ---
> > inet 192.168.1.11 255.255.255.0 NONE 
> > ---
> >
> > /etc/hostname.sis0
> > ---
> > inet 10.0.1.11 255.0.0.0 NONE 
> > ---
> >
> > (tried booting with and without hostname.sis1 configured - didn't make a
> > difference).
> >
> > # pfctl -s rules  
> > ---
> > scrub in all fragment reassemble
> > block return all
> > block return in quick inet6 all
> > pass out all flags S/SA keep state
> > pass in quick on lo all flags S/S

Re: [Soekris] OpenBSD Soekris <-> Soekris PXE boot

2008-02-06 Thread Bill Maas
Hi Yannick,

On Wed, 2008-02-06 at 08:05 -0500, Yannick Gravel wrote:
> Hi Bill,
> 
>   Except for the first one, I have been doing just that with my net4501, 
> net4801, net5501.
> 
>   Do you run "CURRENT" or some 3.8-4.2 on you net4521? The "-l" option 
> for tftpd is new to 4.3, so if you are not running CURRENT that may be 
> the reason for the "rapid re-execution".

I run 4.2 GENERIC, the '-l' was from later tests, without it the same
thing occurs ("Transfer timed out" on the client side, by the way).

Another big difference between PC and Soekris are of course disk access
and program startup speed (net4521 == slo). Maybe some
timeout, inetd waiting for tcpd, or PXE-tftp waiting for inetd. But I
haven't found anything usable/relevant so far.

Here's what it looks like on the server (net4521):

[EMAIL PROTECTED]:~# inetd -d   
ADD: tftp proto=udp, wait.max=1.256 user:group=root:wheel builtin=0
server=/usr/libexec/tcpd
[...]
someone wants tftp
12472 execv /usr/libexec/tcpd
reaping asked for
12472 reaped, status 100
restored tftp, fd 4
someone wants tftp
8305 execv /usr/libexec/tcpd
reaping asked for
8305 reaped, status 100
restored tftp, fd 4
[...]

More 'someone wants tftp' blocks follow in rapid succession. If the
"connect" succeeds, as on the PC, they appear less often and with more
time in between.


On the client (net4801):

> boot f0

BootManage UNDI, PXE-2.0 (build 082)
BootManage PXE-2.0 PROM 1.0, NATSEC 1.0, SDK 3.0/082 (OEM52)
Copyright (C) 1989,2000 bootix Technology GmbH, D-41466 Neuss.
PXE Software Copyright (C) 1997, 1998, 1999, 2000 Intel Corporation.
Licensed to National Semiconductor

CLIENT MAC ADDR: 00 00 24 C2 A3 F0  
CLIENT IP: 192.168.1.4  MASK: 255.255.255.0  DHCP IP: 192.168.1.11
GATEWAY IP: 192.168.1.1 
PXE-E32: TFTP open timeout
TFTP..

I also wonder what that last line means, nothing seems to happen on the
server side - I'd rather have my prompt back;).

Bill

>   Cheers
> 
> Y.
> 
> Bill Maas wrote:
> > Hi,
> >
> > I've been running dhcpd/tftpd on OpenBSD 3.8-4.2 on a PC without
> > problems, for booting my Soekris boxens. However, when I run [almost]
> > the exact same config on a net4521, and try booting another Soekris box
> > off that one, TFTP times out. On the boot server, 'inetd -d' shows a
> > rapid re-execution of tcpd (or of tftpd without using TCP wrappers) when
> > the client tries to fetch the pxeboot image. The same happens by the way
> > when I try to fetch files manually using 'tftp' from Linux or OpenBSD,
> > but those implementations seem to have a reputation for stupidity and
> > brokenness.
> >
> > The really weird thing is that the same config works on a PC as boot
> > server, but not on a Soekris box. Anyone seen this before?
> >
[...]
-- 
"What's a computer?" - MES

___
Soekris-tech mailing list
Soekris-tech@lists.soekris.com
http://lists.soekris.com/mailman/listinfo/soekris-tech


Re: [Soekris] OpenBSD Soekris <-> Soekris PXE boot

2008-02-06 Thread Yannick Gravel
Hi Bill,

  Except for the first one, I have been doing just that with my net4501, 
net4801, net5501.

  Do you run "CURRENT" or some 3.8-4.2 on you net4521? The "-l" option 
for tftpd is new to 4.3, so if you are not running CURRENT that may be 
the reason for the "rapid re-execution".

  Cheers

Y.

Bill Maas wrote:
> Hi,
>
> I've been running dhcpd/tftpd on OpenBSD 3.8-4.2 on a PC without
> problems, for booting my Soekris boxens. However, when I run [almost]
> the exact same config on a net4521, and try booting another Soekris box
> off that one, TFTP times out. On the boot server, 'inetd -d' shows a
> rapid re-execution of tcpd (or of tftpd without using TCP wrappers) when
> the client tries to fetch the pxeboot image. The same happens by the way
> when I try to fetch files manually using 'tftp' from Linux or OpenBSD,
> but those implementations seem to have a reputation for stupidity and
> brokenness.
>
> The really weird thing is that the same config works on a PC as boot
> server, but not on a Soekris box. Anyone seen this before?
>
> Files:
>
> /etc/dhcpd.conf:
> ---
> #   $OpenBSD: dhcpd.conf,v 1.1 1998/08/19 04:25:45 form Exp $
> #
> # DHCP server options.
> # See dhcpd.conf(5) and dhcpd(8) for more information.
> #
>
> # Network:  192.168.1.0/255.255.255.0
> # Domain name:  my.domain
> # Name servers: 192.168.1.3 and 192.168.1.5
> # Default router:   192.168.1.1
> # Addresses:192.168.1.32 - 192.168.1.127
> #
> #shared-network LOCAL-NET {
> #   option  domain-name "my.domain";
> #   option  domain-name-servers 192.168.1.3, 192.168.1.5;
> #
> #   subnet 192.168.1.0 netmask 255.255.255.0 {
> #   option routers 192.168.1.1;
> #
> #   range 192.168.1.32 192.168.1.127;
> #   }
> #}
>
> # Subnetwork declaration
> subnet 192.168.1.0 netmask 255.255.255.0 {
> option domain-name "localdomain";
> option subnet-mask 255.255.255.0;
> option broadcast-address 192.168.1.255;
> option domain-name-servers 194.109.6.66, 194.109.104.104;
> option routers 192.168.1.1;
> }
>
> # Fixed addresses for local hosts
> # (this is for PXE-booting the boor server itself)
> group {
> next-server 192.168.1.2;
> host net4521 {
> hardware ethernet 00:00:24:XX:XX:XX;
> fixed-address 192.168.1.11;
> filename "pxeboot";
> }
> }
>
> # (this is for "my" clients)
> group {
> next-server 192.168.1.11;
> host net4801 {
> hardware ethernet 00:00:24:YY:YY:YY;
> fixed-address 192.168.1.8;
> filename "pxeboot";
> }
> host net5501 {
> hardware ethernet 00:00:24:ZZ:ZZ:ZZ;
> fixed-address 192.168.1.9;
> filename "pxeboot";
> }
> }
> ---
>
> /etc/inetd.conf
> ---
> [...]
> tftpdgram   udp wait
> root/usr/libexec/tcpd /usr/libexec/tftpd -l -s /tftpboot
> [...]
> ---
>
> /etc/hosts.allow
> ---
> # /etc/hosts.allow - see hosts_options(5)
> sshd: ALL : allow
> sendmail: LOCAL 192.168.1. .internal : allow
> tftpd: LOCAL 192.168.1. .internal : allow
> ftpd: LOCAL 192.168.1. .internal : allow
> ALL: ALL : deny
> ---
>
> /etc/hostname.sis0
> ---
> inet 192.168.1.11 255.255.255.0 NONE 
> ---
>
> /etc/hostname.sis0
> ---
> inet 10.0.1.11 255.0.0.0 NONE 
> ---
>
> (tried booting with and without hostname.sis1 configured - didn't make a
> difference).
>
> # pfctl -s rules  
> ---
> scrub in all fragment reassemble
> block return all
> block return in quick inet6 all
> pass out all flags S/SA keep state
> pass in quick on lo all flags S/SA keep state
> block drop in on ! lo inet from 127.0.0.0/8 to any
> block drop in on ! lo inet6 from ::1 to any
> block drop in inet from 127.0.0.1 to any
> block drop in on ! sis0 inet from 192.168.1.0/24 to any
> block drop in inet from 192.168.1.111 to any
> block drop in inet6 from ::1 to any
> block drop in on lo0 inet6 from fe80::1 to any
> block drop in on sis0 inet6 from fe80::200:24ff:fec1:ef50 to any
> pass in quick on sis1 all flags S/SA keep state
> pass in inet proto icmp from  to any icmp-type echoreq code 0 keep
> sta

[Soekris] OpenBSD Soekris <-> Soekris PXE boot

2008-02-06 Thread Bill Maas
Hi,

I've been running dhcpd/tftpd on OpenBSD 3.8-4.2 on a PC without
problems, for booting my Soekris boxens. However, when I run [almost]
the exact same config on a net4521, and try booting another Soekris box
off that one, TFTP times out. On the boot server, 'inetd -d' shows a
rapid re-execution of tcpd (or of tftpd without using TCP wrappers) when
the client tries to fetch the pxeboot image. The same happens by the way
when I try to fetch files manually using 'tftp' from Linux or OpenBSD,
but those implementations seem to have a reputation for stupidity and
brokenness.

The really weird thing is that the same config works on a PC as boot
server, but not on a Soekris box. Anyone seen this before?

Files:

/etc/dhcpd.conf:
---
#   $OpenBSD: dhcpd.conf,v 1.1 1998/08/19 04:25:45 form Exp $
#
# DHCP server options.
# See dhcpd.conf(5) and dhcpd(8) for more information.
#

# Network:  192.168.1.0/255.255.255.0
# Domain name:  my.domain
# Name servers: 192.168.1.3 and 192.168.1.5
# Default router:   192.168.1.1
# Addresses:192.168.1.32 - 192.168.1.127
#
#shared-network LOCAL-NET {
#   option  domain-name "my.domain";
#   option  domain-name-servers 192.168.1.3, 192.168.1.5;
#
#   subnet 192.168.1.0 netmask 255.255.255.0 {
#   option routers 192.168.1.1;
#
#   range 192.168.1.32 192.168.1.127;
#   }
#}

# Subnetwork declaration
subnet 192.168.1.0 netmask 255.255.255.0 {
option domain-name "localdomain";
option subnet-mask 255.255.255.0;
option broadcast-address 192.168.1.255;
option domain-name-servers 194.109.6.66, 194.109.104.104;
option routers 192.168.1.1;
}

# Fixed addresses for local hosts
# (this is for PXE-booting the boor server itself)
group {
next-server 192.168.1.2;
host net4521 {
hardware ethernet 00:00:24:XX:XX:XX;
fixed-address 192.168.1.11;
filename "pxeboot";
}
}

# (this is for "my" clients)
group {
next-server 192.168.1.11;
host net4801 {
hardware ethernet 00:00:24:YY:YY:YY;
fixed-address 192.168.1.8;
filename "pxeboot";
}
host net5501 {
hardware ethernet 00:00:24:ZZ:ZZ:ZZ;
fixed-address 192.168.1.9;
filename "pxeboot";
}
}
---

/etc/inetd.conf
---
[...]
tftpdgram   udp wait
root/usr/libexec/tcpd /usr/libexec/tftpd -l -s /tftpboot
[...]
---

/etc/hosts.allow
---
# /etc/hosts.allow - see hosts_options(5)
sshd: ALL : allow
sendmail: LOCAL 192.168.1. .internal : allow
tftpd: LOCAL 192.168.1. .internal : allow
ftpd: LOCAL 192.168.1. .internal : allow
ALL: ALL : deny
---

/etc/hostname.sis0
---
inet 192.168.1.11 255.255.255.0 NONE 
---

/etc/hostname.sis0
---
inet 10.0.1.11 255.0.0.0 NONE 
---

(tried booting with and without hostname.sis1 configured - didn't make a
difference).

# pfctl -s rules  
---
scrub in all fragment reassemble
block return all
block return in quick inet6 all
pass out all flags S/SA keep state
pass in quick on lo all flags S/SA keep state
block drop in on ! lo inet from 127.0.0.0/8 to any
block drop in on ! lo inet6 from ::1 to any
block drop in inet from 127.0.0.1 to any
block drop in on ! sis0 inet from 192.168.1.0/24 to any
block drop in inet from 192.168.1.111 to any
block drop in inet6 from ::1 to any
block drop in on lo0 inet6 from fe80::1 to any
block drop in on sis0 inet6 from fe80::200:24ff:fec1:ef50 to any
pass in quick on sis1 all flags S/SA keep state
pass in inet proto icmp from  to any icmp-type echoreq code 0 keep
state
pass in log proto tcp from  port = bootps to (sis0) port = bootpc
flags S/SA keep state
pass in log proto tcp from  to (sis0) port = ftp flags S/SA keep
state
pass in log proto tcp from  to (sis0) port > 49151 flags S/SA keep
state
pass in log proto udp from  port = bootps to (sis0) port = bootpc
keep state
pass in log proto udp from  to (sis0) port = tftp keep state
pass in on sis0 proto tcp from any to (sis0) port = ssh flags S/SA keep
state
pass in log on sis0 proto tcp from any to (sis0) port = smtp flags S/SA
keep state
pass out log on sis0 proto