From: ccie2be (ccie2be@nyc.rr.com)
Date: Wed May 18 2005 - 12:48:01 GMT-3
Hey Lee,
I'll take a shot at this.
First, this requirement isn't specific to rip, I believe it applies to all
ipv6 routing protocols except BGP. Also, I think this applies to all nbma
media except isdn.
Here's how I understand it.
Assume r1 and r2 are connected via f/r and are running an ipv6 based routing
protocol, take your pick.
When an ipv6 IGP sends out updates or LSA's or Hello's, by default, the
source ipv6 of the update is the link-local address. Therefore, for the
receiving rtr to be able to respond, it uses the link-local address. If you
don't have a static mapping you'll get an encap failed message.
HTH, Tim
-----Original Message-----
From: nobody@groupstudy.com [mailto:nobody@groupstudy.com] On Behalf Of Lee
Donald
Sent: Wednesday, May 18, 2005 11:26 AM
To: ccielab@groupstudy.com
Subject: IPV6 Link Local FE80
Can anybody tell me why on frame-relay, using RIPng for instance does the
rip route in the table point towards the link local of the originating
router? Shouldn't this route come from the global address of the originating
router?
I know how to get it working over frame, by having a frame-map to the global
and then a frame map to the link local of the remote but I would like to
know why?
Any takers?
Thanks in advance.
Lee.
This archive was generated by hypermail 2.1.4 : Fri Jun 03 2005 - 10:11:58 GMT-3