Contributor III

Question

DHCP relay is not working

Forum|Forum|18 years ago
August 22, 2007
16 replies
18525 views

We have two FGt 3000 in a cluster. We have a DHCP server connected on our management VLAN which serves a lot of the customers connected on individual VLANs. The DHCP requests are relayed to the DHCP server. Shortly after update to MR5 the DHCP relay function stopped working. I can see the DHCP request on the interface where the requesting device is connected, but nothing on the interface where the DHCP server is connected. Is ther anyone who ca give me a hint on what may be wrong? Thanks Magne

Yngve0

New Member

Hei, I had the same issue here om our FG50A after upgrading to MR5, so I had to downgrade all 17 locations to MR3 again. I made a support issue on that in may/june (build 547) and Fortinet claimed few days ago that this was solved in 559. This time I tried more carefully with one location (the closest one) and the problem is still there. Support is informed, but it would be nice if you posted it as well. Yngve

A

Anonymous_UserAuthor

Contributor III

Thanks We have 559 on our boxes. Support is informed. Magne

rwpatterson

New Member

Build 564 is out. Give that a shot.

Yngve0

New Member

Seems like not :( Yngve

Carl_Wallmark

New Member

Im using build 564 and with DHCP Relay enabled on 120 VLANS - and it works perfectly !

Yngve0

New Member

I dont know what Magnnyb' s scenario is, but mine issue was DHCP-relay through VPN in IPSEC interface mode. It was working perfect in MR3, but not with the same configuration in MR5. I got some answer from FG-support that I had to define a new rule: Local-interface (All) -> VPN-interface (All) for DHCP-server. But on a earlier issue, I was told not to use " ALL" as host on VPN-rules, since that will mess up the box. Based on that experience, I asked for some clarification. Yngve

A

Anonymous_UserAuthor

Contributor III

I am using regular DHCP relaying (no IPsec)

Carl_Wallmark

New Member

i was having problems when i configured DHCP relay, but then i played around with STP and RSTP on the switches, then it worked perfect. in my solution i have 2 FG 200A in a cluster A-P with 2 VDOMS (1 NAT, 1 TP) and with 120 VLANS, and every VLAN have its own IP network, and every VLAN is using the same DHCP server. So my problem was the Spanning Tree protocol in the switches (HP Procurve).

A

Anonymous_UserAuthor

Contributor III

That solved my problem. Thank you very much.

Carl_Wallmark

New Member

Ohhh, i have helped my first " customer" , im not a virgin anymore =) Wheres the beer ???????

rwpatterson

New Member

I thought it was a cigarette??

Carl_Wallmark

New Member

thats just in the movies =)

Yngve0

New Member

Unfortunately; My issue is not solved and Fortinet have now idee neither. My issue is -DHCP -server (windows 2k3) is located on HQ, connected to FG200A - All sub-office running DHCP-relay against central DHCP-server. Suboffice have FG50A, FG60 or FG50B with VPN-connection in interface-mode to HQ. Mainly there are a VLAN-switch(D-link 3010) connected to internal interface and multiple VLAN on each site. This was working problemless until upgrading to MR5. Y

A

Anonymous_UserAuthor

Contributor III

Have similar configuration here: - DHCP server (linux based) in main location with a cluster of 2 300A - secondary office with DHCP relay through interface mode VPN with a cluster of 2 100A We had some trouble after the upgrade to MR5 patch 1 (build 564) with both normal traffic and DHCP requests disapparing crossing the VPN tunnel. It seems that a bring down - up on the virtual interface and a disable - enable on the DHCP relay solved the problems, it' s running problemless since 1 day ago. Hope this can help Marco

Sign up

Login with SSO

Login to the community

Login with SSO