[f-nsp] Preserving double-tagged traffic on ICX6100

[Frank Bulk]

Yes, we're using it as an aggregation switch.  

To your questions:
- I'm not sure how that feature is relevant to our situation.  Typically the
Q-in-Q feature is used to describe a mechanism to take in double-tagged
traffic and deliver it non-tagged on access ports.  That's not what we're
trying to do.
- while I think we could find a way to do that with the two other vendor
product that participate in this double-tagged circuit, I'd rather not
deviate from the default to compensate for this bug in ICX6100.  It's always
mystified me why some L2 products seem to choke on double-tagged traffic
when all it should care about is the outer tag.

Thanks,

Frank

-----Original Message-----
From: igor at ergens.org [mailto:igor at ergens.org] On Behalf Of Igor Ybema
Sent: Saturday, May 05, 2012 12:20 PM
To: Frank Bulk
Cc: foundry-nsp at puck.nether.net
Subject: Re: [f-nsp] Preserving double-tagged traffic on ICX6100

> My testing used 0x8100 for both inner and outer and no aggregated-VLAN.
> Nothing special about configuration --
> vlan 1048
>  tagged ethe 1/3/1 ethe 2/3/1
>
> Traffic goes in and out 1/3/1 and out 2/3/1.  VLAN tag 235 is the inner
tag.
> Nothing about VLAN 235 is configured on the ICX6100 -- it doesn't need to
> be.  Yet VLAN tag 235 is stripped out.


Ok I haven't tested this yet. Clearly you are using the ICX as a
provider core switch, not a prodider edge switch which I tested.
Will test this monday if I have some spare time.

Two questions though:
- Why aren't you using the aggregated-vlan feature? To my knowledge
this is needed to support q-in-q. Stripping unknown vlan's from
etherframes could be a 'normal' thing if this feature is disabled.
- Could you try and use another tag-type for the outer tag? Using
0x8100 for both outer and inner should work but it is used less and
less lately. Using 0x9100 or even better the standard for provider
bridging 0x88a8 will keep you and maybe your equipment from getting
headaches.

regards, Igor