On 20 March 2006 at 13:47, Anselm Martin Hoffmeister <[EMAIL PROTECTED]> wrote:
> Am Montag, den 20.03.2006, 07:37 -0500 schrieb Susan G. Kleinmann:
> > On 20 March 2006 at 13:11, Anselm Martin Hoffmeister <[EMAIL PROTECTED]
> wrote:
> > > Am Montag, den 20.03.2006, 06:55 -0500 schrieb Susan G. Kleinmann:
> > > > I have installed ltsp-server-standalone, version 0.82debian2 on 
> > > > a machine running Linux 2.6.15.  I then used ltsp-build-client to build 
> > > > a client system suitable for hosting thin clients.  I have succeeded in 
> > > > getting two thin clients (both VIA MII 6000 diskless computers) to boot 
> > > > using this system.  The problem is, this only works once!!!
> OK, I don't know about any non-labelled buttons, but I'm most probably
> using a completely different set of surrounding software. Which login
> manager is that (kdm, gdm, xdm, wdm, ...)?
The login manager that installs with the Debian ltsp-build-client package is
ldm.  In its package there is in fact a file named "shutdown.png", and this is
probably just what the unlablled button is doing.

> For clients not booting, you could try to find out what happens when the
> client machines are powered up. This should make several DHCP messages
> appear in your system log file, provided it runs properly here. Could
> you check wether anything like "DHCPDISCOVER" appears
> in /var/log/syslog, or /var/log/messages, or /var/log/daemon.log, when
> booting up the terminal? Just in case, is there any difference in wether
> the boot succeeds or not? You might have to reboot the server for
> booting to work again, I'd expect.
Well, the 2nd boot pass never gets into any DHCP questions from the server
or responses; it stops before that stage.  If I subsequently boot up some other
machine using DHCP (not a thin client, just a machine whose kernel image is
stored on the DHCP/TFTP server), then all is well.  

> Can you state which exact Linux distribution you use, and which dhcpd
> version, along with the /etc/dhcpd.conf
> (or /etc/dhcp3-server/dhcpd.conf)?
The server is using Debian with its 'testing' a.k.a. etch distribution;
but the ltsp-server package and packages it depends upon were obtained from
the 'unstable' distribution (they are not currently in 'testing').

But all of the above may be irrelevant:
I've found a note saying that there's a bug in kernels later than 2.6 with 
wake-on-lan, which affects certain via rhine chips, which is exactly what is in 
these boards.  The trick is to google for: "Via Rhine WOL vs PXE".  There is 
apparently some disagreement on how to deal with the problem; in the meantime, 
the quick way to get back to a working system is to unplug it from the wall 
for at least 15 seconds.  

For now, that's a good-enough fix for me, but I do very much appreciate your
quick responses.

Susan


-------------------------------------------------------
This SF.Net email is sponsored by xPML, a groundbreaking scripting language
that extends applications into web and mobile media. Attend the live webcast
and join the prime developer group breaking into this new coding territory!
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=110944&bid=241720&dat=121642
_____________________________________________________________________
Ltsp-discuss mailing list.   To un-subscribe, or change prefs, goto:
      https://lists.sourceforge.net/lists/listinfo/ltsp-discuss
For additional LTSP help,   try #ltsp channel on irc.freenode.net

Reply via email to