PXE with IP Helpers/DHCP Relay

I'm a Sysadmin and I have a question about what is best practice in regards to PXE servers. We are currently using DHCP Options for PXE clients (options 66,67). This works for most clients but is not the recommended method from either of the vendors we have used (Microsoft or Symantec). They recommend using IP Helpers / DHCP relay to forward the DHCP discover request to the PXE servers so that the PXE server is getting the actual request. This is more of an issue now with UEFI-based machines where the boot file would be different based on if the client is UEFI.
My Network team is against using IP Helpers and thinks it can cause issues. This doesn't seem to make much sense to me, as from what I understand, all that happens is both the DHCP server and the PXE servers get the DHCP discover and respond with their relevant info. Can someone clarify what, if any, issues there are using multiple IP helpers/DHCP relay with PXE Servers like SCCM & Altiris? Is this not standard practice?

It's very common to use DHCP relays (IP helpers) in order to centralize DHCP infrastructure. Larger organizations will frequently use this approach in order to avoid having to manually edit DHCP configurations at the router or switch level. Having a few servers with a central DHCP configuration for all segments is a good management proposition.
In most environments, there isn't a problem with doing this, but it is a major architectural consideration and not something you just turn on without consideration. This is largely because DHCP works on a broadcast principle. The clients are going to broadcast for the first DHCP server that answers with an acceptable offer, which they will take. If you have a mixture of local DHCP servers and relays, the local servers will respond faster and may not provide the configuration you want to deploy... at best. At worst, you will have a mix of acceptable responses and a lot of potential for conflicting addresses. On any network segment where you're using DHCP relays, the local server needs to be disabled.
It might be worthwhile going back to your network team and asking what sorts of "issues" that they feel the implementation of DHCP relays would cause. There may be something unique to your environment that makes them reluctant to pursue this approach.

Similar Messages

Setting up a DHCP relay agent

Hello,
I'm trying to setup a relay agent for an XP client to obtain configuration through 2 routers on a VM LAB
I have 3 Segments/subnets 1,2 and 3
the topology is the following:
1- server 2008 R2 AD DS DC on subnet 1 (192.168.1.0) and a DHCP server with 2 scopes
Internal 192.168.1.0 (subnet1)
external 192.168.3.0 (subnet 3)
2- server 2008 R2 with RRAS installed
Interface 1 pointing to the internal subnet1 192.168.1.0
Interface 2 pointing to subnet2 192.168.2.0
3- server 2003 with RRAS installed
Interface 1 pointing to subnet2 192.168.2.0
Interface 2 pointing to subnet3 192.168.3.0
relay agent installed on Interface 2
(servers/Routers 2 and 3 running RIP v2)
4- XP client on subnet3 (192.168.3.0) and the client trying to obtain config.
The XP client is unable to contact the DHCP to obtain config.
server 2003 relay agent receiving requests with no replies.
How do i get this to work?

Hi,
First, let’s see how DHCP relay agent works:
1. The DHCP client broadcasts a DHCPDISCOVER packet.
2. The DHCP relay agent on the client’s subnet forwards the DHCPDISCOVER message to the DHCP server by using unicast.
3. The DHCP server uses unicast to send a DHCPOFFER message to the DHCP relay agent.
4. The DHCP relay agent broadcasts the DHCPOFFER packet to the DHCP client’s subnet.
5. The DHCP client broadcasts a DHCPREQUEST packet.
6. The DHCP relay agent on the client’s subnet forwards the DHCPREQUEST message to the DHCP server by using unicast.
7. The DHCP server uses unicast to send a DHCPACK message to the DHCP relay agent.
8. The DHCP relay agent broadcasts the DHCPACK to the DHCP client’s subnet.
We can see in the second step, DHCP relay agent send unicast to DHCP server after receiving DHCP request. So confirm unicast communication between DHCP server
and DHCP relay agent works fine. At least ping should be working. You can use the following commend to add the route entry.
Add a static IP route
http://technet.microsoft.com/en-us/library/cc757323(v=ws.10).aspx
The result should be based on your test. If it doesn’t work it just indicates that we cannot configure another DHCP relay agent behind a relay agent.
Hope this helps.

DHCP Relay forwarded to Secondary when Scope is not available in Primaray

Two ip helper-addresses (let suppose DHCPServer1 and DHCPServer2) are defined on each of the branch router, and customer want to divide the load of DHCP request on two different DHCP Servers. The propose solution by customer is to disable some scope from one DHCP Server (DHCPServer1) and define the similar scope in second DHCP Server (DHCPServer2). Does the DHCP Relay request would be forwarded to secondary server (DHCPServer2), if the scope is disabled on first DHCP Server (DHCPServer1)?

The DHCP request is forwarded to all the addresses defined with the ip helper-address command.
So if you have 2 ip helper-addresses then the DHCP request is sent to both at the same time by the router. First one to respond is usually the one accepted by the client.
Jon

DHCP Relay Cisco SG500X

Hi,
I've create 2 vlan in a cisco SG500X-24 and a DHCP server on vlan 2. I just want to dhcp server assign ip to devices on vlan 3. I've configured the vlan and dhcp server relay commands.
ip dhcp relay address 192.168.1.11
ip dhcp relay enable
ip dhcp information option
interface vlan 2
ip address 192.168.1.250 255.255.255.0
interface vlan 3
ip address 192.168.51.254 255.255.255.0
ip dhcp relay enable
The dhcp server gets the request from pc, and sends a new address, but the offer packet not comes to device. With Wireshark a see like offer dhcp packet can't jump to vlan 3.
It's the first time a work with SMB series, and this never happens with catalyst. I'm turning crazy.
Anyone can help me? Thank you in advances.
Victor.

Hi,
Yes, also I configured ip dhcp relay on intefaces.
Yesterday I found the trouble. I was using the tftpd32 dhcp server, and I tested with a Windows DHCP server and everything works like a champ. I didn't know what have do, and i turned crazy, so I began to change every element on solution, finally the DHCP server that it was the key.
Thank you so much for your answer.
KR!

WRVS4400Nv2 DHCP Relay on 2nd VLAN

Hi,
Here's what I'm trying to figure out:
My network is set up such that I have a Wireless Network in VLAN 1, which is the primary network that we use. The subnet is 10.5.1.x.
My goal is to set up a completely isolated Guest Wireless Network, however it would work best. What I am trying to do now is I created a seperate VLAN (VLAN 2, IP range 10.5.2.x) and turned on DHCP on the WRVS4400N. However, in the Guest Network, it is always picking up a 10.5.1.x IP which is handed out by the DHCP server (10.5.1.5, Win 2003) and still routing all of the traffic to/from our private network.
Here's What I have set:
Wireless>Security Settings>Guest Network (SSID 2)
Wireless Isolation (between SSID w/o VLAN): Enabled
Wireless Isolation (within SSID): Enabled
Setup>LAN>VLAN 1
Router IP 10.5.1.1, WLAN IP 10.5.1.3
DHCP Relay for 10.5.1.5
Setup>LAN>VLAN 2
Router IP 10.5.2.1
DHCP Enabled for 10.5.2.x subnet
DHCP Relay option is grayed out (not sure why)
Setup>Advanced Routing
Inter-VLAN Routing: Disabled
Any way to solve this would be fine. I just do not want traffic routing through our internal network. Ideally, if I could get the Windows server to hand out 10.5.2.x addresses, that would be perfect, but I'm not sure how to configure it for such.
If anyone has any ideas, that'd be great- thanks!
Matt

Yes...here's an answer I got from Cisco's Engineering support:
The issue you reported is a know issue.
Engineering and development are aware of this issue, and have provided the following information:
PROBLEM DESCRIPTION:
If the WRVS4400N is configured with multiple VLANs, and these VLANs are mapped to different SSID, the user cannot use an external DHCP server to provide IP scopes for these VLANs.
Hosts connected to both SSID will obtain IP address from native DHCP server only.
The workaround for this is to use the embedded DHCP server for all VLANs defined on the WRVS4400N.
Note: This is not considered a bug but rather a product limitation. The developer has confirmed the WRVS4400N is functioning as designed.
Regarding a fix:
Due to wireless and trunk switch port using different chip set, it is not possible to provide a fix for this issue.
In future product, Engineering & Dev teams will strive to use the same chip set (same vendor).
This functionality has been targeted for next new Product. No fix will be made on the current hardware.
Note: If this feature/function is mission critical to your deployment, and you would like to recover the cost of the WRVS4400N, please forward the serial number and a copy of the proof of purchase, and we will gladly provide a refund.
Best regards,
Alex Delano

DHCP Relay using Brocade Switches

Hi
I have a large project with 3 UCM cluster with unity cluster and UCCX
The network is a Brocade switch environement,
The Core is using OSPF and distribution is Layer 2.
I have configured the Cluster with a dedicated DHCP and TFTP service.
DHCP relay is not working, but when i configure one of the phones with a static IP address it registers and I have full functionality
When I connect my server dirrectly to the core which is not the design then the dhcp relay works and I get an IP address, but when i traverse the layer2 then i do not get an address.
In the same topology I connected a Windows DHCP server on the same vlan as my UCM cluster and change the relay address to point to the windows dhcp the i do get an address.
In an additional test i configured the same setup on cisco switches then the relay works great.
If anyone has seen or knows of any bug regarding DHCP relay i would be greatful for info
Thanks
Lance

Hi Experts,
i forgot to mention that i was reading an interesting document on Cisco website "network virtualization design guide", and they clearly mentioned the below:
"""VRF-awareness for DHCP-relay functionality is currently not supported on any Catalyst platform, but it is required only for supporting overlapping IP addresses"""
So i would like to ask you if you have any workaround to be done in such deployments
Thank you in advance
Samer Labaky
CCIE # 24675

DHCP relay for SGE2010

hi
I have SGE2010 switch in layer 3 mode and im unable to assign IP addresses in second VLAN 2.
My setup
Port 1: TRUNK , VLAN 1 and VLAN 2 --> connected to Cisco 887 FE3 Trunk with VLAn 1 and VLAN 2 and DHCP pool for VLAN2
Port 2: TRUNK , VALN1 and WLAN2 --> Cisco Aironet 1040 LAN WIFI VLAN 1 and Guest WIFI VLAN2
Port 3: Access , VLAN1 Windows DHCP server for VLAN 1 subnet
My LAN WIFI clients can get IP from Windows DHCP server for VLAN 1
My Guest WIFI clients on VLAN 2 cant get an IP from the Cisco 887 router on VLAN 2.
I did try turning on DHCP relay etc but didnt make any difference.
Can anyone give me some pointers on wat im missing ?
Thank you

Hi, if the symptom is that when you assign an IP address to vlan 2 the switch "locks up" then the reason is because the vlan 1 did not have an IP address assigned by user.
To fix that, you'd need to add an IP address for vlan 1 as you like then try to make an IP for the additional vlans.
-Tom
Please mark answered for helpful posts

SGE2000 DHCP Relay Issue

I am looking for some help with DHCP Relay on a SGE2000 switch.
I have configured two VLANs on the switch, VLAN2 (192.168.10.x/24) and VLAN3 (192.168.9.x/24). I have the switch in Layer 3 mode. I have configured the DHCP relay server of 192.168.10.4 and the DHCP Interfaces as VLAN3. All of the IP Static Routes were generated by the switch.
If I put a client computer on a port that is Untagged VLAN 3 and try to get a DHCP address from the server on an Untagged VLAN 2 port I never get a response back.
I have done some packet captures and here is what I have found:
I see the DHCP broadcast on the client computer
I see the DHCP Request on the DHCP server coming from the IP assiged to the switch on VLAN 3 (192.168.9.254)
I see the DHCP server respond with a DHCP Offer
The DHCP offer never gets to the client computer
I can't seem to get a DHCP address to any system not on the same VLAN as the DHCP server. Option 82 is disable and I did try enabling it, which made no difference.
Any help would be great.
Thanks,
Phil

Hi phil,
Have you created a static route on the DHCP server that points back to the 192.168.9.0/24 network.
The gateway for that network, from the DHCP relay servers perspective, is the VLAN2 IP address of the SGE2000 switch.
i would think that if you tried to ping the VLAN3 switch IP address from the DHCP server now, you will not get a reply.
When you create a static or persistant route in the DHCP server, you then should be able to ping VLAN3 IP address of the switch.
regards Dave

VWLC DHCP relay flooding

Hello.
I have a big problem with a vWLC installation and the DHCP relay. I have a interface for guest access in central Switching mode where is a DHCP Server address which is not in the same subnet. The Proxy mode is on, and the vWLC can reach the DHCP over its default gateway on the Interface. When a clients connects to ssid it sends a DHCP request and the controller sends it with its interface ip to the DHCP. So far so good, but the problem is that the vWLC does not send one packet, it sends thousends of same DHCP requests to the Server. The DHCP Server also ansers with thousends of DHCP acks.
When i give the DHCP which is also a vm, a interface in the guest vlan everthing works correct. Does anybody know if there is a problem with DHCP relay over Layer 3?
Regards Stefan
I use the latest verion of vWLC.

Hello.
I have a big problem with a vWLC installation and the DHCP relay. I have a interface for guest access in central Switching mode where is a DHCP Server address which is not in the same subnet. The Proxy mode is on, and the vWLC can reach the DHCP over its default gateway on the Interface. When a clients connects to ssid it sends a DHCP request and the controller sends it with its interface ip to the DHCP. So far so good, but the problem is that the vWLC does not send one packet, it sends thousends of same DHCP requests to the Server. The DHCP Server also ansers with thousends of DHCP acks.
When i give the DHCP which is also a vm, a interface in the guest vlan everthing works correct. Does anybody know if there is a problem with DHCP relay over Layer 3?
Regards Stefan
I use the latest verion of vWLC.

Sonicwall DHCP relay not working

I recently set up a new vlan and am trying to get the dhcp server on the existing vlan to issue IPs. I have no trouble getting dhcp working with the sonicwall as the server, but I can't get it to pass it along to the actual server. I've set up the IP of the server as a trusted DHCP relay host but I don't see anything in the logs for the relay for the test host on the vlan. I tagged the port the server is connected to for the new vlan so I'm lost as to why it isn't working. DNS also isn't working using that server for the lookup. I am able to ping it by IP though. I'm sure it's something simple I'm missing but vlans are new to me so I'm still learning.
This topic first appeared in the Spiceworks Community

I recently set up a new vlan and am trying to get the dhcp server on the existing vlan to issue IPs. I have no trouble getting dhcp working with the sonicwall as the server, but I can't get it to pass it along to the actual server. I've set up the IP of the server as a trusted DHCP relay host but I don't see anything in the logs for the relay for the test host on the vlan. I tagged the port the server is connected to for the new vlan so I'm lost as to why it isn't working. DNS also isn't working using that server for the lookup. I am able to ping it by IP though. I'm sure it's something simple I'm missing but vlans are new to me so I'm still learning.
This topic first appeared in the Spiceworks Community

Switches that do dhcp relay (ip helper address)

Hi all,
I'm looking at this switch the 1810 24g and I was wondering whether it dose dhcp relay (ip helper address) ie forward dhcp addresses from the dhcp server to the switch?
Many thanks
Rob

Hi,
on all the vlan interfaces where the DHCP server doesn't reside you must configure
interface Vlan
description Client VLAN
ip helper-address
When a client now sends DHCP request, the router
forwards this request to the ip helper-address.
Because the router also puts in its own interface ip address as source ip, the packet finds the way back.
DHCP uses UDP port 67 and 68. With the command ip helper-address, there are also some other ports which are opened for udp. To close this ports you configure (global command)
no ip forward-protocol udp tftp
no ip forward-protocol udp nameserver
no ip forward-protocol udp domain
no ip forward-protocol udp time
no ip forward-protocol udp netbios-ns
no ip forward-protocol udp netbios-dgm
no ip forward-protocol udp tacacs
On your DHCP Server you have to configure a scope for each ip subnet.
If your dhcp server is located at ther server vlan, do NOT configure a helper-address there.
For the migration I would suggest to use two different ip subnets. Image all your clients are now in VLAN2 10.2.0.0/16. If you have this IP subnet on your Router you can't add a new VLAN with 10.2.1.0/24, because this overlaps.
So make the new VLANs with 10.3.1.0/24, 10.3.2.0/24, ... and move the clients to the new
vlans by change the vlan of the port where the PC is conneted to. When you then reboot the PC it shoud get a new ip from the dhcp and everything sould be fine.
Bye
Jo

ASA Hub-and-spoke VPN dhcp-relay

Hi!
Have anyone implemented a solution with a hub-and-spoke IPSEC VPN (running ASA) with dhcp relay for the inside clients on the spoke. With the DHCP server on the hub site?
Normal LAN-LAN IPSEC VPN is a bit cumbersome to configure something like below:
SPOKE
<snip>
access-list CRYPTO_ALLOWED extended permit ip INSIDE-NETWORKS any
#ALL INTERNET ACCESS GOES THROUGH THE SPOKE SITE
access-list CRYPTO_ALLOWED extended permit udp host OUTSIDE_IF_ADDR host HUB_DHCP_SERVER_ADDR eq bootps
access-list CRYPTO_ALLOWED extended permit udp host OUTSIDE_IF_ADDR host HUB_DHCP_SERVER_ADDR eq bootpc
nat (INSIDE,OUTSIDE) source static CRYPTO_ALLOWED CRYPTO_ALLOWED destination static OSKO-INTERNET OSKO-INTERNET route-lookup
dhcprelay DHCP-SERVER outside
dhcprelay enable INSIDE
dhcprelay setroute INSIDE
dhcprelay timeout 60
HUB
<snip>
access-list CRYPTO_ALLOWED_TO_SPOKE extended permit ip 0.0.0.0 0.0.0.0 HUB_NETWORKS
access-list CRYPTO_ALLOWED_TO_SPOKE extended permit udp host HUB_DHCP_SERVER_ADDR host SPOKE_OUTSIDE_ADDR eq 67
access-list CRYPTO_ALLOWED_TO_SPOKE extended permit udp host HUB_DHCP_SERVER_ADDR host SPOKE_OUTSIDE_ADDR eq 68
nat (INSIDE,OUTSIDE) source static ANY ANY destination static SPOKE_NETWORKS SPOKE_NETWORKS
nat (INSIDE,OUTSIDE) source static HUB_DHCP_SERVER_ADDR HUB_DHCP_SERVER_ADDR destination static SPOKE_OUTSIDE_ADDR SPOKE_OUTSIDE_ADDR
### HUB INTERNET ACCESS ##
nat (OUTSIDE,OUTSIDE) source dynamic SPOKE_NETWORKS interface
I can't really apply this to a hub-and-spoke configuration.
Any ideas?
Regards
Daniel

Thanks. That's what I thought. I'm trying to configure this a my lab and having trouble though. Here's what I am trying to accomplish: HUB should communication with spoke1 and spoke2 via ipsec vpn using their own internal addresses HUB: 192.100.10.0/24, SPOKE1 10.142.0.0/24, SPOKE2 10.25.0.0/24) Communication between SPOKE1 and SPOKE2 should be nat'ed by the HUB so SPOKE2's addresses appear to be 172.16.128.0/24. SPOKE1's interesting traffic rule will allow the entire 172.16.128.0 255.255.128.0 subnet. Any new SPOKE's will use another subnet of that network. In my head I think I might need to let SPOKE2 NAT it's own traffic before it gets to HUB, but I'm dealing with multiple different devices as spokes so I want to handle everything on the HUB. Ideally the HUB would translate all traffic in both directions so both business partners and clients would only need one supernet in their interesting traffic rules.

DHCP relay issues - WLC4400 series

Hi all,
I'm experiencing some strange problems with my WLC 4400 – and hope you guys can give me a hand.
There is an issue while connecting a WLAN Client to the WLC for the first time. I pinpointed the source of the problem to the dhcp, but I wondering why this happens…
As stated above – the issue occurs only during the first time registration of a WLAN client with the WLC. If I do another registration right after the failed connection attempt, the session is established and I can start working in my network environment.
Because we use 802.1x authentication, my first idea was that there is an issue – but the authentication process completes successfully.
Another debug for the dhcp process showed an issue during the initial registration process. I'll paste an extract of the NOT working connection attempt below (DHCP DISCOVER msg and DHCP OFFER msg passed successfully – I'll focus on the DHCP REQUEST msg):
###### Extract one ######
Tue Mar 9 09:51:31 2010: 00:21:6a:00:35:9c dhcpProxy: Received packet: Client 00:21:6a:00:35:9c
                        DHCP Op: BOOTREQUEST(1), IP len: 303, switchport: 29, encap: 0xec03
Tue Mar 9 09:51:31 2010: 00:21:6a:00:35:9c dhcp option len, including the magic cookie = 67
Tue Mar 9 09:51:31 2010: 00:21:6a:00:35:9c dhcp option: received DHCP REQUEST msg
Tue Mar 9 09:51:31 2010: 00:21:6a:00:35:9c dhcp option: skipping option 61, len 7
Tue Mar 9 09:51:31 2010: 00:21:6a:00:35:9c dhcp option: requested ip = 10.64.153.66
Tue Mar 9 09:51:31 2010: 00:21:6a:00:35:9c dhcp option: server id = 1.1.1.1
Tue Mar 9 09:51:31 2010: 00:21:6a:00:35:9c dhcp option: skipping option 12, len 12
Tue Mar 9 09:51:31 2010: 00:21:6a:00:35:9c dhcp option: vendor class id = MSFT 5.0 (len 8)
Tue Mar 9 09:51:31 2010: 00:21:6a:00:35:9c dhcp option: skipping option 55, len 12
Tue Mar 9 09:51:31 2010: 00:21:6a:00:35:9c dhcpParseOptions: options end, len 67, actual 67
Tue Mar 9 09:51:31 2010: 00:21:6a:00:35:9c dhcpProxy: dhcp request, client: 00:21:6a:00:35:9c:
                        dhcp op: 1, port: 29, encap 0xec03, old mscb port number: 29
Tue Mar 9 09:51:31 2010: 00:21:6a:00:35:9c Determing relay for 00:21:6a:00:35:9c
                                                                                                        dhcpServer: 10.49.143.8, dhcpNetmask: 0.0.0.0,
                        dhcpGateway: 0.0.0.0, dhcpRelay: 0.0.0.0 VLAN: 0
Tue Mar 9 09:51:31 2010: 00:21:6a:00:35:9c Relay settings for 00:21:6a:00:35:9c
                                                                                                        Local Address: 0.0.0.0, DHCP Server: 10.49.143.8,
                        Gateway Addr: 10.64.153.1, VLAN: 0, port: 29
Tue Mar 9 09:51:31 2010: 00:21:6a:00:35:9c dhcpProcessPacket return an error,chaddr: 00:21:6a:00:35:9c
The process stops working after the last line above. The client reports connection successfully, but no IP address was assigned to the client. A second connection attempt was successful (again – I'll focus on the dhcp REQUEST msg – ignoring DISCOVER, OFFER and ACK msg):
                        DHCP Op: BOOTREQUEST(1), IP len: 303, switchport: 29, encap: 0xec03
Tue Mar 9 09:53:02 2010: 00:21:6a:00:35:9c dhcp option len, including the magic cookie = 67
Tue Mar 9 09:53:02 2010: 00:21:6a:00:35:9c dhcp option: received DHCP REQUEST msg
Tue Mar 9 09:53:02 2010: 00:21:6a:00:35:9c dhcp option: skipping option 61, len 7
Tue Mar 9 09:53:02 2010: 00:21:6a:00:35:9c dhcp option: requested ip = 10.64.153.66
Tue Mar 9 09:53:02 2010: 00:21:6a:00:35:9c dhcp option: server id = 1.1.1.1
Tue Mar 9 09:53:02 2010: 00:21:6a:00:35:9c dhcp option: skipping option 12, len 12
Tue Mar 9 09:53:02 2010: 00:21:6a:00:35:9c dhcp option: vendor class id = MSFT 5.0 (len 8)
Tue Mar 9 09:53:02 2010: 00:21:6a:00:35:9c dhcp option: skipping option 55, len 12
Tue Mar 9 09:53:02 2010: 00:21:6a:00:35:9c dhcpParseOptions: options end, len 67, actual 67
Tue Mar 9 09:53:02 2010: 00:21:6a:00:35:9c dhcpProxy: dhcp request, client: 00:21:6a:00:35:9c:
                        dhcp op: 1, port: 29, encap 0xec03, old mscb port number: 29
Tue Mar 9 09:53:02 2010: 00:21:6a:00:35:9c Determing relay for 00:21:6a:00:35:9c
                                                                                                        dhcpServer: 10.49.143.8, dhcpNetmask: 0.0.0.0,
                        dhcpGateway: 0.0.0.0, dhcpRelay: 10.64.153.6 VLAN: 300
Tue Mar 9 09:53:02 2010: 00:21:6a:00:35:9c Relay settings for 00:21:6a:00:35:9c
                                                                                                        Local Address: 10.64.153.6, DHCP Server: 10.49.143.8,
The major difference seems to be in line 16:
Not Working:
                        dhcpGateway: 0.0.0.0, dhcpRelay: 0.0.0.0 VLAN: 0
Working:
                        dhcpGateway: 0.0.0.0, dhcpRelay: 10.64.153.6 VLAN: 300
For me it seems that the WLC is not able to forward this request to the appropriate dhcp server.
Does anyone of you have an idea, why this happens? And why does this happen only during the first time login of every client? Or am I misinterpreting the debug output?!
Thx a lot in advance!
Cheers
Martin

Hi,
thx for your comment so far.
I did some additional troubleshooting yesterday and I guess I fixed the problem. The management interface was configured with two dhcp server IPs (0.0.0.0 and 1.1.1.1).
Within the Cisco documentation it is stated that the dhcp relay proxy feature uses a virtual IP 1.1.1.1.
0.0.0.0    seems to be used for the internal communication.
When I changed the dhcp address (primary & secondary) to IP 1.1.1.1 the problem was solved. We tested it yesterday evening and this morning.
My assumption is that the virtual 1.1.1.1 IP is mandatory to match the dhcp responses to the proxy relaying feature. Or the WLC uses the DHCP addresses on the management interface to forward the traffic to the appropriate feature (where 1.1.1.1 triggers the proxy feature and 0.0.0.0 is used to forward the traffic to the internal dhcp service). But this is just a guesswork – I do not know the Cisco WLAN good enough to provide a valuable explanation.
Cheers
Martin

Windows DHCP Server and Linux DHCP Relay Agent

We are trying to organize a VLAN (say VLAN 1) for guests who must be assigned IP addresses from a DHCP server in a different VLAN (VLAN 2). This DHCP server is configured with two scopes - 172.16.0.0/24 (for VLAN 2) and 172.16.4.0/24 (for the Guests
VLAN 1). The DHCP server successfully distributes addresses to clients in its VLAN (it has the IP address 172.16.0.2). For the clients in the other VLAN a DHCP Relay Agent has been setup on the router. It is DHCRELAY running on Linux (CentOS) which has
been configured to accept the DHCPDISCOVER broadcasts coming on the VLAN1 interface of the router and forward these to the DHCP server. The IP address of the VLAN1 interface of the router is 172.16.4.254 and on the VLAN2 interface - 172.16.0.254
The problem is that the DHCP server won't respond with a DHCPOFFER message to the relay agent. I have traced the frames on the router and on the DHCP server. They arrive on the DHCP server with the correct GIADDR of the relay agent. According to all documentation,
if a scope has been configured on the DHCP server and it receives a unicast message with the GIADDR set by a relay agent that matches one of the configured scopes, the DHCP server must send a unicast DHCPOFFER to the relay agent. But it doesn't.
Here is what Wireshark reports (ignore the Destination port unreachable messages, the DHCP service was stopped at the time Wireshark was running)
When the service is running, there are just DHCPDISCOVERs - no OFFER. You can see that the server has the two scopes configured:
The relay agent seems to work normally - it forwards the DHCPDISCOVERs to the server continuously (tried many times with ipconfig /renew on the client).
I read many posts about this problem. Some users had other services running on the DHCP server that used the DHCP port, but I don't have such an issue (you see that when the service is stopped, an ICMP port unreachable is sent which is correct). Others however
did not find a solution. Am I missing something? Is there something specific when using the DHCRELAY agent from DHCPD? Can I turn on some verbose logging to track this down? Thanks in advance.

WIth DHCP, there is really nothing to configure. If the Relay Agent/IP Helper is pointing to it, and the VLAN subnet exactly matches the scope subnet, then it should just work.
What I've seen in the VLAN config is either a static route back to the subnet the DHCP server itself is sitting on is not configured or incorrectly configured, or there are ports blocked (need UDP, too, since that's what DHCP uses to pass the OFFER), and
other necessary ports are opened, then it should just work.
Sometimes NIC teaming on the DHCP server will cause it. Not sure. Microsoft doesn't support teaming prior to Windwos 2012, but it doesn't mean that it doesn't work. Don't get me wrong, teaming works nicely, but they just don't support it because they never
certified the drivers, that's all.
The issues I've seen with DHCP relays and VLANs in the forums are usually based on misconfigs in the VLAN or ports blocked. Sometimes we'll refer to call Microsoft Support for specific, hands-on assistance. And searching the threads, from what
I've found that if they did call support, they've never posted back what the problem was based on or the resolution. I can post a couple of them for you to read through, but there were never any response with the actual resolution.
If you like, you also have the option to contact Microsoft Support. Here's a list of phone numbers if you choose this option:
http://support.microsoft.com/contactus/
Ace Fekay
MVP, MCT, MCITP/EA, MCTS Windows 2008/R2 & Exchange 2007, Exchange 2010 EA, MCSE & MCSA 2003/2000, MCSA Messaging 2003
Microsoft Certified Trainer
Microsoft MVP - Directory Services
Technical Blogs & Videos: http://www.delawarecountycomputerconsulting.com/
This post is provided AS-IS with no warranties or guarantees and confers no rights.

CSVPN 3002 and DHCP Relay

I saw an answer to this questions back in Jul 2004 that stated DHCP Relay was not supported on the Cisco 3002 HW client. Is this still the case? If not, what version of software must be used to enable this feature?
Thanks.

DHCP relay is not supported yet, you could check with the release notes.

PXE with IP Helpers/DHCP Relay

Similar Messages

Maybe you are looking for