[Fwd:] Nvidia NICs with duplicate mac addresses
etow at yorktel.com
Fri Sep 5 10:22:39 CDT 2008
Back in 1999, I experienced an issue when we ghosted a number of Compaq
desktops and the Intel NICs all ended up with the same MAC address.
It turned out that the Intel NIC that was used in the later models was
different from the earlier models, so changing the driver resolved the
E-mail: etow at yorktel.com
From: Kevin Oberman [mailto:oberman at es.net]
Sent: Fri 9/5/2008 11:05 AM
To: Scott Berkman
Cc: 'Robert E. Seastrom'; nanog at nanog.org
Subject: Re: [Fwd:] Nvidia NICs with duplicate mac addresses
> From: "Scott Berkman" <scott.berkman at reignmaker.net>
> Date: Fri, 5 Sep 2008 10:53:28 -0400 (EDT)
> This reminds me of a story I was told a while back that there was a batch
> of 3com NIC's that all went out with the same MAC from the factory. I
> never found out if that was a rumor/urban legend or the truth. Anyone
> know firsthand or have an article about that?
> -----Original Message-----
> From: Robert E. Seastrom [mailto:rs at seastrom.com]
> Sent: Friday, September 05, 2008 10:33 AM
> To: nanog at nanog.org
> Subject: [Fwd:] Nvidia NICs with duplicate mac addresses
> 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...
> Just when you thought this couldn't happen any more...
> Copying from a different email list...
> mac address 04:4b:80:80:80:03, was showing up in multiple places
> across the network. I googled the mac address and discovered that
> other people are having the same issue with this mac address. Below
> are some links describing the problem:
> I just wanted everyone to know about this problem in case you run
> across similar slow "connectivity" issues. I believe the network card
> is made by NVIDIA.
I can attest that it happened. We had several of them when I was at
LLNL back in the early '80s. It was amazingly hard to convince the 3Com
folks that the cards were not usable. (This is not the only case of
this, either, but was probably the biggest in terms of number of cards
as 3Com made a very large percentage of all cards in those days.
R. Kevin Oberman, Network Engineer
Energy Sciences Network (ESnet)
Ernest O. Lawrence Berkeley National Laboratory (Berkeley Lab)
E-mail: oberman at es.net Phone: +1 510 486-8634
Key fingerprint:059B 2DDF 031C 9BA3 14A4 EADA 927D EBB3 987B 3751
More information about the NANOG