Subject: Re: kern/32900
To: None <kern-bug-people@netbsd.org, gnats-admin@netbsd.org,>
From: Christos Zoulas <christos@zoulas.com>
List: netbsd-bugs
Date: 02/22/2006 15:30:02
The following reply was made to PR kern/32900; it has been noted by GNATS.

From: christos@zoulas.com (Christos Zoulas)
To: gnats-bugs@netbsd.org, kern-bug-people@netbsd.org,
	gnats-admin@netbsd.org, netbsd-bugs@netbsd.org, ndehne@gmail.com
Cc: 
Subject: Re: kern/32900
Date: Wed, 22 Feb 2006 10:28:34 -0500

 On Feb 22,  2:50pm, ndehne@gmail.com (Nino Dehne) wrote:
 -- Subject: Re: kern/32900
 
 |  Can this apply even when the switch is 802.1Q-compliant? I've been using
 |  that switch and VLANs for some time now and never noticed anything like
 |  this. The switch certainly wasn't cheap in the sense of inexpensive. :)
 |  
 |  I also tested the whole setup with an even more advanced switch (D-Link
 |  DES-3225G) which provides sophisticated traffic and error statistics per
 |  port and couldn't see any errors on the participating ports. The problem
 |  still persisted.
 |  
 |  Also, why doesn't this apply to packets that merely pass the router and
 |  are not originating from it?
 |  
 |  The vlan(4) man page says
 |  
 |    vlan can be used with devices not supporting the IEEE 802.1Q MTU, but
 |    then the MTU of the vlan interface will be 4 bytes too small and will not
 |    interoperate properly with other IEEE 802.1Q devices, unless the MTU of
 |    the other hosts on the VLAN are also lowered to match.
 |  
 |  However, sip(4) is listed as being supported.
 |  
 |  I'm at a loss now and still think this is a bug somewhere.
 
 If you are certain that the problem is not in the switch then it can
 certainly be a bug in the sip driver. Do you have another card to test
 with?
 
 christos