kickstart install of redhat enterprise 3 via pxe fails

[tom van overbeke]

Hi,

I'm trying to setup a dhcp/tftp server for
automatically installing linux rh es3 on our clients.

I've setup dhcpd correctly because the client gets the
ip address i reserved for it based on the mac address,
the kernel and initrd (i used the ones from the
pxeboot directory) are correctly transferred to the
client, it starts to boot, mounts the root filesystem
and then it stops, the last line it prints is:
Freeing unused kernel memory

It's supposed to fetch the kickstart file on the
server and start installing, but it doesn't even make
an attempt to do so: i verified this by using http to
serve the installation files and the access_log and
error_log files  are empty. (before that, i tried it
with nfs and it seemed to mount the share but also
stopped at the same point)

this is my dhcpd.conf:

ddns-update-style ad-hoc;
ignore unknown-clients;
not authoritative;

allow bootp;
allow booting;

subnet 172.16.6.0 netmask 255.255.255.0 {
option routers 172.16.6.1;
}

group {
next-server 172.16.6.40;
filename "pxelinux.0";

host test1 {
        hardware ethernet 00:10:DC:9C:3B:91;
        fixed-address 172.16.6.41;
           }

host hpbak1 {
        hardware ethernet 00:11:85:E7:1B:22;
        fixed-address 172.16.6.42;
           }
}

my tftp directory looks like this:

[root at proliant /]# ls -lR tftpboot/
tftpboot/:
total 3380
-rw-r--r--    1 root     root      2538426 Dec 16
20:17 initrd.img
-rw-r--r--    1 root     root            0 Feb 25
10:48 Kickstart_end
-rw-r--r--    1 root     root        11304 Feb 25
10:32 pxelinux.0
drwxr-xr-x    2 root     root         4096 Feb 25
15:50 pxelinux.cfg
-rw-r--r--    1 root     root       895078 Dec 16
20:17 vmlinuz

tftpboot/pxelinux.cfg:
total 4
-rw-r--r--    1 root     root          232 Feb 25
15:50 default


the default file looks like this:

default linux
serial 0,38400n8
label linux
kernel vmlinuz
append ksdevice=eth0 console=tty0 console=ttyS0,38400
\ load_ramdisk=1 ramdisk_size=8192 root=/dev/ram0 \
initrd=initrd.img network
ks=http://172.16.6.40:/Kickstart/ks-rhes.cfg


the relevant lines on the server's messages files are
these:


1) when i used nfs:

Feb 25 13:55:18 proliant in.tftpd[19932]: RRQ from
172.16.6.41 filename vmlinuz
Feb 25 13:55:19 proliant in.tftpd[19933]: RRQ from
172.16.6.41 filename initrd.img
Feb 25 14:55:37 proliant dhcpd: DHCPDISCOVER from
00:10:dc:9c:3b:91 via eth0
Feb 25 14:55:37 proliant dhcpd: DHCPOFFER on
172.16.6.41 to 00:10:dc:9c:3b:91 via eth0
Feb 25 14:55:37 proliant dhcpd: DHCPDISCOVER from
00:10:dc:9c:3b:91 via eth0
Feb 25 14:55:37 proliant dhcpd: DHCPOFFER on
172.16.6.41 to 00:10:dc:9c:3b:91 via eth0
Feb 25 14:55:37 proliant dhcpd: DHCPREQUEST for
172.16.6.41 (172.16.6.40) from 00:10:dc:9c:3b:91
 via eth0
Feb 25 14:55:37 proliant dhcpd: DHCPACK on 172.16.6.41
to 00:10:dc:9c:3b:91 via eth0
Feb 25 14:55:47 proliant rpc.mountd: authenticated
mount request from xxxxxxxx:612 for /n
etinstall/Kickstart (/netinstall)
Feb 25 14:56:02 proliant rpc.mountd: authenticated
mount request from xxxxxxxx:616 for /n
etinstall/RedHat (/netinstall)


and when i tried it with http: simply this:

Feb 25 14:55:45 proliant in.tftpd[20128]: RRQ from
172.16.6.41 filename vmlinuz
Feb 25 14:55:46 proliant in.tftpd[20129]: RRQ from
172.16.6.41 filename initrd.img


Any pointers ?


thanks,

Tom.

PS. While the machine is doing nothing, i can ping it
on the ip address that the dhcp server offered, so
network communications still seem to be ok at that
point.



	
	
		
___________________________________________________________ 
ALL-NEW Yahoo! Messenger - all new features - even more fun! http://uk.messenger.yahoo.com