Searched the forum and saw this error many times but no fixes have helped me.

Setup:
isolated lan with:
Server1: dhcp server and tftpboot server

Server2 boots with Network boot from bios. gets hung at PXE-E32" TFTP Open time out:

TFTP....


It gets an IP from Server1 so I know its communicating.


Server1: selinux is disabled
chkconfig --list | grep tft
tftp: on

chkconfig --list | grep dhcp
dhcpd: on


root 2247 1 0 17:45 ? 00:00:00 xinetd -stayalive -pidfile /var/run/xinetd.pid

root 2704 1 0 18:12 ? 00:00:00 /usr/sbin/dhcpd
root 2857 2545 0 18:23 tty1 00:00:00 grep dhcp

dhcpd.conf
ddns-update-style interim;
ignore client-updates;
allow booting;
allow bootp;
subnet 10.1.1.0 netmask 255.255.255.0 {

# --- default gateway
option routers 10.1.1.1;
option subnet-mask 255.255.255.0;
filename "pxelinux.0";

#option nis-domain "domain.org";
#option domain-name "domain.org";
#option domain-name-servers 192.168.1.1;

#option time-offset -18000; # Eastern Standard Time
# option ntp-servers 192.168.1.1;
# option netbios-name-servers 192.168.1.1;
# --- Selects point-to-point node (default is hybrid). Don't change this unless
# -- you understand Netbios very well
# option netbios-node-type 2;

range dynamic-bootp 10.1.1.128 10.1.1.254;
default-lease-time 21600;
max-lease-time 43200;

# we want the nameserver to appear at a fixed address
#host ns {
# next-server marvin.redhat.com;
# hardware ethernet 12:34:56:78:AB:CD;
# fixed-address 207.175.42.254;
#}
}


default:

default vesamenu.c32
prompt 1
timeout 0
display boot.msg
F1 boot.msg
F4 fedora.msg
label local
localboot 0
append -


label centosserver
kernel centos51_i386/vmlinuz
append ks=http://10.1.1.66/CentOS/5.1/ks/ks-server initrd=centos51_i386/initrd.img ramdisk_size=8192 selinux=0 text


ks-server=
/LinuxPXE/Centos/5.1/ks/ks-server


also copied the ks-server to:
/tftpboot/Centos/5.1/ks/ks-server

Server1:
netstat -an | fgrep -w 69

udp 0 0 0.0.0.0:69 0.0.0.0:*



any suggestions?? thanks!


Both servers are running Centos 5.1

more info

nmap 10.1.1.66 -sU

Starting Nmap 4.20 ( http://insecure.org ) at 2008-04-23 08:08 EDT
Interesting ports on 10.1.1.66:
Not shown: 1480 closed ports
PORT STATE SERVICE
67/udp open|filtered dhcps
69/udp open|filtered tftp
111/udp open|filtered rpcbind
631/udp open|filtered unknown
857/udp open|filtered unknown
860/udp open|filtered unknown
5353/udp open|filtered zeroconf
32768/udp open|filtered omad

Nmap finished: 1 IP address (1 host up) scanned in 1.349 seconds


also came back with some DNS errors but I was not able to capture them, went by too quickly.



/etc/xinetd.d/tftp

# default: off
# description: The tftp server serves files using the trivial file transfer \
# protocol. The tftp protocol is often used to boot diskless \
# workstations, download configuration files to network-aware printers, \
# and to start the installation process for some operating systems.
service tftp
{
disable = no
socket_type = dgram
protocol = udp
wait = yes
user = root
server = /usr/sbin/in.tftpd
server_args = -s /tftpboot
per_source = 11
cps = 100 2
flags = IPv4
}


Whats the "filtered" mean in the nmap scan?



EDIT:

system-config-securitylevel-tui
disable firewall


Now it is stuck at:
tftp prefix:
Trying to load: pxelinux.cfg/"mac address"
Trying to load: pxelinux.cfg/0a0101fe
Trying to load: pxelinux.cfg/0a0101f
Trying to load: pxelinux.cfg/0a0101
Trying to load: pxelinux.cfg/0a010
Trying to load: pxelinux.cfg/0a01
Trying to load: pxelinux.cfg/0a0
Trying to load: pxelinux.cfg/0a
Trying to load: pxelinux.cfg/0
Trying to load: pxelinux.cfg/default
Could not find kernel image: linux
boot

created a boot.msg in /tftpboot (copied from an old pxe boot server, I plan on adding centos51 once I get it going)
0bAvailable boot commands:

0aDOS0f - Start a DOS environment (with tools)
0aFC60f - Start Fedora Core 6 Installer/Updater [see note]
0aMEMTEST0f - Start memtest86

07Pressing <ENTER> will attempt to boot from the local drive

Note: For Fedora notes and options, press F4



default:
default vesamenu.c32
prompt 1
timeout 0
display boot.msg
F1 boot.msg
F4 fedora.msg
label local
localboot 0
append -

label fc6
kernel fc6/vmlinuz
append ks=http://10.1.1.66/FC6/core/i386/ks.fc6 initrd=fc6/initrd.img ramdisk_size=8192 selinux=0




"mac address" is the actual mac address, just substituted for privacy




still getting stuck at trying to load pxelinux.cfg









EDIT:
added next-server <10.1.1.66>;
to dhcpd.conf and it gives me the boot menu but it can't find the ks file. Does httpd need to be running in order to server the ks file?
Our old pxe server has httpd running and our location in default to the ks file is:

append ks=http://10.1.1.66/CentOS/5.1/ks/ks.install initrd=centos51_i386/initrd.img ramdisk_size=8192 selinux=0 text

but I get the Error downloading kickstart file:









NEW ERROR!

I always love it when I find a new error. At least this one gives me hope.

checked /var/log/messages and see this when I try to boot from the PC i want to install.

in.tftpd[2882]: tftp: client does not accept options.

its followed by a few DHCP messages
DHCPDISCOVER
DHCPOFFER
DHCPREQUEST
DHCPACK


then on the PC that fails to load it crashes:
loaded received SIGSEGV! BACKTRACE
"some stuff"
install exited abnormally.
umounting some filesystems

you may safely reboot your system.

At least now it gets me to my boot menu and loads the ata_pixx driver, then probes the ethernet card, then crashes.

Still having problems.

Gets to the boot menu screen. I put in my label name it loads some stuff, it gets to the "/sbin/loader" the loads the ata_piix



label info in "default"
label centosserver
kernel centos51_i386/vmlinuz
append ks=http://10.1.1.66/CentOS/5.1/ks/ks-server initrd=centos51_i386/initrd.img ramdisk_size=8192 selinux=0 text
kernel centos51_i386/vmlinuz
append ks=http://10.1.1.66/CentOS/5.1/ks/ks-games initrd=centos51_i386/initrd.img ramdisk_size=8192 selinux=0 text


httpd.conf alias definitions
Alias /icons/ "/var/www/icons/"
Alias /CentOS/ "/LinuxPXE/CentOS"
Alias /CentOS51_i386/ "/LinuxPXE/CentOS/5.1/"
Alias /FC6/ "/storage/FC6/"
ScriptAlias /cgi-bin/ "/var/www/cgi-bin/"
Alias /error/ "/var/www/error/"



/var/log/messages:

Apr 24 14:21:06 hostname dhcpd: DHCPDISCOVER from 00:07:e9:9e:38:e3 via eth1
Apr 24 14:21:07 hostname dhcpd: DHCPOFFER on 10.1.1.247 to 00:07:e9:9e:38:e3 via eth1
Apr 24 14:21:08 hostname dhcpd: DHCPREQUEST for 10.1.1.247 (10.1.1.66) from 00:07:e9:9e:38:e3 via eth1
Apr 24 14:21:08 hostname dhcpd: DHCPACK on 10.1.1.247 to 00:07:e9:9e:38:e3 via eth1
Apr 24 14:21:09 hostname xinetd[3064]: START: tftp pid=5989 from=10.1.1.247
Apr 24 14:21:09 hostname in.tftpd[5990]: tftp: client does not accept options
Apr 24 14:21:28 hostname dhcpd: DHCPDISCOVER from 00:07:e9:9e:38:e3 via eth1
Apr 24 14:21:28 hostname dhcpd: DHCPOFFER on 10.1.1.247 to 00:07:e9:9e:38:e3 via eth1
Apr 24 14:21:28 hostname dhcpd: DHCPREQUEST for 10.1.1.247 (10.1.1.66) from 00:07:e9:9e:38:e3 via eth1
Apr 24 14:21:28 hostname dhcpd: DHCPACK on 10.1.1.247 to 00:07:e9:9e:38:e3 via eth1



httpd/access_log
10.1.1.247 "GET /CentOS/5.1/ks/ks-server HTTP/1.0 302 299 "-" anaconda/11.1.2.87"



WTF is wrong???







EDIT:

I removed the ks= from my default file

label centos
kernel centos51_i386/vmlinuz
append initrd=centos51_i386/initrd.img ramdisk_size=8192 selinux=0 text

it loaded vmlinuz and initrd.img but it was doing that before.

It asked for kb, lang, etc and moved along fine till it hit the installation method. When I told it to use HTTP, it asked for server/location

10.1.1.66
CentOS/5.1/core/i386

then it crashed at the same spot



The next time I booted I chose Local CDROM as the boot method and its installing packages as we speak.


When using HTTP method:
HTTPD is running and DHCP is running. I see in the logs that an IP was given out and a GET request was received for the Centos install

ugh....


figured it out...



http was redirecting to https!


working fine now. thanks!