[c-nsp] Bridging multiple inner VLAN's

[Brian Christopher Raaen]

You mentioned an ASR9k.  I'm pretty sure that you can do the requested
task, based on my testing with bridging on an ASR.  What was it that
didn't work for you with the ASR?

----
Brian Raaen
Network Architect
Zcorum

On Wed, Jan 18, 2012 at 7:08 PM, Pshem Kowalczyk <pshem.k at gmail.com> wrote:
> Hi,
>
> On 19 January 2012 12:43, Dave Weis <djweis at internetsolver.com> wrote:
>
> {cut}
>
>>>I'm not entirely sure if I understand you correctly, but if you strip
>>>the inner vlan on the ingress how would you know what inner vlan to
>>>assign to all egress packets that should go towards that internal tag
>>>of 54? For that to work the device that does the operation would have
>>>to remember not only the MAC address but also the inner vlan from the
>>>ingress packets.
>>>We solved this by making the BRAS interpret double tags and that
>>>worked well for us.
>>
>> That's a great question, I would suppose it has to keep a table of MAC to inner tag as they are stripped.
>
> We looked at doing that on ASR9k and it turned out that the vlans are
> never stored. Perhaps ME switches offer more in that space.
>
>> Are you using PPPoE on your BRAS? This is for a legacy setup with 1000+ bridged modems at the far end.
>
> Yes, we are. The frames come double-tagged (as in your case) and the
> BRAS keeps track of all the tags (we use Huawei ME60 as the BRAS, but
> I suspect this is quite standard feature), that allows us to simply
> carry the traffic across our network in a VPLS.
>
> kind regards
> Pshem
> _______________________________________________
> cisco-nsp mailing list  cisco-nsp at puck.nether.net
> https://puck.nether.net/mailman/listinfo/cisco-nsp
> archive at http://puck.nether.net/pipermail/cisco-nsp/