Examples: Configuring the Multicast Forwarding Cache

Before you begin:

• Configure the router interfaces.

• Configure an interior gateway protocol. See the Junos OS Routing Protocols Library for Routing Devices.

• Configure a multicast protocol. This feature works with the following multicast protocols:

  • DVMRP

  • PIM-DM

  • PIM-SM

  • PIM-SSM

This example includes the following statements:

• forwarding-cache—Specifies how forwarding entries are aged out and how the number of entries is controlled.

• timeout—Specifies an idle period after which entries are aged out and removed from inet.1. You can specify a timeout in the range from 1 through 720 minutes.

• threshold—Enables you to specify threshold values on the forwarding cache to suppress (suspend) entries from being added when the cache entries reach a certain maximum and begin adding entries to the cache when the number falls to another threshold value. By default, no threshold values are enabled on the routing device.

  The suppress threshold suspends the addition of new multicast forwarding cache entries. If you do not specify a suppress value, multicast forwarding cache entries are created as necessary. If you specify a suppress threshold, you can optionally specify a reuse threshold, which sets the point at which the device resumes adding new multicast forwarding cache entries. During suspension, forwarding cache entries time out. After a certain number of entries time out, the reuse threshold is reached, and new entries are added. The range for both thresholds is from 1 through 200,000. If configured, the reuse value must be less than the suppression value. If you do not specify a reuse value, the number of multicast forwarding cache entries is limited to the suppression value. A new entry is created as soon as the number of multicast forwarding cache entries falls below the suppression value.

• Procedure
• Results

set routing-options multicast forwarding-cache threshold suppress 150000 
set routing-options multicast forwarding-cache threshold reuse 34 
set routing-options multicast forwarding-cache timeout 60

user@host# show routing-options
multicast {
    forwarding-cache {
        threshold {
            suppress 150000;
            reuse 70000;
        }
        timeout 60;
    }
}

To verify the configuration, run the show multicast route extensive command.

user@host> show multicast route extensive
Family: INET
Group: 232.0.0.1
    Source: 11.11.11.11/32
    Upstream interface: fe-0/2/0.200
    Downstream interface list:
        fe-0/2/1.210
    Downstream interface list rejected by CAC:
        fe-0/2/1.220
    Session description: Source specific multicast
    Statistics: 0 kBps, 0 pps, 0 packets
    Next-hop ID: 337
    Upstream protocol: PIM
    Route state: Active
    Forwarding state: Forwarding
    Cache lifetime/timeout: 60 minutes
Wrong incoming interface notifications: 0

Before you begin:

• Configure the router interfaces.

• Configure an interior gateway protocol. See the Junos OS Routing Protocols Library for Routing Devices.

• Configure a multicast protocol. This feature works with the following multicast protocols:

  • DVMRP

  • PIM-DM

  • PIM-SM

  • PIM-SSM

Flow maps are typically used for fast multicast source failover when there are multiple sources for the same group. For example, when one video source is actively sending the traffic, the forwarding states for other video sources are timed out after a few minutes. Later, when a new source starts sending the traffic again, it takes time to install a new forwarding state for the new source if the forwarding state is not already there. This switchover delay is worsened when there are many video streams. Using flow maps with longer timeout values or permanent cache entries helps reduce this switchover delay.

This example includes the following statements:

• bandwidth—Specifies the bandwidth for each flow that is defined by a flow map to ensure that an interface is not oversubscribed for multicast traffic. If adding one more flow would cause overall bandwidth to exceed the allowed bandwidth for the interface, the request is rejected. A rejected request means that traffic might not be delivered out of some or all of the expected outgoing interfaces. You can define the bandwidth associated with multicast flows that match a flow map by specifying a bandwidth in bits per second or by specifying that the bandwidth is measured and adaptively modified.

  When you use the adaptive option, the bandwidth adjusts based on measurements made at 5-second intervals. The flow uses the maximum bandwidth value from the last 12 measured values (1 minute).

  When you configure a bandwidth value with the adaptive option, the bandwidth value acts as the starting bandwidth for the flow. The bandwidth then changes based on subsequent measured bandwidth values. If you do not specify a bandwidth value with the adaptive option, the starting bandwidth defaults to 2 megabits per second (Mbps).

  For example, the bandwidth 2m adaptive statement is equivalent to the bandwidth adaptive statement because they both use the same starting bandwidth (2 Mbps, the default). If the actual flow bandwidth is 4 Mbps, the measured flow bandwidth changes to 4 Mbps after reaching the first measuring point (5 seconds). However, if the actual flow bandwidth rate is 1 Mbps, the measured flow bandwidth remains at 2 Mbps for the first 12 measurement cycles (1 minute) and then changes to the measured 1 Mbps value.

• flow-map—Defines a flow map that controls the forwarding cache timeout of specified source and group addresses, controls the bandwidth for each flow, and specifies redundant sources. If a flow can match multiple flow maps, the first flow map applies.

• forwarding-cache—Enables you to configure the forwarding cache properties of entries defined by a flow map. You can specify a timeout of never to make the forwarding entries permanent, or you can specify a timeout in the range from 1 through 720 minutes. If you set the value to never, you can specify the non-discard-entry-only option to make an exception for entries that are in the pruned state. In other words, the never non-discard-entry-only statement allows entries in the pruned state to time out, while entries in the forwarding state never time out.

• policy—Specifies source and group addresses to which the flow map applies.

• redundant-sources—Specify redundant (backup) sources for flows identified by a flow map. Outbound interfaces that are admitted for one of the forwarding entries are automatically admitted for any other entries identified by the redundant source configuration. in the example that follows, the two forwarding entries, (10.11.11.11) and (10.11.11.12,) match the flow map defined for flowMap1. If an outbound interface is admitted for entry (10.11.11.11), it is also automatically admitted for entry (10.11.11.12) so one source or the other can send traffic at any time.

• Procedure
• Results

set policy-options prefix-list permanentEntries1 232.1.1.0/24 
set policy-options policy-statement policyForFlow1 from source-address-filter 11.11.11.11/32 exact 
set policy-options policy-statement policyForFlow1 from prefix-list-filter permanentEntries1 orlonger 
set policy-options policy-statement policyForFlow1 then accept 
set routing-options multicast flow-map flowMap1 policy policyForFlow1 
set routing-options multicast flow-map flowMap1 bandwidth 2m 
set routing-options multicast flow-map flowMap1 bandwidth adaptive 
set routing-options multicast flow-map flowMap1 redundant-sources 10.11.11.11 
set routing-options multicast flow-map flowMap1 redundant-sources 10.11.11.12 
set routing-options multicast flow-map flowMap1 forwarding-cache timeout never non-discard-entry-only

user@host# show policy-options
prefix-list permanentEntries1 {
    232.1.1.0/24;
}
policy-statement policyForFlow1 {
    from {
        source-address-filter 11.11.11.11/32 exact;
        prefix-list-filter permanentEntries1 orlonger;
    }
    then accept;
}

user@host# show routing-options
multicast {
    flow-map flowMap1 {
        policy policyForFlow1;
        bandwidth 2m adaptive;
        redundant-sources [ 10.11.11.11 10.11.11.12 ];
        forwarding-cache {
            timeout never non-discard-entry-only;
        }
    }
}

To verify the configuration, run the following commands:

• show multicast flow-map

• show multicast route extensive