Bekijk de volledige versie : Peculiar DHCP Behavior in Client Mode

19-04-2007, 21:12
Hello everyone,

I've been a long time follower of Asus hardware and it's derivatives, and typically lurking about these forums has yielded fantastic fixes to any issues, and I'd just like to say thanks to the countless people who have helped in the past and never even knew it.

But, I've reached an oddball problem here. I've been trying to setup a Rio Receiver on my network for the past week, and have been ruthlessly fighting endless problems in putting together a server to host music to it on my Debian box, finally, today, I hit money, but it was using my wireless game bridge. Now, as I need that bridge for my Xbox, I made the logical step of moving the receiver over to my hacked MN-700 with Oleg's latest custom firmware on it. The MN-700 (which behaves identically to a WL-500g sans the USB port, just for the record) is setup as a client on my existing network, and has served countless uses as a "bridge" for wired clients in the past with no issue.

Here is my problem, though. In the battle to setup the receiver, I tolled all DHCP serving functions over to my Debian box (at, and disabled the DHCP functions on my gateway router. So, at boot, the MN-700 attaches to the wireless, requests an IP, and receives it's static address ( from the Debian box. Now, when I run the receiver, it boots, and begins DHCP requesting, to which the Debian box responds, as per the norm.

The killer here is that these responses don't seem to reach the receiver, as it eventually times out with no address given from the DHCP. My syslog on the Debian box confirms that it is seeing the requests from the proper MAC address, and it is responding with the proper assigned IP (

It may also be pertinent to point out that when the router ( did the DHCP serving, the receiver grabbed an address with no issues (however it consistently got the WRONG one, thus my Debian machine was given the job instead.)

This is an absolutely vital step, as without the properly assigned IP from the Debian box, the receiver cannot mount the NFS directory it gets it's boot files from (which are also on the Debian box). This directory is determined by the actual address of the receiver on the network. Apparently Rio did their best to make this an oddball of a system.

Thank you loads for any help! I can see the light with this damned thing, but at this point I'm stumped!