Hi, 1) Could it be some problem with internal firewall, AppArmor or SELinux? 2) How you are doing the relay? I was auditing our relay with our Cisco Firewall recently and found some nasty side effects due to some lack of understanding of the relay process when configuring the firewall. 3) This packet trace is not evidently the first requests, but a renewal? Regards, Rui Ribeiro

Message: 4 Date: Tue, 01 Apr 2014 14:32:52 -0400 From: Alan DeKok <aland@deployingradius.com> To: FreeRadius users mailing list <freeradius-users@lists.freeradius.org> Subject: Re: Freeradius DHCP IP pool and not correct port for DHCP reply Message-ID: <533B0654.8060306@deployingradius.com> Content-Type: text/plain; charset=ISO-8859-1

Thomas Bru wrote:

...

On the Bug fixes list , I see the problem was solved using Freeradius 3.0.4 (http://freeradius.org/version3.html, Use correct port when DHCP relaying, ) So I reinstall the Freeradius 3.0.4 on my server but the problem is still present and my server dropped the packets.

17:42:23.875867 IP 10.239.0.2.55175 > 10.225.2.8.67: BOOTP/DHCP, Request from c9:31:cf:d8:af:ec, length 308 17:42:23.932246 IP 10.225.2.8.67 > 10.239.0.2.67: BOOTP/DHCP, Reply, length 300

AS you can see, the request packet from 10.239.0.2 with source port 55175 but will go back to 10.239.0.2 BUT with 67 port.

The correct destination port for DHCP relay packets is 67.

What exactly do you think the problem is?

Alan DeKok.