时间戳

首页 > 技术文章 > 组播——优选RPF(反向路径转发)路由原则验证

组播——优选RPF(反向路径转发)路由原则验证

xiaohuihui4956 2021-08-12 16:57 原文

验证理论

 

 RPF路由选举规则：

 

 

优选RPF路由原则：

  1. 掩码最长匹配
  2. 路由最优优先级（Pre值）
  3. 组播静态路由>MBGP路由>单播路由

 

实验拓扑

 

 

 

 

 

 

初始配置

配置接口地址以及AR123之间的OSPF

 

初始结果

 

 

 

 

实验步骤

第一步，配置组播

在全局下开启组播，在所有接口下配置pim dm

以AR1为例

multicast routing-enable
#
interface Serial1/0/0
link-protocol ppp
ip address 155.1.13.1 255.255.255.0
pim dm
ospf enable 1 area 0.0.0.0
#
interface GigabitEthernet0/0/0
ip address 155.1.12.1 255.255.255.0
pim dm
ospf enable 1 area 0.0.0.0
#
interface GigabitEthernet0/0/1
ip address 10.1.1.254 255.255.255.0
pim dm

 

第二步：

验证最长掩码匹配

首先设置组播静态路由出接口为S1/0/0

[AR3]ip rpf-route-static 10.1.1.0 24 Serial 1/0/0
Info: Succeeded in modifying route.

[AR3]dis multicast rpf-info 10.1.1.0
VPN-Instance: public net
RPF information about source: 10.1.1.0
RPF interface: Serial1/0/0, RPF neighbor: 155.1.13.1
Referenced route/mask: 10.1.1.0/24
Referenced route type: mstatic
Route selection rule: preference-preferred
Load splitting rule: disable

然后增加设置组播静态路由掩码增加到25，下一跳到AR2

[AR3]ip rpf-route-static 10.1.1.0 25 155.1.23.2
[AR3]dis multicast rpf-info 10.1.1.0
VPN-Instance: public net
RPF information about source: 10.1.1.0
RPF interface: GigabitEthernet0/0/1, RPF neighbor: 155.1.23.2
Referenced route/mask: 10.1.1.0/25
Referenced route type: mstatic
Route selection rule: preference-preferred
Load splitting rule: disable

此时再查看RPF，下一跳就变更到了AR2上，可见掩码越长越优先

 

第三步：

验证路由优先级，优先级越小越优先

首先删除前面的两条静态，然后将10.1.1.0宣告进OSPF，同时在AR2,4之间建立BGP，将10.1.11.0放入BGP中进行传递

AR1:

interface GigabitEthernet0/0/1
ip address 10.1.1.254 255.255.255.0
pim dm
ospf enable 1 area 0.0.0.0

bgp 100
peer 150.1.3.3 as-number 100
peer 150.1.3.3 connect-interface LoopBack0
#
ipv4-family unicast
undo synchronization
undo peer 150.1.3.3 enable
#
ipv4-family multicast
undo synchronization
peer 150.1.3.3 enable

 

AR3:

bgp 100
timer connect-retry 5
peer 150.1.1.1 as-number 100
peer 150.1.1.1 connect-interface LoopBack0
#
ipv4-family unicast
undo synchronization
undo peer 150.1.1.1 enable
#
ipv4-family multicast
undo synchronization
peer 150.1.1.1 enable
#

此时可以看到AR3优选从OSPF学到的，因为OSPF的优先级10比BGP的255更优先

[AR3]dis multicast rpf-info 10.1.1.0
VPN-Instance: public net
RPF information about source: 10.1.1.0
RPF interface: GigabitEthernet0/0/1, RPF neighbor: 155.1.23.2
Referenced route/mask: 10.1.1.0/24
Referenced route type: unicast
Route selection rule: preference-preferred
Load splitting rule: disable

此时注意，通过和BGP比较OSPF的优先级更小，更优先，但是从AR1,2均可通过OSPF学习到10.1.1.0，AR4却选择下一跳为AR2,由此可见，相同协议下，度量值越小越优先

[AR3]dis multicast routing-table
Multicast routing table of VPN-Instance: public net
Total 1 entry

00001. (10.1.1.1, 239.1.1.1)
Uptime: 00:00:14
Upstream Interface: GigabitEthernet0/0/1

 

 

第四步：

验证：组播静态路由>MBGP路由>单播路由

增加组播静态路由，在掩码相同的情况下，将静态路由，OSPF的优先级都更改为255

[AR3-ospf-1]preference 255

[AR3]ip rpf-route-static  10.1.1.0 24 155.1.13.1 preference 255

此时优选静态的路由

[AR3]dis multicast rpf-info 10.1.1.0
VPN-Instance: public net
RPF information about source: 10.1.1.0
RPF interface: Serial1/0/0, RPF neighbor: 155.1.13.1
Referenced route/mask: 10.1.1.0/24
Referenced route type: mstatic
Route selection rule: preference-preferred
Load splitting rule: disable

删除静态路由之后，此时优选MBGP:

[AR3]UNDO ip rpf-route-static 10.1.1.0 24
[AR3]dis multicast rpf-info 10.1.1.0
VPN-Instance: public net
RPF information about source: 10.1.1.0
RPF interface: GigabitEthernet0/0/1, RPF neighbor: 155.1.23.2
Referenced route/mask: 10.1.1.0/24
Referenced route type: mbgp
Route selection rule: preference-preferred
Load splitting rule: disable

可见，组播静态路由>MBGP>IGP

 

第五步

验证通过IGP选举，开销相同时选择下一跳较大IP的邻居

首先，删除MBGP

[AR1-bgp-af-multicast]undo network 10.1.1.0 24

然后设置AR3的接口开销为2

[AR3]dis ip routing-table 10.1.1.0
Route Flags: R - relay, D - download to fib
------------------------------------------------------------------------------
Routing Table : Public
Summary Count : 2
Destination/Mask  Proto Pre Cost Flags NextHop Interface

10.1.1.0/24　　　 OSPF 255 3 D 155.1.23.2 GigabitEthernet0/0/1
　　 　　　　　　OSPF 255 3 D 155.1.13.1 Serial1/0/0

此时负载均衡，但是AR3还是选择了AR2，因此负载均衡时选择下一跳较大IP的邻居

[AR3]dis multicast rpf-info 10.1.1.0
VPN-Instance: public net
RPF information about source: 10.1.1.0
RPF interface: GigabitEthernet0/0/1, RPF neighbor: 155.1.23.2
Referenced route/mask: 10.1.1.0/24
Referenced route type: unicast
Route selection rule: preference-preferred
Load splitting rule: disable
[AR3]

 

第六步，在AR3的0/0/0口上开启IGMP

[AR3-GigabitEthernet0/0/0]igmp en 

 

 

 

总结

RPF检测规则：

1.掩码越长越优先

2.优先级越小越优先（组播静态优先级默认为1）

3.组播静态>MBGP>IGP

4.相同IGP开销越小越优先

5.相同IGP开销相同，则下一跳IP地址越大越优先

推荐阅读