Hi all,

I have a problem with DHCP relay on mikrotik. Our DHCP server is on linux machine connected over 2 MT routers with DHCP relay.
I'm trying to get address on ubiquiti bullet 2 device connected directly with MT DHCP relay router.

What happens, is that DHCP relay gets the request from client, and log writes that discover is sent to DHCP server but nothing comes there.
I tried to configure DHCP server on relay's neighbour MT router and same situation happened, no request seen on dhcp server.

When DHCP server is on the same router where I'm trying to build relay, client gets it's address.
Linux DHCP server is tested and works with other clients.
Btw, we have ospf routing between MT's

have you configured your server to accept packets from relay? At least that is required for DHCP server in RouterOS.

It accepts packets from relay - tested.
MT DHCP server was also configured to allow requests from relay, but didn't work.

I saw this on a v5.x router the other day. I've used the DHCP relay a number of times previously on version 2.x, 3.x, & 4.x but it seems to be broken with v5.x. I haven't had a chance to test further, but in the limited testing I did get to run, I see the DHCP relay count the request, but it never seems to send the DHCP request on up to the DHCP server.

5.8 will have dhcp fixes. But tat does not affect linux server. So you can see requests from relay but there is no response?

Sent from my GT-I9100 using Tapatalk

I can see the DHCP relay service on RouterOS count up the request coming in and you can see the request coming into the DHCP relay via a packet sniff, but the DHCP relay never seems to send out the request to the upstream DHCP server. In the case the other day, the DHCP server was a Windows 2008 server with a fully working DHCP server and the link was via an IPSEC connection to another Mikrotik router at the site with the Windows 2008 server. The DHCP relay at the remote site never relayed the DHCP request across the IPSEC tunnel.

I see requests in relay status and sending discovers in log but that doesn't mean that it actually sends it, right?
Running tcpdump on DHCP server doesn't catch any packets from relay.

We tested this on RB750G, 4.11 and 5.7 versions.
On RB450, version 4.9 it works.

5.8 will have dhcp fixes.

I didn't think this was worth creating a new thread, but I can't get DHCP to work reliably at all in 5.7. This scenario has two Mikrotiks. One is an access point using Virtual Access Points with VLANS. The request goes to a different Mikrotik that's the DHCP server.

Steps to repeat:

* Configure a DHCP server using the wizard. Works fine.
* Reboot Mikrotik. DHCP no longer works.
* Disable and re-enable your DHCP server. If you have multiple, you must disable/enable ALL of them
* It starts working again

I was able to see the DHCP packet come in to the Mikrotik performing as the DHCP server, but DHCP debug would never show anything until you disable/enable DHCP.

At least one point while testing this, the DHCP servers went red after enabling them. I was only able to fix this by disabling/enabling the local IPv4 address related to the pool.

After a few hours of troubleshooting I finally downgraded to 5.6 and it works perfectly fine.

It's also worth noting that the changelog does not list any DHCP changes between 5.6 and 5.7.

Problem is solved that was on wifi equipment.
We put it from bridge to WDS mode which is true bridge.
I'm talking about Ubiquiti.

I am seeing a similar issue on RouterOS 5.12.

I understand 5.13 just came out, I am currently working both to boil down a testcase for this problem on the bench, and to confirm it's still occurring in 5.13.

Confirmed on RB1200 router, and two different RB1100 routers thus far.

We are converting our large L2 network to L3, and one of those milestones is setting up DHCP relay at our erstwhile routers (starting with the core router which is currently installed) to centralize all of our DHCP business to a single ISC-dhcp-server running on Debian Squeeze.

We had the problem so intermittently, that I had to spend several weeks making a copy of Nagios work and crafting tests to fire packets both at the relay and directly to the server. (DHCP is hard to test when you're not sabotaging the network settings of a workstation by hand. ;D) Shiny new set of calipers crafted, I was able to sniff traffic and confirm that most of the time it worked, some of the time the relay received the DHCPREQUEST packets and did not increment either counter, sometimes it received the request packets, incremented the request counter, and never forwarded them, and sometimes it forwarded the request, incremented the counter, got the reply "DHCP OFFER" packet but never incremented the "responses" counter nor relayed the response.

In order to get customers off the hook, as soon as we were able to measure the problem we just extended client vlans past this router to the DHCP server which functions flawlessly but that bandaid blocks our L3 rollout. I'm emailing Mikrotik support, and will keep this thread up to date as to the resolution. 8I

Thanks, guys!

Did you get a response from mikrotik support on this? I've upgraded a RB1200 to 5.19 and I'm seeing similar behavior as well.

Running a dhcp-relay on address 192.168.4.1 vlan4 pointed at a dhcp server at 192.168.0.16.

My ISC dhcpd is receiving the discover requests and sending out offers, however these are not being forwarded back to the other 192.168.4.0/24 clients.

Pings from 192.168.0.0/24 to 192.168.4.0/24 work and other traffic is routed successfully.

Self inflicted.

Adding a firewall rule to the input chain to allow UDP 67 to router addresses must allow the relay helper to send the dhcp offer back to the clients. Was getting blocked by a drop all filter on the input chain.

Hi

I have this issue, I configured a Dhcp relay, but the DHCP Server alike the several examples on internet is not directly connected to the DHCP Relay. there is at least 3 routers, there is no nat between them and all network have connection between each others.

I will try to detail the escenario

R1 (DHCP Server) is an CCR1036
has many IP networks on his local interface for customers, a loopback address for the ospf, and already has a DHCP Server on the same interface
the IP 192.168.1.1/30 on the same bridge interface connect to

R2 (Tower Router) CCR1036
who has 192.168.1.2/30 as the wan interface, the routing is with ospf and it has a loopback interface and IP, in the bridge interface for customer has different networks and the Ip 192.168.1.5/30 who connects to

R3 (DHCP Relay) RB4011
who has 192.168.1.6/30 as the wan interface, the routing is with ospf and it has a loopback interface and IP, in the bridge interface for customer has different networks and the Ip 192.168.0.1/24 that it is the network I try to provide the DHCP From the dhcp server

I create the pool on DHCP Router R1 as 192.168.0.2-192.168.0.254, and the relay as 192.168.0.1
in the dhcp relay I try to set up the address of the dhcp server as 192.168.1.1, also try the loopback IP, and then the first IP created on the bridge interface but didn't work, I set the local address as the 192.168.0.1, but the dhcp relay shows the sended reques but there is no reply from the dhcp server and the host doesn't get the IP.