Ricky Beam <jfbeam@bluetopia.net> said:
> On Wed, 3 May 2000, Bernd Markgraf wrote:
> >> Isn't there a problem with this in that it is possible (although not
> >> necessary all that likely) for two nics to have the same MAC address?
> >iirc, mac adresses are unique! thats how we identify nic world wide. it's
> >a mix of manufacturer, type and serial number. so there is no problem
> >unless you try to burn the nic's prom yourself ;-)
> No, they are NOT unique. They _are_ "relatively unique" -- which is to say
> the manufacturer sequentially numbers the MAC when the card is ready to
> leave the plant (i.e. after QC, etc.) Eventually, the counter will roll
> over -- how many NICs has 3Com and Intel made?
IEEE (I think it is them) allocates a range to each manufacturer. At least
one ran out and requested another one. No roll over.
> Additionally, the MAC is "just another number." In almost all cases, the
> MAC address can be changed -- after all, the manufacturer had to be able
> to set it following construction. AND, sparc hardware doesn't have a MAC
> address as part of the NIC -- it's part of the _machine_ of all things.
Makes sense: Different NICs will normally sit on different nets, where they
can't interfere with one another if they have the same MAC. All you need
really is that all MACs on each network segment are different, and a simple
way to ensure this is to make them depend on the machine serial number.
Also simplifies network management a bit if a machine with a fried NIC
doesn't show up with a new MAC later, and shows up everywhere with the same
MAC.
-- Horst von Brand vonbrand@sleipnir.valparaiso.cl Casilla 9G, Viņa del Mar, Chile +56 32 672616- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Sun May 07 2000 - 21:00:13 EST