Hi,

I have three machines on a network.

1. Old Dell running e-smith 5.0 (192.168.1.1 - gw1)
2. HP running Red Hat 7.2 (192.168.1.2 - ahnold)
3. IBM which has a 3Com PXE 2 boot ROM and runs either Windows 98SE or NT 4 
Workstation (192.168.1.11 - ibm2)

Before the HP came into the picture, I was able to netboot the IBM from the 
Dell using the recipe from 
<http://cui.unige.ch/info/pc/remote-boot/howto.html>. The Dell was serving 
the BpBatch interpreter via tftp which I used to image and reimage the 
drive on the IBM. In order to do this, the IBM was forced to boot from the 
PXE boot ROM. So, we know that the boot ROM is fine and the system works.

I am ultimately shooting for being able to load any version of Windows or 
Linux on the IBM from images that are stored on the HP. In order to do 
this, I need to be able to run MrZip to create the disk images of the IBM 
in the first place. I figured booting a minimal diskless Linux environment 
was the best way to approach this. I have used ltsp <http://www.ltsp.org> 
before so I did a base installation of ltsp on the HP.

Here is the problem. I have the stock tftpd that came with RH 7.2 running 
on the HP. I am not running dhcp on the HP because I want to keep the dhcp 
server separate from the tftp server. The dhcp server continues to run on 
the Dell running e-smith. When I boot the IBM client from the PXE boot ROM, 
it manages to find the dhcp server and acquire an address, which is no 
surprise since it could do that in the original scenario. However, it 
cannot find the tftp server. Here is what the IBM reports:

CLIENT IP: 192.168.1.11
MASK: 255.255.255.255.0
DHCP IP: 192.168.1.2 (This is the HP but I have no idea what this is 
supposed to signify.)
GATEWAY IP: 192.168.1.1

PXE-E32: TFTP open timeout

I started tcpdump on the HP to see what was going on and the IBM keeps 
looking for the tftp server on the e-smith box. It should not be because I 
have specified in dhcpd.conf that it should be looking at the HP for the 
tftp server by using the next-server parameter. Here is dhcpd.conf:

---------------------------
# Beginning of dhcpd.conf
option dhcp-class-identifier "PXEClient";
option vendor-encapsulated-options 01:04:00:00:00:00:ff;

subnet 192.168.1.0 netmask 255.255.255.0 {
     option broadcast-address    192.168.1.255;
     allow bootp;
     option domain-name          "dinamis.com";
     option domain-name-servers  192.168.1.1;
     default-lease-time          86400;
     max-lease-time              604800;
     option subnet-mask          255.255.255.0;
     range                       192.168.1.40 192.168.1.250;
     option routers              192.168.1.1;
     }

host ahnold {
     hardware ethernet 00:10:83:f9:ed:c8;
     fixed-address 192.168.1.2;
     }

host ibm2 {
     hardware ethernet 00:04:76:e0:f6:cf;
     fixed-address 192.168.1.11;
     next-server 192.168.1.2;
     filename "bpbatch";
     option option-135 "-i";
     }
# End of dhcpd.conf
---------------------------

According to man dhcpd.conf,

>The next-server statement is used to specify the host address of the 
>server from which the initial boot file (specified in the filename 
>statement) is to be loaded. Server-name should be a numeric IP address or 
>a domain name. If no next-server parameter applies to a given client, the 
>DHCP server's IP address is used.

I think I have explicitly specified that the boot file, bpbatch, should be 
loaded from 192.168.1.2, and yet, the client insists on looking for it at 
192.168.1.1. The client is behaving as if it did not even know that the HP 
exists. What am I doing wrong? Any ideas?

Regards,

Clifford Ilkay
Dinamis Corporation
3266 Yonge Street, Suite 1419
Toronto, Ontario
Canada M4N 3P6

Tel: 416-410-3326

mailto:[EMAIL PROTECTED]


_____________________________________________________________________
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.openprojects.net

Reply via email to