Solved DHCP Reservation Config

---

Hello ITPro TV ! I have a question that came up after working with DHCP in my lab.. Working from the CCENT course 100-105 specifically. I have a router setup with the following DHCP config.

ip dhcp excluded-address 192.168.80.1 192.168.80.99

ip dhcp pool 192.168.80.0
network 192.168.80.0 255.255.255.0
default-router 192.168.80.1
dns-server 192.168.99.20
domain-name mydomain.local

ip dhcp pool 192.168.80.144
host 192.168.80.144 255.255.255.0
hardware-address 68f7.2878.6eb2
default-router 192.168.80.1
dns-server 192.168.99.20
domain-name mydomain.local

Things seemed to be working great until I configured the 2nd scope to test DHCP reservations. For some reason the PC that is supposed to get the 192.168.80.144 address still seems to obtain an address out of the general scope. Here is the output of the "show ip dhcp binding" command:

R2#show ip dhcp binding
Bindings from all pools not associated with VRF:
IP address Client-ID/ Lease expiration Type
Hardware address/
User name
192.168.80.100 0168.f728.786e.b2 Oct 11 2016 01:47 PM Automatic
192.168.80.144 68f7.2878.6eb2 Infinite Manual

Notice how the the same MAC has 80.100, but the system added this funny "01" to the front of the MAC.. If I do a clear ip dhcp binding on the 80.100 address,. then ipconfig /release - /renew on the client it just keeps gettting the 80.100 address.

Thanks for clarifying where I went wrong!

-Adam

ip dhcp pool 192.168.80.0
network 192.168.80.0 255.255.255.0
default-router 192.168.80.1
dns-server 192.168.99.20
domain-name mydomain.local
ip dhcp pool 192.168.80.144
host 192.168.80.144 255.255.255.0
hardware-address 68f7.2878.6eb2
default-router 192.168.80.1
dns-server 192.168.99.20
domain-name mydomain.local

Things seemed to be working great until I configured the 2nd scope to test DHCP reservations. For some reason the PC that is supposed to get the 192.168.80.144 address still seems to obtain an address out of the general sco Here is the output of the "show ip dhcp binding" command:

R2#show ip dhcp binding
Bindings from all pools not associated with VRF:
IP address Client-ID/ Lease expiration Type
Hardware address/
User name
192.168.80.100 0168.f728.786e.b2 Oct 11 2016 01:47 PM Automatic
192.168.80.144 68f7.2878.6eb2 Infinite Manual

Notice how the the same MAC has 80.100, but the system added this funny "01" to the front of the MAC.. If I do a clear ip dhcp binding on the 80.100 address,. then ipconfig /release - /renew on the client it just keeps gettting the 80.100 address.

Thanks for clarifying where I went wrong!

-Adampe.

Your scopes are overlapping. 192.168.80.100/24 and 192.168.80.144/24 are two addresses in the 192.168.80.0/ 24 network. If you want to use something like this you must subnet that address space and use exclusions to help you. For example:
You could do 192.168.80.0/25 and 192.168.80.128/25. Make exclusions for .1-99, for the first network to give you the 192.168.80.100 through 192.168.80.126 For the Second pool, you could create exclusions for .129-143.
Also you would need different default gateways for both ranges, one within each of the /25 networks to allow them to connect out. With your configuration, the router doesn't exactly know that you have a second pool because it's within the first pool. This is from your subnet mask being a /24 for both pools.

Ah, thanks. I didn't realize that the MAC reservation wouldn't automatically clue the router in that it needed to assign the address in the second pool even though they overlapped. I'm used to working with the Windows Server DHCP service which can assign these reservations out of an existing scope/network while it still remained a member of the same network and used the same default gateway.

Just to confirm, ANY DHCP reservation you create on the Cisco router has to be a separate subnet/network for each reservation? Is this just the nature of the DHCP service being somewhat limited on the routers? For branch office and small deployments mostly?

Ps.. As a test I deleted the 192.168.80.0 pool entirely and it still doesn't seem to be assigning the reserved IP address to my client. Did I make any mistakes or leave anything off of my reservation DHCP pool?

ip dhcp pool 192.168.80.144
host 192.168.80.144 255.255.255.0
hardware-address 68f7.2878.6eb2
default-router 192.168.80.1
dns-server 192.168.99.20
domain-name mydomain.local

-Adam

@Adam-Tyler,

I think you may have to give the pool a name. I'm not sure if the 192.168.80.144 would be considered a name or not...I've never configured it this way.

This configuration seems like you're trying to do a static IP address assignment to the hardware-address 68f7.2878.6eb2 you may find that you have to configure the hardware-address differently using something like a client-id which is a bit more complex.
You may have to check the syntax of the for doing this dhcp binding for that device. I don't have access to a cisco switch right now at home to check. but recheck both those ideas:

name
and syntax for the static binding.

Thanks.. I changed the pool name to something a little more friendly and tried using the "client-identifier", still not quite working.. I enabled DHCP debugging and am getting this output when the client checks for an IP. Just getting a 169 address now...

Oct 10 16:02:09 PDT: DHCPD: Sending notification of DISCOVER:
Oct 10 16:02:09 PDT: DHCPD: htype 1 chaddr 68f7.2878.6eb2
Oct 10 16:02:09 PDT: DHCPD: remote id 020a0000c0a80a0201000001
Oct 10 16:02:09 PDT: DHCPD: circuit id 00000000
Oct 10 16:02:09 PDT: DHCPD: Seeing if there is an internally specified pool class:
Oct 10 16:02:09 PDT: DHCPD: htype 1 chaddr 68f7.2878.6eb2
Oct 10 16:02:09 PDT: DHCPD: remote id 020a0000c0a80a0201000001
Oct 10 16:02:09 PDT: DHCPD: circuit id 00000000
Oct 10 16:02:09 PDT: DHCPD: there is no address pool for 192.168.80.1.

It's interesting that it is stating there is no address pool for 192.168.80.1. This is the interface IP of the router connected to that 192.168.80.0/24 network. The DHCP request is coming from a client workstation out of that network.

-Adam

Breakthrough!! I managed to get this working, and with overlapping subnets! I get an award for persistence. | .. So, above in this thread I mentioned that the output of the command "show ip dhcp binding" displayed my client MAC address as "01"68.f728.786e.b2.. Turns out that I actually had to use "01" in the client-identifier. I'm not exactly sure why at this point. I found a different post that must have dealt with different hardware as it explained that you had to include another command after client-identifier 68f7.2878.6eb2 (ethernet) or (IEEE...something.).. My 2911 lab router didn't support this however. The MAC address was the end of the command. The client in my test was a Windows 10 workstation.

So below is how I have my DHCP pool(s) configured. If I change the MAC address from 68f7.2878.6eb2 to something else on the local adapter it chooses the first available IP in the scope 192.168.80.0/24 network. When the MAC address is moved back it selects the 192.168.80.144 IP after releasing and renewing the address.

ip dhcp pool testres
host 192.168.80.144 255.255.255.0
client-identifier 0168f7.2878.6eb2
default-router 192.168.80.1
dns-server 192.168.99.20
domain-name domain.local
!
ip dhcp pool 192.168.80.0
network 192.168.80.0 255.255.255.0
default-router 192.168.80.1
dns-server 192.168.99.20
domain-name domain.local

Credit to this thread for pointing me in the right direction: http://www.routerie.com/cgi-bin/ultimatebb.cgi?ubb=get_topic;f=50;t=001115

-Adam

@Adam-Tyler,

Well glad you figured it out. If you're studying for CCNA you're going beyond the requirement. which is great!

I've implemented this when working with certain thin clients . This took forever when I created static reservations for 50 - 100 thin clients from Wyse. You'll not want to do this more than once in your life. I wouldn't want to do it again.