From: John Conzone (jkconzone@xxxxxxxx)
Date: Sun May 28 2000 - 19:25:19 GMT-3
Kevin, thanks. I was applying basic layer 2 and layer 3 reasoning, and
it wasn't making sense. So actually, it should work like ip on the same
segment (cable range)? Routing shouldn't have to take place? When I turned
of split horizon and it worked, it really threw me.
I'm going to read up on the bug and take a break!
----- Original Message -----
From: "Kevin M. Woods" <kev@nil.org>
To: "John Conzone" <jkconzone@home.com>
Cc: "ccielab" <ccielab@groupstudy.com>
Sent: Sunday, May 28, 2000 5:53 PM
Subject: Re: Appletalk ove frame relay
> John,
>
> The reason it works now is because of a bug in IOS (CSCdk20406). This
will not
> be fixed per the bug ID.
>
> I was pointing you towards `appletalk local-routing' myself, but now
realize it
> too is a bug (fix). In other words, always disable split-horizon to get
around
> the bug above and only use local-routing if you still can't ping a spoke
(older
> IOS images I'm guessing).
>
> Someone who really knows AT can shed more light on this topic...
>
> Kevin
>
> // I just disabled RTMP split horizon on the hub sub-int, and the ping
> // works. But I don't undertand why.
> // I know RTMP is a distance vector protocol, but why would routing
even
> // occur in this case? I'm thinking that they are on the same cable range,
so
> // it is a directly connected device. I'm reading Caslows AT chapter as we
> // speak (errrr, write), but I'm missing how RTMP would affect a ping on
the
> // same newtork (cable range). Unless of course AT doesn't look at it that
way.
This archive was generated by hypermail 2.1.4 : Thu Jun 13 2002 - 08:23:32 GMT-3