From: Swan, Jay (jswan@sugf.com)
Date: Wed May 09 2007 - 14:05:55 ART
The key thing to understand here is that there are two components to the
issue: 1) source registration, and 2) traffic flow.
The use of shared trees centralizes source registration on the RP(s).
That way, non-RPs only need to know about sources for those groups for
which they are forwarding traffic.
The traffic flow component is a totally different issue. As you write,
the default is to join the source tree for packet forwarding, although
this can be changed with the spt-threshold command.
-----Original Message-----
From: nobody@groupstudy.com [mailto:nobody@groupstudy.com] On Behalf Of
sirus MOGHADASIAN
Sent: Wednesday, May 09, 2007 5:44 AM
To: Cisco certification
Subject: multicast funny question!! shared trees leads to source tree at
the end of story!!!
Hello group,
I have a "why from first implement sth that at the ends leads to first
place
problem?" question
we use share trees to prevent populating routing table with so many
(S,G)
entries, when I study PIM-SM ,I realized that each source and
destination
builds source tree (actually after bypassing RP and send packet to each
other directly) so what's the point to use share tree when eventually
PIM-SM
cisco implementation (with default of zero threshold to make SPT) leads
to
use source tree model?
Thanks
Sirus Moghadasian
This archive was generated by hypermail 2.1.4 : Fri Jun 01 2007 - 06:55:20 ART