Categories
CCNP Enterprise Core (350-401) Cisco Routing

PIM Bootstrap Router: Candidate Rendezvous Points

A PIM bootstrap router that has been configured as a candidate rendezvous point will receive messages from the bootstrap router.

The bootstrap router contains information that will identify the current active bootstrap router.

The message can be used for the candidate rendezvous point to forward group to rendezvous point mappings from its cache to the boot strap router.

The bootstrap router will collect this information from all candidate rendezvous points in the network and forward them to all PIM routers in the network.

The PIM Bootstrap router mechanism does not elect an active rendezvous point for the entire multicast network. Each router will use a identical hashing algorithm to elect a current active rendezvous point for a particular group range. The hashing algorithm will return a priority value where the lowest priority becomes the active rendezvous point for that group range. If the priority is tied, the IP address which is the highest is selected instead.

Categories
CCNP Enterprise Core (350-401) Cisco Routing

PIM Bootstrap Router

The bootstrap mechanism in a PIM router, is a non-proprietary technology providing a fault tolerant, automated rendezvous point discovery and distribution mechanism. The non-proprietary technology is described in RFC 5059

PIM uses the bootstrap router mechanism to discover and announce rendezvous point information for each group prefix to all routers within a PIM domain.

The object that PIM Bootstrap router accomplishes is that same as the Cisco proprietary ‘Auto-RP’. Bootstrap router mechanism is part of the PIM Version 2 specification.

The rendezvous point information announcements contain the following information:

  • Multicast group range
  • Rendezvous point address
  • Rendezvous point priority
  • Hash mask length
  • SM/Bidir Flag

Bootstrap router messages originate on the bootstrap router, and are flooded to each hop by intermediate routers.

When a bootstrap message is received on an intermediate router, it is forwarded out of every PIM-enabled interface on that router that has a PIM neighbour, including the interface where the message was received.

Bootstrap messages are sent to the multicast address of 224.0.0.13 with a time to live of 1.

To avoid a single point of failure with a single bootstrap router, multiple candidate bootstrap routers can be deployed through a PIM domain. All candidate bootstrap routers will participate in a bootstrap router process by sending PIM bootstrap router messages containing their priority out of all PIM enabled interfaces.

Categories
CCNP Enterprise Core (350-401) Cisco Routing

Rendezvous Point Mapping Agents

A Rendezvous Point Mapping Agent joins the multicast group 224.0.1.39 to receive rendezvous point advertisements.

When an announcement has been received, it will store it in a group to rendezvous point mapping cache along with a hold timer. If there are multiple of the same group range advertised, the candidate rendezvous points advertisement with the highest IP address will be selected and stored.

The mapping agents advertise their stored mappings in the cache to another multicast address, 224.0.1.40, the Cisco rendezvous point discovery address.

The messages are advertised every 60 seconds or when a change in the cache is detected. The mapping agents advertisements contain the elected rendezvous points and their group to rendezvous point mappings.

PIM enabled routers listen on the multicast address 224.0.1.40 and will store these rendezvous point mapping messages in their own caches.

Multiple mapping agents can be configured on the same network to provide a backup in case of a network failure. Mapping agents act independently with no mechanism to elect a primary mapping agent between them – they will all advertise their own collected cache information on the multicast address 224.0.1.40

Categories
CCNP Enterprise Core (350-401) Cisco Routing

Candidate Rendezvous Point

A rendezvous point configured as a ‘Candidate Rendezvous Point’ or C-RP, advertises via an RP announcement message that it is willing to become a rendezvous point.

The candidate rendezvous point advertises this willingness every 60 seconds by default to the multicast address 224.0.1.39. This is controlled by the rendezvous point announce interval.

The announcement message contains three parts, the default group range of 224.0.0.0/4, the candidate rendezvous points address, and the hold timer.

The hold timer is three times that of the rendezvous point announcement interval.

If multiple candidate rendezvous points are advertising their ability to become a rendezvous point, then the rendezvous point with the highest IP address is selected as the preferred rendezvous point.

Categories
CCNP Enterprise Core (350-401) Cisco Routing

Automatic Rendezvous Point

‘Auto-RP’ is a Cisco propriety technology that automates distribution of multicast group to rendezvous point distribution in a PIM topology.

Auto-RP has some benefits:

  • Makes it easier to use multi rendezvous points inside of a network to service different group ranges
  • Allows for load splitting across multiple rendezvous points
  • Simplifies rendezvous point placement towards the location of group participants
  • Prevents inconsistent manual static rendezvous point configuration
  • Multi rendezvous points can act as backups for the other rendezvous point
  • The automatic rendezvous point technology has two basis components, candidate rendezvous points and rendezvous point mapping agents.
Categories
CCNP Enterprise Core (350-401) Cisco Routing

Static Rendezvous Point

A rendezvous point can be statically configured on a multicast range by configuring the address of the rendezvous point on every router in the multicast domain.

This can be the simplest method of configuring a rendezvous point and is suitable for when there are a low number of routers in the network or if the network is small.

Like other protocols, static configuration can increase administration overheads. Every router will be required to have the same rendezvous point address, and if changed, means updating every router with a new static configuration.

If there are several rendezvous points for different groups, every router will need separate configuration in order to be fully informed on where to reach different multicast groups.

Static configuration can run into a problem if a rendezvous point fails; there is no automatic failover or splitting of loads.

Categories
CCNP Enterprise Core (350-401) Cisco Routing

Rendezvous Points

In PIM Sparse Mode, it is required that one router or more operate as a rendezvous point, or RP.

The RP is a single common root placed at a selected point of a shared distribution tree.

The RP can be configured statically or learned through a dynamic mechanism.

The PIM router can be configured to act as a rendezvous point statically or dynamically by configuring automatic rendezvous point or as a PIM bootstrap router.

Categories
CCNP Enterprise Core (350-401) Cisco Routing

PIM Assert Mechanism

There can be scenarios in a multicast network where duplicate multicast packets can flow onto the network. There is a mechanism in PIM called the ‘PIM assert mechanism’ that can stop these duplicate flows. This mostly happens in PIM Dense Mode networks due its flooding properties, and rarely happens in PIM Sparse Mode.

A PIM assert message is sent when multiple routers receive a packet into their outgoing interface that is meant to be going out of the same interface for the same (S, G).

When this occurs, a PIM assert message is sent into the LAN. It contains the administrative distance and metric back towards the source, and it determines which router should be forwarding multicast traffic to that network segment.

Preference is given to the router with the lower route metric. If metrics are tied, the router with the higher IP address is used.

The losing router prunes its interface to prevent further duplicate multicast packets from arriving on that interface. The prune lasts for around three minutes before the process will start again.

Categories
CCNP Enterprise Core (350-401) Cisco Revision Topics Routing

Reverse Path Forwarding (RPF)

Reverse Path Forwarding is an algorithm that helps prevent loops and ensures that multicast traffic arrives on the correct interface.

The RPF algorithm has several features:

  • When a router receives a multicast packet on an interface that it uses to send unicast packets towards the source, that interface is a RPF interface.
  • When a packet arrives on the RPF interface, the router forwards the packet out of an outgoing interface list interface.
  • When a packet arrives on an interface that is not the RPF interface, it discards it to try prevent loops from occuring

PIM will utilise source trees between the source and last hop router, and between the source and the rendezvous point. Shared trees are established between the rendezvous point and the last hop routers. RPF checks are performed differently for each tree:

  1. If a PIM router has a (S, G) entry in the multicast routing table, a shortest path tree, the router will perform a RPF check against the IP address for the source of the multicast packet.
  2. If a PIM router has no source-tree state, it is considered a shared tree state. The router will perform a RPF check on the address of the rendezvous point – the rendezvous address is known when members join the group

Spare mode uses the RPF lookup function to determine where joins and prunes need to be sent. (S, G) joins are sent towards the source, (*, G) joins are sent towards the rendezvous point.

Categories
CCNP Enterprise Core (350-401) Cisco Revision Topics Routing

PIM Sparse Mode – Designated Routers

Where there are multiple PIM-SM routers on a subnet, an election is called to determine a designated router (DR). The designated router helps prevent duplicates of multicast traffic from being sent to to the rendezvous point.

In an election between PIM sparse mode routers, the highest priority wins the election. By default the priority of PIM routers is 1. If all routers in a subnet have the same priority, the election tiebreaker is the highest IP address in the subnet.

On the first hop router, the designated router is responsible for encapsulating register messages (in unicast) that are originated by a source, with a destination of the rendezvous point.

On a last hop router, the designated router is responsible for sending PIM join and prune messages towards the rendezvous point to inform it about group membership. The last hop router is also responsible for a shortest tree path switch-over.

If there are no designated routers, all the of the last hop routers on the same subnet would send PIM joins upstream. This may result in duplicate multicast traffic arriving on the LAN.

For the first hop routers, if there are no elected designated routers, multiple register messages to the rendezvous point can occur.

If there are no hello messages sent in 105 seconds, 3.5 times the hello interval, a new designated router will be elected.