bin/111493: routed doesn't use multicasts for RIPv2 via P2P interfaces

[Bruce M Simpson]

The following reply was made to PR bin/111493; it has been noted by GNATS.

From: Bruce M Simpson <bms at incunabulum.net>
To: James Carlson <carlsonj at workingcode.com>
Cc: Vernon Schryver <vjs at calcite.rhyolite.com>,  dan at obluda.cz, 
 freebsd-gnats-submit at FreeBSD.org
Subject: Re: bin/111493: routed doesn't use multicasts for RIPv2 via P2P interfaces
Date: Fri, 10 Aug 2007 13:54:18 +0100

 James Carlson wrote:
 > Vernon Schryver writes:
 >   
 >> What do you think of the enclose patch?  It differs from the original
 >> in two ways that should reduce behavior changes seen by people with
 >> IFF_POINTOPOINT interfaces without IFF_MULTICAST.  If it makes sense,
 >> I'll generate a bundle on http://www.rhyolite.com/src/
 >>     
 >
 > It seems reasonable to me for RIP.  Looking at the code, it seems that
 > rdisc already handles this case -- despite the comment to the contrary
 > on send_adv's and send_rdisc's 'dst' argument, the value can be a
 > multicast destination (not just 0 or unicast), and often is.
 >
 >   
 
 +1 here.
 
 It is OK to allow the use of multicast addressing on links which do not 
 natively multicast. Indeed it is necessary for IPv4 and IPv6 multicast 
 forwarding to work across generic point-to-point links, as no special 
 encapsulation is performed.
 
 RIPv2's multicast address falls in the 224.0.0.0/8 link local scope, so 
 it should never be seen outside of the link anyway.
 
 I would just suggest that it should default to off to avoid confusing 
 folk. If Zebra/Quagga is disallowing multicasted RIP across 
 non-multicast links, that's another story entirely.
 
 Regards,
 BMS