Login
- Table of Contents
- Related Documents
-
| Title | Size | Download |
|---|---|---|
| 02-Multicast routing and forwarding configuration | 188.17 KB |
Contents
Configuring multicast routing and forwarding
About multicast routing and forwarding
Usages of static multicast routes
Multicast forwarding across unicast subnets
Restrictions: Hardware compatibility with multicast routing and forwarding
Restrictions and guidelines: Multicast routing and forwarding configuration
Multicast routing and forwarding tasks at a glance
Prerequisites for multicast routing and forwarding
Configuring static multicast routes
Specifying the longest prefix match principle
Configuring multicast load splitting
Configuring a multicast forwarding boundary
Configuring RPF check failure processing
Configuring the device to flood RPF-check-failed multicast data packets in all VLANs
Configuring the device to send RPF-check-failed multicast data packets to the CPU
Setting the maximum number of cached unknown multicast packets
Display and maintenance commands for multicast routing and forwarding
Troubleshooting multicast routing and forwarding
Static multicast route failure
Configuring multicast routing and forwarding
About multicast routing and forwarding
Each multicast routing protocol has its own routing table. Multicast routing information in routing entries generated by the multicast routing protocols and statically configured multicast routing entries are summarized in a set of (S, G) and (*, G) entries. All the (S, G) and (*, G) entries form a general multicast routing table. The optimal multicast routing entries in the general multicast routing table are added to the multicast forwarding table to guide multicast data forwarding.
RPF check mechanism
A multicast routing protocol uses reverse path forwarding (RPF) check to ensure the multicast data delivery along the correct path and to avoid data loops.
RPF check process
A multicast device performs the RPF check on a multicast packet as follows:
1. Chooses an optimal route back to the packet source separately from the unicast, MBGP, and static multicast routing tables.
The term "packet source" means different things in different situations:
¡ For a packet that travels along the SPT, the packet source is the multicast source.
¡ For a packet that travels along the RPT, the packet source is the RP.
¡ For a bootstrap message originated from the BSR, the packet source is the BSR.
For more information about the concepts of SPT, RPT, source-side RPT, RP, and BSR, see "Configuring PIM."
2. Selects one of the three optimal routes as the RPF route as follows:
¡ If the device uses the longest prefix match principle, the route with the highest subnet mask becomes the RPF route. If the routes have the same mask, the route with the highest route preference becomes the RPF route. If the routes have the same route preference, the unicast route becomes the RPF route. If equal cost routes exist, the route with the highest next hop IP address becomes the RPF route.
For more information about the route preference, see Layer 3—IP Routing Configuration Guide.
¡ If the device does not use the longest prefix match principle, the route with the highest route preference becomes the RPF route. If the routes have the same preference, the unicast route becomes the RPF route. If equal cost routes exist, the route with the highest next hop IP address becomes the RPF route.
The RPF route contains the RPF interface and RPF neighbor information.
¡ If the RPF route is a unicast route or MBGP route, the outgoing interface is the RPF interface and the next hop is the RPF neighbor.
¡ If the RPF route is a static multicast route, the RPF interface and RPF neighbor are specified in the route.
3. Determines whether the packet arrived at the RPF interface.
¡ If the packet arrived at the RPF interface, the RPF check succeeds and the packet is forwarded.
¡ If the packet arrived at the non-RPF interface, the RPF check fails and the packet is discarded.
RPF check implementation in multicast
Implementing an RPF check on each received multicast packet brings a big burden to the device. The use of a multicast forwarding table is the solution to this issue. When the device creates a multicast forwarding entry for an (S, G) packet, it sets the RPF interface of the packet as the incoming interface of the (S, G) entry. After the device receives another (S, G) packet, it looks up the multicast forwarding table for a matching (S, G) entry.
· If no match is found, the device first determines the RPF route back to the packet source and the RPF interface. Then, it creates a forwarding entry with the RPF interface as the incoming interface and makes the following judgments:
¡ If the receiving interface is the RPF interface, the RPF check succeeds and the device forwards the packet out of all the outgoing interfaces.
¡ If the receiving interface is not the RPF interface, the RPF check fails and the device discards the packet.
· If a match is found and the matching forwarding entry contains the receiving interface, the device forwards the packet out of all the outgoing interfaces.
· If a match is found but the matching forwarding entry does not contain the receiving interface, the device determines the RPF route back to the packet source. Then, the device performs one of the following actions:
¡ If the RPF interface is the incoming interface, it means that the forwarding entry is correct but the packet traveled along a wrong path. The packet fails the RPF check, and the device discards the packet.
¡ If the RPF interface is not the incoming interface, it means that the forwarding entry has expired. The device replaces the incoming interface with the RPF interface and matches the receiving interface against the RPF interface. If the receiving interface is the RPF interface, the device forwards the packet out of all outgoing interfaces. Otherwise, it discards the packet.
As shown in Figure 1, assume that unicast routes are available on the network, MBGP is not configured, and no static multicast routes have been configured on Device C. Multicast packets travel along the SPT from the multicast source to the receivers. The multicast forwarding table on Device C contains the (S, G) entry, with Port A as the incoming interface.
· If a multicast packet arrives at Device C on Port A, the receiving interface is the incoming interface of the (S, G) entry. Device C forwards the packet out of all outgoing interfaces.
· If a multicast packet arrives at Device C on Port B, the receiving interface is not the incoming interface of the (S, G) entry. Device C searches its unicast routing table and finds that the outgoing interface to the source (the RPF interface) is Port A. In this case, the (S, G) entry is correct, but the packet traveled along a wrong path. The packet fails the RPF check and Device C discards the packet.
Usages of static multicast routes
A static multicast route can change an RPF route or create an RPF route.
Changing an RPF route
Typically, the topology structure of a multicast network is the same as that of a unicast network, and multicast traffic follows the same transmission path as unicast traffic does. You can configure a static multicast route for a multicast source to change the RPF route. As a result, the device creates a transmission path for multicast traffic that is different from the transmission path for unicast traffic.
Figure 2 Changing an RPF route
As shown in Figure 2, when no static multicast route is configured, Device C's RPF neighbor on the path back to the source is Device A. The multicast data from the source travels through Device A to Device C. You can configure a static multicast route on Device C and specify Device B as Device C's RPF neighbor on the path back to the source. The multicast data from the source travels along the path: Device A to Device B and then to Device C.
Creating an RPF route
When a unicast route is blocked, multicast forwarding might be stopped due to lack of an RPF route. You can configure a static multicast route to create an RPF route. In this way, a multicast routing entry is created to guide multicast forwarding.
Figure 3 Creating an RPF route
As shown in Figure 3, the RIP domain and the OSPF domain are unicast isolated from each other. For the receiver hosts in the OSPF domain to receive multicast packets from the multicast source in the RIP domain, you must configure Device C and Device D as follows:
· On Device C, configure a static multicast route for the multicast source and specify Device B as the RPF neighbor.
· On Device D, configure a static multicast route for the multicast source and specify Device C as the RPF neighbor.
Multicast forwarding across unicast subnets
Devices forward the multicast data from a multicast source hop by hop along the forwarding tree, but some devices might not support multicast protocols in a network. When the multicast data is forwarded to a device that does not support IP multicast, the forwarding path is blocked. In this case, you can enable multicast forwarding across two unicast subnets by establishing a tunnel between the devices at the edges of the two unicast subnets.
Figure 4 Multicast data transmission through a tunnel
As shown in Figure 4, a tunnel is established between Device A and Device B. Device A encapsulates the multicast data in unicast IP packets, and forwards them to Device B across the tunnel through unicast devices. Then, Device B strips off the unicast IP header and continues to forward the multicast data to the receiver.
To use this tunnel only for multicast traffic, configure the tunnel as the outgoing interface only for multicast routes.
Restrictions: Hardware compatibility with multicast routing and forwarding
|
Hardware |
Multicast routing and forwarding compatibility |
|
F5010, F5020, F5020-GM, F5030, F5030-6GW, F5040, F5060, F5080, F5000-AI-20, F5000-AI-40, F5000-V30, F5000-C, F5000-S, F5000-M, F5000-A |
Yes |
|
F1000-AI-20, F1000-AI-30, F1000-AI-50, F1000-AI-60, F1000-AI-70, F1000-AI-80, F1000-AI-90 |
Yes |
|
F1003-L, F1005-L, F1010-L |
Yes |
|
F1005, F1010 |
Yes |
|
F1020, F1020-GM, F1030, F1030-GM, F1050, F1060, F1070, F1070-GM, F1070-GM-L, F1080, F1090, F1000-V70 |
Yes |
|
F1000-AK1110, F1000-AK1120, F1000-AK1130, F1000-AK1140 |
Yes |
|
F1000-AK1212, F1000-AK1222, F1000-AK1232, F1000-AK1312, F1000-AK1322, F1000-AK1332 |
Yes |
|
F1000-AK1414, F1000-AK1424, F1000-AK1434, F1000-AK1514, F1000-AK1524, F1000-AK1534, F1000-AK1614 |
Yes |
|
F1000-AK108, F1000-AK109, F1000-AK110, F1000-AK115, F1000-AK120, F1000-AK125, F1000-AK710 |
Yes |
|
F1000-AK130, F1000-AK135, F1000-AK140, F1000-AK145, F1000-AK150, F1000-AK155, F1000-AK160, F1000-AK165, F1000-AK170, F1000-AK175, F1000-AK180, F1000-AK185, F1000-GM-AK370, F1000-GM-AK380, F1000-AK711 |
Yes |
|
LSU3FWCEA0, LSUM1FWCEAB0, LSX1FWCEA1 |
Yes |
|
LSXM1FWDF1, LSUM1FWDEC0, IM-NGFWX-IV, LSQM1FWDSC0, LSWM1FWD0, LSPM6FWD, LSQM2FWDSC0 |
Yes |
|
vFW1000, vFW2000 |
No |
Restrictions and guidelines: Multicast routing and forwarding configuration
The device can route and forward multicast data only through the primary IP addresses of interfaces, rather than their secondary addresses or unnumbered IP addresses. For more information about primary and secondary IP addresses, and IP unnumbered, see Layer 3—IP Services Configuration Guide.
Multicast routing and forwarding tasks at a glance
To configure multicast routing and forwarding, perform the following tasks:
1. Enabling IP multicast routing
2. (Optional.) Configuring static multicast routes
3. (Optional.) Specifying the longest prefix match principle
4. (Optional.) Configuring multicast load splitting
5. (Optional.) Configuring a multicast forwarding boundary
6. (Optional.) Configuring RPF check failure processing
¡ Configuring the device to flood RPF-check-failed multicast data packets in all VLANs
¡ Configuring the device to send RPF-check-failed multicast data packets to the CPU
7. (Optional.) Setting the maximum number of cached unknown multicast packets
Prerequisites for multicast routing and forwarding
Before you configure multicast routing and forwarding, configure a unicast routing protocol so that all devices in the domain can interoperate at the network layer.
Enabling IP multicast routing
About this task
Enable IP multicast routing before you configure any Layer 3 multicast functionality on the public network or VPN instance.
Procedure
1. Enter system view.
system-view
2. Enable IP multicast routing and enter MRIB view.
multicast routing [ vpn-instance vpn-instance-name ]
By default, IP multicast routing is disabled.
Configuring static multicast routes
About this task
To configure a static multicast route for a multicast source, you can specify an RPF interface or an RPF neighbor for the multicast traffic from that source.
Restrictions and guidelines
Static multicast routes take effect only on the multicast devices on which they are configured, and will not be advertised throughout the network or redistributed to other devices.
Procedure
1. Enter system view.
system-view
2. Configure a static multicast route.
ip rpf-route-static [ vpn-instance vpn-instance-name ] source-address { mask-length | mask } { rpf-nbr-address | interface-type interface-number } [ preference preference ]
3. (Optional.) Delete all static multicast routes.
delete ip rpf-route-static [ vpn-instance vpn-instance-name ]
You can use the undo ip rpf-route-static command to delete a specific static multicast route or use the delete ip rpf-route-static command to delete all static multicast routes.
Specifying the longest prefix match principle
About this task
You can enable the device to use the longest prefix match principle for RPF route selection. For more information about RPF route selection, see "RPF check process."
Procedure
1. Enter system view.
system-view
2. Enter MRIB view.
multicast routing [ vpn-instance vpn-instance-name ]
3. Specify the longest prefix match principle.
longest-match
By default, the route preference principle is used.
Configuring multicast load splitting
About this task
You can enable the device to split multiple data flows on a per-source basis or on a per-source-and-group basis. This optimizes the traffic delivery.
Restrictions and guidelines
This feature does not take effect on BIDIR-PIM.
Procedure
1. Enter system view.
system-view
2. Enter MRIB view.
multicast routing [ vpn-instance vpn-instance-name ]
3. Configure multicast load splitting.
load-splitting { source | source-group }
By default, multicast load splitting is disabled.
Configuring a multicast forwarding boundary
About this task
You can configure an interface as a multicast forwarding boundary for a multicast group range. The interface cannot receive or forward multicast packets for the group range.
Restrictions and guidelines
You do not need to enable IP multicast before this configuration.
Procedure
system-view
2. Enter interface view.
interface interface-type interface-number
3. Configure the interface as a multicast forwarding boundary for a multicast group range.
multicast boundary group-address { mask-length | mask }
By default, an interface is not a multicast forwarding boundary.
Configuring RPF check failure processing
Configuring the device to flood RPF-check-failed multicast data packets in all VLANs
About this task
In some networks, multicast receivers might exist in VLANs to which RPF-check-failed multicast data packets belong. For the receivers to receive the packets, you can configure the device to flood the packets in all VLANs.
Restrictions and guidelines
You do not need to enable IP multicast before this configuration.
Procedure
1. Enter system view.
system-view
2. Configure the device to flood RPF-check-failed multicast data packets in all VLANs.
multicast rpf-fail-pkt flooding
By default, RPF-check-failed multicast data packets are not flooded in all VLANs.
3. Return to user view.
quit
4. Clear all multicast forwarding entries.
reset multicast [ vpn-instance vpn-instance-name ] forwarding-table all
The multicast rpf-fail-pkt flooding command takes effect only after you perform this step.
Configuring the device to send RPF-check-failed multicast data packets to the CPU
About this task
In the following cases, multicast data packets that fail the RPF check must be sent to the CPU:
· If a multicast data packet arrives on an outgoing interface of the corresponding multicast forwarding entry, the packet fails the RPF check. Such packets must be delivered to the CPU to trigger the assert mechanism to prune the unwanted branch. For more information about the assert mechanism, DR, and RPT-to-SPT switchover, see "Configuring PIM."
· Assume that the SPT and RPT have different incoming interfaces on the receiver-side DR in a PIM-SM domain. Before the switchover to SPT finishes, the RPF interface of the route on the DR to the multicast source remains as the RPT incoming interface. The multicast packets that travel along the SPT will fail the RPF check and be discarded. If the RPT is pruned at this moment, the multicast service is instantaneously interrupted.
To avoid this problem, you can configure the device to send the packets that travel along the SPT and fail the RPF check to the CPU. When the packets arrive at the CPU, the system determines whether the packets are expected. If they are expected, the device initiates an RPT prune.
Restrictions and guidelines
You do not need to enable IP multicast before this configuration.
Procedure
1. Enter system view.
system-view
2. Configure the device to send RPF-check-failed multicast data packets to the CPU.
multicast rpf-fail-pkt trap-to-cpu
By default, RPF-check-failed multicast data packets are not sent to the CPU.
3. Return to user view.
quit
4. Clear all multicast forwarding entries.
reset multicast [ vpn-instance vpn-instance-name ] forwarding-table all
The multicast rpf-fail-pkt trap-to-cpu command takes effect only after you perform this step.
Setting the maximum number of cached unknown multicast packets
About this task
The device caches a multicast packet for a period of time if no matching multicast forwarding entry is found for the packet. If a multicast forwarding entry is established for the packet within the time period, the device forwards the packet. This mechanism prevents the device from mistakenly dropping multicast packets when the multicast forwarding entries for these packets are to be created.
You can set the maximum number of unknown multicast packets that can be cached for an (S, G) entry, in total, or both.
Restrictions and guidelines
As a best practice, set the value in the multicast forwarding-table cache-unknown total command to be far greater than the value in the multicast forwarding-table cache-unknown per-entry command.
Procedure
1. Enter system view.
system-view
2. Set the maximum number of unknown multicast packets that can be cached for an (S, G) entry.
multicast forwarding-table cache-unknown per-entry per-entry-limit
By default, the device can cache only one unknown multicast packet for an (S, G) entry.
3. Set the maximum number of unknown multicast packets that can be cached in total.
multicast forwarding-table cache-unknown total total-limit
By default, the device can cache 1024 unknown multicast packets in total.
Display and maintenance commands for multicast routing and forwarding
|
|
CAUTION: The reset commands might cause multicast data transmission failures. |
Execute display commands in any view and reset commands in user view.
|
Task |
Command |
|
Display information about the interfaces maintained by the MRIB. |
display mrib [ vpn-instance vpn-instance-name ] interface [ interface-type interface-number ] |
|
Display multicast boundary information. |
display multicast [ vpn-instance vpn-instance-name ] boundary [ group-address [ mask-length | mask ] ] [ interface interface-type interface-number ] |
|
Display multicast fast forwarding entries. |
display multicast [ vpn-instance vpn-instance-name ] fast-forwarding cache [ source-address | group-address ] * [ slot slot-number ] |
|
Display DF information. |
display multicast [ vpn-instance vpn-instance-name ] forwarding df-info [ rp-address ] [ verbose ] [ slot slot-number ] |
|
Display statistics for multicast forwarding events. |
display multicast [ vpn-instance vpn-instance-name ] forwarding event [ slot slot-number] |
|
Display multicast forwarding entries. |
display multicast [ vpn-instance vpn-instance-name ] forwarding-table [ source-address [ mask { mask-length | mask } ] | group-address [ mask { mask-length | mask } ] | incoming-interface interface-type interface-number | outgoing-interface { exclude | include | match } interface-type interface-number | slot slot-number | statistics ] * |
|
Display information about the DF list in the multicast forwarding table. |
display multicast [ vpn-instance vpn-instance-name ] forwarding-table df-list [ group-address ] [ verbose ] [ slot slot-number ] |
|
Display multicast routing entries. |
display multicast [ vpn-instance vpn-instance-name ] routing-table [ source-address [ mask { mask-length | mask } ] | group-address [ mask { mask-length | mask } ] | incoming-interface interface-type interface-number | outgoing-interface { exclude | include | match } interface-type interface-number ] * |
|
Display static multicast routing entries. |
display multicast [ vpn-instance vpn-instance-name ] routing-table static [ source-address { mask-length | mask } ] |
|
Display RPF information for a multicast source. |
display multicast [ vpn-instance vpn-instance-name ] rpf-info source-address [ group-address ] |
|
Clear multicast fast forwarding entries. |
reset multicast [ vpn-instance vpn-instance-name ] fast-forwarding cache { { source-address | group-address } * | all } [ slot slot-number ] |
|
Clear statistics for multicast forwarding events. |
reset multicast [ vpn-instance vpn-instance-name ] forwarding event |
|
Clear multicast forwarding entries. |
reset multicast [ vpn-instance vpn-instance-name ] forwarding-table { { source-address [ mask { mask-length | mask } ] | group-address [ mask { mask-length | mask } ] | incoming-interface { interface-type interface-number } } * | all } |
|
Clear multicast routing entries. |
reset multicast [ vpn-instance vpn-instance-name ] routing-table { { source-address [ mask { mask-length | mask } ] | group-address [ mask { mask-length | mask } ] | incoming-interface interface-type interface-number } * | all } |
|
|
NOTE: · When you clear a multicast routing entry, the associated multicast forwarding entry is also cleared. · When you clear a multicast forwarding entry, the associated multicast routing entry is also cleared. |
Troubleshooting multicast routing and forwarding
Static multicast route failure
Symptom
No dynamic routing protocol is enabled on the routers, and the physical status and link layer status of interfaces are both up, but the static multicast route fails.
Solution
To resolve the problem:
1. Use the display multicast routing-table static command to display information about static multicast routes. Verify that the static multicast route has been correctly configured and that the route entry exists in the static multicast routing table.
2. Check the type of interface that connects the static multicast route to the RPF neighbor. If the interface is not a point-to-point interface, be sure to specify the address for the RPF neighbor.
3. If the problem persists, contact H3C Support.
