I have a multi-use RHEL 5.5 host which I want to place our DHCP service on.
At first, I thought it would be as simple as binding the DHCP instance to that interface and walla. Turns out, I don't think it's quite that simple. I had problems trying to get it working.
I started playing with the route tables, and here is where I think there may be some adjustments needed.
10.111.0.0 * 255.255.255.0 U 0 0 0 eth0
10.112.0.0 * 255.255.255.0 U 0 0 0 eth1
default 10.101.0.1 0.0.0.0 UG 0 0 0 eth0
The table currently looks like that, and I know that's wrong. I had made other adjustments and ultimately couldn't get the broadcasts to work correctly and gave up.
There are 6 interfaces on this system. I found that you can't expect Linux (RHEL) to take a NIC card (Intel, in this case) and have each port be in a numerical order! That was one problem.
In any case, I believe I must be overlooking something rather simple. I am revisiting this issue as I want to get the DHCP service onto this machine -- but I'm curious if anyone might share some insight as to how to get this working correctly and perhaps what I may be doing wrong.
Here is the dhcpd.conf - which is pretty basic.
Thanks,
Forrest
authoritative;
server-identifier 10.123.0.18;
# Route DHCP broadcasts only on the appropriate interface
option ip-forwarding off;
server-name "dhcp.our.domain";
ddns-update-style none;
ignore client-updates;
ping-check true;
ping-timeout 3;
subnet 10.123.0.0 netmask 255.255.255.0 {
range dynamic-bootp 10.123.0.20 10.123.0.100;
option routers 10.123.0.1;
option subnet-mask 255.255.255.0;
option domain-name "subdom.our.domain";
option domain-name-servers 10.111.0.32, 10.111.0.33;
option time-offset -18000; # Eastern Standard Time
option ntp-servers 10.111.0.32;
default-lease-time 86400;
max-lease-time 259200;
}
At first, I thought it would be as simple as binding the DHCP instance to that interface and walla. Turns out, I don't think it's quite that simple. I had problems trying to get it working.
I started playing with the route tables, and here is where I think there may be some adjustments needed.
10.111.0.0 * 255.255.255.0 U 0 0 0 eth0
10.112.0.0 * 255.255.255.0 U 0 0 0 eth1
default 10.101.0.1 0.0.0.0 UG 0 0 0 eth0
The table currently looks like that, and I know that's wrong. I had made other adjustments and ultimately couldn't get the broadcasts to work correctly and gave up.
There are 6 interfaces on this system. I found that you can't expect Linux (RHEL) to take a NIC card (Intel, in this case) and have each port be in a numerical order! That was one problem.
In any case, I believe I must be overlooking something rather simple. I am revisiting this issue as I want to get the DHCP service onto this machine -- but I'm curious if anyone might share some insight as to how to get this working correctly and perhaps what I may be doing wrong.
Here is the dhcpd.conf - which is pretty basic.
Thanks,
Forrest
authoritative;
server-identifier 10.123.0.18;
# Route DHCP broadcasts only on the appropriate interface
option ip-forwarding off;
server-name "dhcp.our.domain";
ddns-update-style none;
ignore client-updates;
ping-check true;
ping-timeout 3;
subnet 10.123.0.0 netmask 255.255.255.0 {
range dynamic-bootp 10.123.0.20 10.123.0.100;
option routers 10.123.0.1;
option subnet-mask 255.255.255.0;
option domain-name "subdom.our.domain";
option domain-name-servers 10.111.0.32, 10.111.0.33;
option time-offset -18000; # Eastern Standard Time
option ntp-servers 10.111.0.32;
default-lease-time 86400;
max-lease-time 259200;
}