3Com Switch 7754, Руководство по настройке

Страница: 487 / 940

Configuring MSDP Basic Functions 487
■
Redundancy backup between RPs: When an RP fails, the multicast source
previously registered with it or the receivers previous joined it will register with
or join another nearest RP, thus achieving redundancy backup between RPs.
c
CAUTION:
■
Be sure to configure a 32-bit subnet mask (255.255.255.255) for the Anycast
RP address, namely configure the Anycast RP address into a host address.
■
An MSDP peer address must be different from the Anycast RP address.
Protocols and Standards MSDP is documented in the following specifications:
■
RFC 3618: Multicast Source Discovery Protocol (MSDP)
■
RFC 3446: Anycast Rendezvous Point (RP) mechanism using Protocol
Independent Multicast (PIM) and Multicast Source Discovery Protocol (MSDP)
Configuring MSDP
Basic Functions
To enable exchange of information from the multicast source S between two
PIM-SM domains, you need to establish MSDP peering relationships between RPs
in these PIM-SM domains. In this way, the information from the multicast source
can be sent through SA messages between the MSDP peers, and the receivers in
other PIM-SM domains can finally receive the multicast source information.
A BGP or MBGP route is required between two routers that are MSDP peers to
each other. Through this route, the two routers can transfer SA messages between
PIM-SM domains, so BGP peers are the basic for establishing MSDP peers.
For an area containing only one MSDP peer, known as a stub area, the BGP or
MBGP route is not compulsory. SA messages are transferred in a stub area through
the static RPF peers. In addition, the use of static RPF peers can avoid RPF check on
the received SA messages, thus saving resources.
Before configuring static RPF peers, you must create an MSDP peering connection.
If you configure only one MSDP peer on a router, the MSDP peer will act as a static
RPF peer. If you configure multiple static RPF peers, you need to handle them by
using different rules according to whether the
rp-policy keyword is used to
configure the filtering policies.
When configuring multiple static RPF peers for the same router, you must follow
the following two configuration methods:
■
In the case that all the peers use the rp-policy keyword: Multiple static RPF
peers function at the same time. RPs in SA messages are filtered based on the
configured prefix list, and only the SA messages whose RP addresses pass the
filtering are received. If multiple static RPF peers using the same
rp-policy
keyword are configured, when any of the peers receives an SA message, it will
forward the SA message to other peers.
■
None of the peers use the rp-policy keyword: Based on the configured
sequence, only the first static RPF peer whose connection state is UP is active.
All the SA messages from this peer will be received, while the SA messages
from other static RPF peers will be discarded. Once the active static RPF peer
fails (because the configuration is removed or the connection is terminated),
based on the configuration sequence, the subsequent first static RPF peer