Re: kern/40139: bridge(4) does not transmit DHCP replies to tap(4)

[David Young <dyoung%pobox.com@localhost>]

The following reply was made to PR kern/40139; it has been noted by GNATS.

From: David Young <dyoung%pobox.com@localhost>
To: gnats-bugs%NetBSD.org@localhost
Cc: kern-bug-people%netbsd.org@localhost, gnats-admin%netbsd.org@localhost,
        netbsd-bugs%netbsd.org@localhost, Taylor R Campbell 
<campbell%mumble.net@localhost>
Subject: Re: kern/40139: bridge(4) does not transmit DHCP replies to tap(4)
Date: Sat, 21 Mar 2009 15:50:54 -0500

 On Sat, Mar 21, 2009 at 07:10:06PM +0000, Quentin Garnier wrote:
 > The following reply was made to PR kern/40139; it has been noted by GNATS.
 > 
 > From: Quentin Garnier <cube%cubidou.net@localhost>
 > To: gnats-bugs%NetBSD.org@localhost
 > Cc: 
 > Subject: Re: kern/40139: bridge(4) does not transmit DHCP replies to tap(4)
 > Date: Sat, 21 Mar 2009 20:09:51 +0100
 > 
 >  On Tue, Dec 09, 2008 at 09:00:01AM +0000, Taylor R Campbell wrote:
 >  > >Number:         40139
 >  > >Category:       kern
 >  > >Synopsis:       bridge(4) does not transmit DHCP replies to tap(4)
 >  
 >  It's not an issue with tap(4) (the same happens if you replace it with
 >  any other Ethernet interface).  If you run ktrace on dhclient, you'll
 >  see that it receives the packet on its udp socket for port 68.  I don't
 >  know why it rejects it though.
 
 dhclient does not log any complaints about those packets?
 
 Some NICs pass the FCS to the host in promiscuous mode, but not
 otherwise.  Their drivers neither apply M_HASFCS nor trim the ethernet
 checksum from the input mbufs when the NIC is in promiscuous. bridge(4)
 may conceal or compound the defect in those input mbufs.
 
 Dave
 
 -- 
 David Young             OJC Technologies
 dyoung%ojctech.com@localhost      Urbana, IL * (217) 278-3933