配置公网IP FRR(快速重路由)

简介:

 

与我们所知的VRRP一样,IP FRR也是一种用于路由备份的技术。IP FRR适用与公网IP网络中对于丢包、延时非常敏感的业务。

 

在传统的IP网络上,转发链路出现底层故障后,最为直观的表现是在设备上的物理接口状态变为Down状态。设备检测到这种故障后,会通知上层路由系统进行相应更新,并重新计算路由。通常从链路故障发生到路由系统完成路由收敛,要经过几秒钟的时间。

 

但对于网络上的某些对延时、丢包等非常敏感的业务来说,秒级的收敛时间是不能忍受的,可能导致当前业务的中断。如VoIP业务所能容忍网络中断时间为毫秒级。IP FRR特性能都保证转发系统快速应对出现的链路故障,直接启用备份路由进行数据转发,尽快让业务流恢复正常。

 

局限性:

 

IPFRR备用方案的建立,不会对原有的流量产生影响。但是在切换到备份路由后,流量的状态和去向便不得而知。如下,链路的cost如图分布:

wKiom1P8mCSBOzj1AAFk0K_T2lQ664.jpg

R1的流量要到R4去。正常情况下是走R1―R2―R4这条路(红色箭头所示),但是当R1R2之间的链路发生故障或是R2出现故障时,在IP FRR的作用下回直接切换到备用链路R1―R3这条链路。

我们假设R2发生故障,此时流量行走路线会立即切换到备用线路,经过R3,接下来根据cost值,很明显,流量会转发到R2R4,而不会直接到R4。但是,此时的R2出现了故障,所以流量是不能到达R4的。

对于以上不足的解决办法本文档暂不说明。请查找其他资料。

 

 

配置实例:

 

一、实验目的

 

  • 了解IP FRR的原理;

  • 掌握IP FRR的应用;

  • 理解试试IP FRR的意义;

  • 比较与VRRP的区别。

 

二、实验拓扑

wKiom1P8mEPhVmZfAAIMqZR5NHA936.jpg

三、实验要求

 

如上图:网络中有4台路由器,全都运行OSPF协议。R1R4有两条路由可达,且LinkBLinkA的备份链路。当LinkA出现故障时,流量可以快速切换到备份链路上,从而保证流量从R1R4的正常转发。

 

四、实验步骤

 

1.基本配置

 

R1配置:

<Huawei>system-view

Enter system view, return userview with Ctrl+Z.

[Huawei]sys R1

[R1]inter g0/0/0

[R1-GigabitEthernet0/0/0]ipadd 12.12.12.1 24

Aug 26 2014 18:04:31-05:13R1 %%01IFNET/4/LINK_STATE(l)[0]:The line protocol IP on the interfaceGigabitEthernet0/0/0 has entered the UP state.

[R1-GigabitEthernet0/0/0]interg0/0/1

[R1-GigabitEthernet0/0/1]ipadd 13.13.13.1 24

Aug 26 2014 18:04:43-05:13R1 %%01IFNET/4/LINK_STATE(l)[1]:The line protocol IP on the interfaceGigabitEthernet0/0/1 has entered the UP state.

[R1-GigabitEthernet0/0/1]interloo0

[R1-LoopBack0]ip add1.1.1.1 24

[R1-LoopBack0]quit

 

R2配置:

 

<Huawei>system-view

Enter system view, returnuser view with Ctrl+Z.

[Huawei]sys R2

[R2]inter g0/0/0

[R2-GigabitEthernet0/0/0]ipadd 12.12.12.2 24

Aug 26 2014 18:07:41-05:13R2 %%01IFNET/4/LINK_STATE(l)[0]:The lineprotocol IP on the interfaceGigabitEthernet0/0/0 has entered the UP state.

[R2-GigabitEthernet0/0/0]interg0/0/1

[R2-GigabitEthernet0/0/1]ipadd 24.24.24.1 24

Aug 26 2014 18:07:55-05:13R2 %%01IFNET/4/LINK_STATE(l)[1]:The line protocol IP on the interfaceGigabitEthernet0/0/1 has entered the UP state.

 

R3配置:

 

<Huawei>system-view

Enter system view, returnuser view with Ctrl+Z.

[Huawei]sys R3

[R3]inter g0/0/0

[R3-GigabitEthernet0/0/0]ipadd 13.13.13.2 24

Aug 26 2014 18:09:48-05:13R3 %%01IFNET/4/LINK_STATE(l)[0]:The line protocol IP on the interfaceGigabitEthernet0/0/0 has entered the UP state.

[R3-GigabitEthernet0/0/0]interg0/0/1

[R3-GigabitEthernet0/0/1]ipadd 34.34.34.1 24

Aug 26 2014 18:10:03-05:13R3 %%01IFNET/4/LINK_STATE(l)[1]:The line protocol IP on the interfaceGigabitEthernet0/0/1 has entered the UP state.

 

R4配置:

 

<Huawei>system-view

Enter system view, returnuser view with Ctrl+Z.

[Huawei]sys R4

[R4]inter g0/0/0

[R4-GigabitEthernet0/0/0]ipadd 24.24.24.2 24

Aug 26 2014 18:11:23-05:13R4 %%01IFNET/4/LINK_STATE(l)[0]:The line protocol IP on the interfaceGigabitEthernet0/0/0 has entered the UP state.

[R4-GigabitEthernet0/0/0]interg0/0/1

[R4-GigabitEthernet0/0/1]ipadd 34.34.34.3 24

Aug 26 2014 18:11:34-05:13R4 %%01IFNET/4/LINK_STATE(l)[1]:The line protocol IP on the interfaceGigabitEthernet0/0/1 has entered the UP state.

[R4-GigabitEthernet0/0/1]interloo 0

[R4-LoopBack0]ip add4.4.4.4 24

[R4-LoopBack0]quit

 

2.路由配置

 

配置公网IP FRR之前,前置任务之一就是配置静态路由或是IGP协议,保证各节点路由可达。本实验我们就用OSPF协议。

 

R1配置:

 

[R1]ospf

[R1-ospf-1]area 0

 [R1-ospf-1-area-0.0.0.0]network 12.12.12.00.0.0.255

 [R1-ospf-1-area-0.0.0.0]network 1.1.1.00.0.0.255

 [R1-ospf-1-area-0.0.0.0]network 13.13.13.00.0.0.255

 

R2配置:

 

[R2]ospf

[R2-ospf-1]area 0

[R2-ospf-1-area-0.0.0.0]network12.12.12.0 0.0.0.255

[R2-ospf-1-area-0.0.0.0]network24.24.24.0 0.0.0.255

 

 

R3配置:

 

[R3]ospf

[R3-ospf-1]area 0

[R3-ospf-1-area-0.0.0.0]network13.13.13.0 0.0.0.255

[R3-ospf-1-area-0.0.0.0]network34.34.34.0 0.0.0.255

 

R4配置:

 

[R4]ospf

[R4-ospf-1]area 0

[R4-ospf-1-area-0.0.0.0]network4.4.4.0 0.0.0.255

[R4-ospf-1-area-0.0.0.0]network24.24.24.0 0.0.0.255

[R4-ospf-1-area-0.0.0.0]network34.34.34.0 0.0.0.255

 

验证R1―R4连通性:

 

[R1]ping 4.4.4.4

  PING 4.4.4.4: 56  data bytes, press CTRL_C to break

    Reply from 4.4.4.4: bytes=56 Sequence=1ttl=254 time=100 ms

    Reply from 4.4.4.4: bytes=56 Sequence=2ttl=254 time=20 ms

    Reply from 4.4.4.4: bytes=56 Sequence=3ttl=254 time=10 ms

    Reply from 4.4.4.4: bytes=56 Sequence=4ttl=254 time=20 ms

    Reply from 4.4.4.4: bytes=56 Sequence=5ttl=254 time=20 ms

 

  --- 4.4.4.4 ping statistics ---

    5 packet(s) transmitted

    5 packet(s) received

    0.00% packet loss

    round-trip min/avg/max = 10/34/100 ms

 

[R4]ping 1.1.1.1

  PING 1.1.1.1: 56  data bytes, press CTRL_C to break

    Reply from 1.1.1.1: bytes=56 Sequence=1ttl=254 time=80 ms

    Reply from 1.1.1.1: bytes=56 Sequence=2ttl=254 time=20 ms

    Reply from 1.1.1.1: bytes=56 Sequence=3ttl=254 time=10 ms

    Reply from 1.1.1.1: bytes=56 Sequence=4ttl=254 time=20 ms

    Reply from 1.1.1.1: bytes=56 Sequence=5ttl=254 time=20 ms

  --- 1.1.1.1 ping statistics ---

    5 packet(s) transmitted

    5 packet(s) received

    0.00% packet loss

    round-trip min/avg/max = 10/30/80 ms

 

R1---R4连接正常。

 

 

3.路径设置

 

配置公网IP FRR之前的另一个任务是生成两条不等价的路由,一条为另一条的备份路径,本实验我们把LinkB设置为LinkA的备份路径,所以我们要让到达R4的流量在正常情况下经过R2到达R4,通过修改链路的cost值实现。

 

修改R1G0/0/1接口上的Cost值,使OSPF优先选择  LinkA为主链路。

 

[R1]inter g0/0/1

[R1-GigabitEthernet0/0/1]ospfcost 100

[R1-GigabitEthernet0/0/1]quit

 

修改R4G0/0/1接口上的Cost值,使OSPF优先选择  LinkA为主链路。

 

[R4]inter g0/0/1

[R4-GigabitEthernet0/0/1]ospfcost 100

[R4-GigabitEthernet0/0/1]quit

 

验证:

[R1]tracert 4.4.4.4

traceroute to  4.4.4.4(4.4.4.4), max hops: 30 ,packetlength: 40,press CTRL_C to break

1 12.12.12.2 20 ms  20 ms 10 ms

2 24.24.24.2 30 ms  30 ms 20 ms

 

R1R4的流量确实是从R2经过的。

 

4.配置路由策略并启用IP FRR

 

[R1]ip ip-prefix frrpermit 4.4.4.4 32   //用前缀列表抓取到4.4.4.4的流量

 [R1]route-policy ip_frr permit node 10  //启用路由策略

Info: New Sequence of thisList.

[R1-route-policy]if-matchip-prefix frr

[R1-route-policy]applybackup-nexthop 13.13.13.2

[R1-route-policy]applybackup-interface GigabitEthernet 0/0/1

以上三行的意思为:如果匹配前缀列表的流量,就设置下一跳的备份IP为13.13.13.2(即R3的IP)、下一跳的备份接口为R1的G0/0/1(也就是让流量经过R3)

[R1-route-policy]quit

 

启用IP FRR

[R1]ip frr route-policyip_frr

 

 

验证:

 

R1上查看备份出接口和备份下一跳信息

[R1]display iprouting-table 4.4.4.4 verbose

Route Flags: R - relay, D- download to fib

------------------------------------------------------------------------------

Routing Table : Public

Summary Count : 1

 

Destination: 4.4.4.4/32

     Protocol: OSPF             Process ID: 1

   Preference: 10                     Cost: 2

      NextHop: 12.12.12.2        Neighbour: 0.0.0.0

        State: Active Adv              Age: 00h00m11s

          Tag: 0                  Priority: medium

        Label: NULL                QoSInfo: 0x0

   IndirectID: 0x0             

 RelayNextHop: 0.0.0.0           Interface: GigabitEthernet0/0/0

     TunnelID: 0x0                   Flags:  D

    BkNextHop:13.13.13.2      BkInterface:GigabitEthernet0/0/1

      BkLabel: NULL            SecTunnelID: 0x0             

 BkPETunnelID: 0x0         BkPESecTunnelID: 0x0             

 BkIndirectID: 0x0      

 

 

五、实验总结

 

  • IP FRR适用与公网IP网络中对于丢包、延时非常敏感的业务;

  • IP FRR可以在原有链路Down之后,不用经过路由重新计算选路的时间而损失流量,达到毫秒级的快速切换备用链路,保证业务的正常运行;

  • IP FRR虽然好,但是也存在缺陷。流量经过备份路径后的状态和去向我们不得而知,需要经过严格的组网设计解决这种缺陷;


你可能感兴趣的:(技术,接口,检测,流量)