[107550] in North American Network Operators' Group
Re: [Fwd:] Nvidia NICs with duplicate mac addresses
daemon@ATHENA.MIT.EDU (Robert E. Seastrom)
Fri Sep 5 13:19:49 2008
To: "David W. Hankins" <David_Hankins@isc.org>
From: "Robert E. Seastrom" <rs@seastrom.com>
Date: Fri, 05 Sep 2008 13:19:44 -0400
In-Reply-To: <20080905163606.GE5333@isc.org> (David W. Hankins's message of
"Fri, 5 Sep 2008 09:36:06 -0700")
Cc: nanog@nanog.org
Errors-To: nanog-bounces@nanog.org
"David W. Hankins" <David_Hankins@isc.org> writes:
> On Fri, Sep 05, 2008 at 10:32:46AM -0400, Robert E. Seastrom wrote:
>> Forwarded to NANOG in the interests of wider awareness... having been
>> there and torn out my already scarce hair, duplicate MAC addresses can
>> really mess up your day...
>
> Out of curiosity, does this happen often enough we might want to
> consider an automated means to negotiate out of the problem state
> (e.g. detect collisions and negotiate MAC address by DHCP)?
>
> It would take years to deploy but might save thousands of hairs.
The same DHCP server (ip helper-address blah) serves my office, my
home, and the colo. Can you give me an idea of a good heuristic for
telling the difference between moving my laptop around and finding MAC
address collisions? Or are you suggesting that you hand out a MAC
address along with an IP address when the client DHCPs and the client
then changes it?
-r
