On Tue, 24 Jul 2012, Sam Wilson wrote:
> > P.S. We also commonly use the subnet number as the VLAN number. The
> > range for this (in Ethernet at least) is 12bits wide. This will fit in
> > both decimal and hex into the 'subnet' word of the IPv6 address. So I
> > don't think it has much bearing on the choice.
>
> We also align VLAN numbers with IPv4 subnet numbers where we can (mainly
> /24s within our /16 but we have schemes to match up parts of the private
> space too). We've (almost) plumped for putting the BCD representation of
> the VLAN number in the IPv6 subnet field - 4 digits, easily human
> readable and plenty of space for expansion or other uses.
I'm averse to using the VLAN number, simply because I have chopped and
changed VLAN allocations over the years to account for topological changes
(a process largely invisible to end users), and often those changes
haven't required IP renumbering (a process very visible to some end users
or their technical support staff), and I wouldn't want to have introduced
that relationship without a particularly good reason.
In (our) IPv4 world, carving out classic /24 chunks of a /16 and matching
the third octet with the VLANid is common, but starts to break down as
soon as you start doing networks of sizes other than /24. Once you've
broken that link for one not-insignificant set of 'exceptions', then it
seems to me to be a bit pointless introducing that same relationship into
a potentially pristine new scheme.
Putting it another way; while I might chose to take the opportunity to do
IPv6 numbering in a more logical structured way according to topologies,
potential for route aggregation, and so on, I'm extremely unlikely to
start changing the extant VLAN ids to match that neat IPv6 addressing in
many cases.
Jethro.
. . . . . . . . . . . . . . . . . . . . . . . . .
Jethro R Binks, Network Manager,
Information Services Directorate, University Of Strathclyde, Glasgow, UK
The University of Strathclyde is a charitable body, registered in
Scotland, number SC015263.
|