Huge thanks to our Platinum Members Endace and LiveAction,
and our Silver Member Veeam, for supporting the Wireshark Foundation and project.

Ethereal-users: Re: [Ethereal-users] Regarding 802.1q

Note: This archive is from the project's previous web site, ethereal.com. This list is no longer active.

From: Scott Lowrey <slowrey@xxxxxxxxxxx>
Date: Wed, 09 Mar 2005 13:49:57 -0500
Good point.  Sorry, I was thinking about host interfaces.   As Marco points out, this shouldn't happen on a simple switch.  It remains as only a suspicion on my part and I have no conclusive evidence that it actually does happen!

Marco van den Bovenkamp wrote:
Scott Lowrey wrote:

Some hubs/switches will drop VLAN-tagged frames because they don't understand the tag (which appears as a bogus type field to a non-VLAN-aware device).

Shouldn't happen for any frame that's not longer than 1518 bytes including VLAN tag. Why should a switch care about what the frame looks like, besides size, source & destination MAC adresses and CRC? The only frames such a switch should drop are those that have more than 1496 payload bytes, as they end up as well-formed frames longer than 1518 bytes ('baby giants').

I have, however, seen small NetGear gigabit switches that will pass tagged packets even though they are not VLAN-aware.

And so they should, with the proviso that they are otherwise OK (not longer than 1518 bytes).


--
Scott Lowrey
Test Engineering Manager
NexTone Communications
Gaithersburg, Maryland USA

1.240.912.1369