MPLS
{Back to Index}

1. 术语
2. 标签格式
3. LDP
4. 原理
5. MPLS 中 BGP 路由黑洞问题研究
6. 路由汇总对 MPLS 的影响

1 术语

1.1 FEC ( Forwarding Equivalence Class )

MPLS 实际上是一种 分类转发 的技术，它将具有相同转发处理方式（目的地相同，使用的转发路径相同，具有相同的服务等级等）的分组归为一类，这种类别就称为 转发等价类 。

1.2 CEF

Figure 1: CEF 组成

Figure 2: CEF 原理示意

使用 MPLS 首先要启用 CEF ，它的优点在于：

可被硬件(ASIC)调用，转发效率高
拓扑驱动，具有较快的收敛速度
CEF 将路由和下一跳关联，解决路由递归查询问题
CEF 是唯一一种提供 MPLS 标签插入的转发机制
CEF 形成 FIB ，而标签分发协议(TDP/LDP)根据 FIB 来分发标签

1.3 LSR

标签交换路由器。运行 MPLS 的路由器即为 LSR 。
LSR 主要负责标签的交换。

Figure 3: 在 LSR 上，数据包只需要查 LFIB

1.3.1 Edge LSR ¹

运行于 MPLS Domain 的边界的 LSR 即为 Edge LSR ，负责标签的压入 (Push) 和弹出 (Pop) 操作。

Figure 4: Edge LSR

Figure 5: Edge 上，FIB 和 LFIB 都会用到 (参见 14)

1.4 FIB/LIB/LFIB

FIB 是基于路由表 (RIB) 形成的，其中包含标签信息 ²
LIB 用于存放本地分配的标签信息和 所有学习到 的标签信息，起到保存标签信息的作用
LFIB 即标签转换信息表，由 LIB 和 FIB 推导出来，用于查询某条路由的标签转换规则，建立入标签与出标签的映射关系

Figure 6: 三张表的内在联系

2 标签格式

MPLS Label
00	01	02	03	04	05	06	07	08	09	10	11	12	13	14	15	16	17	18	19	20	21	22	23	24	25	26	27	28	29	30	31
Label																				TC: Traffic Class (QoS and ECN)			S: Bottom-of-Stack	TTL: Time-to-Live

0-15 是保留标签，一般从 16 开始分配
bottom-of-stack 主要用于多层标签的场景:

一层标签
MPLS

两层标签
MPLS VPN

三层标签
MPLS TE (流量工程)

3 LDP

Figure 7: 实验环境

LDP Route ID 建议使用 /32 Loopback 地址. ³

3.1 基本配置

配置清单

----------------- R1 ----------------------
en
conf t
 hostname R1
 no ip do lo
 line con 0
  exec-timeout 0 0
  logging synchronous
 exit
 int lo0
  ip address 1.1.1.1 255.255.255.255
 exit
 int e0/0
  ip address 10.1.12.1 255.255.255.0
  mpls ip ! switch to establish LDP on this ifc
  no sh
 exit
 router ospf 110
  router-id 1.1.1.1
  network 10.1.12.1 0.0.0.0 area 0
  network 1.1.1.1 0.0.0.0 area 0
 exit
 
 ip cef ! must enable this
 mpls ldp router-id lo0
 mpls label range 100 199 ! make it easy to debug
 
end

----------------- R2 ----------------------
en
conf t
 hostname R2
 no ip do lo
 line con 0
  exec-timeout 0 0
  logging synchronous
 exit
 int lo0
  ip address 2.2.2.2 255.255.255.255
 exit
 int e0/0
  ip address 10.1.12.2 255.255.255.0
  mpls ip
  no sh
 exit
 int e0/1
  ip address 10.1.23.2 255.255.255.0
  mpls ip
  no sh
 exit
 router ospf 110
  router-id 2.2.2.2
  network 10.1.12.2 0.0.0.0 area 0
  network 10.1.23.2 0.0.0.0 area 0
  network 2.2.2.2 0.0.0.0 area 0
 exit
 ip cef ! must enable this
 mpls ldp router-id lo0
 mpls label range 200 299
end

----------------- R3 ----------------------
en
conf t
 hostname R3
 no ip do lo
 line con 0
  exec-timeout 0 0
  logging synchronous
 exit
 int lo0
  ip address 3.3.3.3 255.255.255.255
 exit
 int e0/0
  ip address 10.1.23.3 255.255.255.0
  mpls ip
  no sh
 exit
 router ospf 110
  router-id 3.3.3.3
  network 10.1.23.3 0.0.0.0 area 0
  network 3.3.3.3 0.0.0.0 area 0
 exit
 ip cef ! must enable this
 mpls ldp router-id lo0
 mpls label range 300 399
end

3.2 查看命令

3.2.1 查看邻居

R2#sh mpls ldp neighbor
    Peer LDP Ident: 3.3.3.3:0; Local LDP Ident 2.2.2.2:0
        TCP connection: 3.3.3.3.43796 - 2.2.2.2.646
        State: Oper; Msgs sent/rcvd: 49/49; Downstream
        Up time: 00:35:31
        LDP discovery sources:
          Ethernet0/1, Src IP addr: 10.1.23.3
        Addresses bound to peer LDP Ident:
          10.1.23.3       3.3.3.3
    Peer LDP Ident: 1.1.1.1:0; Local LDP Ident 2.2.2.2:0
        TCP connection: 1.1.1.1.646 - 2.2.2.2.38627
        State: Oper; Msgs sent/rcvd: 48/48; Downstream
        Up time: 00:35:29
        LDP discovery sources:
          Ethernet0/0, Src IP addr: 10.1.12.1
        Addresses bound to peer LDP Ident:
          10.1.12.1       1.1.1.1
R2#

3.2.2 查看是否收到 Hello

R2#sh mpls ldp discovery
 Local LDP Identifier:
    2.2.2.2:0
    Discovery Sources:
    Interfaces:
        Ethernet0/0 (ldp): xmit/recv
            LDP Id: 1.1.1.1:0
        Ethernet0/1 (ldp): xmit/recv
            LDP Id: 3.3.3.3:0

假设 transport IP 不可达(比如 R1 没有通告 1.1.1.1/32 进 IGP)，也能收到 Hello :

R2#sh mpls ldp discovery
 Local LDP Identifier:
    2.2.2.2:0
    Discovery Sources:
    Interfaces:
        Ethernet0/0 (ldp): xmit/recv
            LDP Id: 1.1.1.1:0; no route
        Ethernet0/1 (ldp): xmit/recv
            LDP Id: 3.3.3.3:0

3.2.3 查看 FIB 表

show ip cef detail

R1#sh ip cef 3.3.3.3 detail
3.3.3.3/32, epoch 0
  dflt local label info: global/102 [0x0]
  nexthop 10.1.12.2 Ethernet0/0 label 201-(local:102)

3.2.4 查看 LIB 表

R2#sh mpls ldp bindings
  lib entry: 1.1.1.1/32, rev 2
        local binding:  label: 200
        remote binding: lsr: 1.1.1.1:0, label: imp-null
        remote binding: lsr: 3.3.3.3:0, label: 300
  lib entry: 2.2.2.2/32, rev 4
        local binding:  label: imp-null ! imp-null 即为 Label:3 ，表明该路由是本地产生的，收到该标签的路由器就是针对该路由的次末跳
        remote binding: lsr: 1.1.1.1:0, label: 100
        remote binding: lsr: 3.3.3.3:0, label: 301
  lib entry: 3.3.3.3/32, rev 6
        local binding:  label: 201
        remote binding: lsr: 3.3.3.3:0, label: imp-null
        remote binding: lsr: 1.1.1.1:0, label: 102
  lib entry: 10.1.12.0/24, rev 8
        local binding:  label: imp-null
        remote binding: lsr: 1.1.1.1:0, label: imp-null
        remote binding: lsr: 3.3.3.3:0, label: 302
  lib entry: 10.1.23.0/24, rev 10
        local binding:  label: imp-null
        remote binding: lsr: 1.1.1.1:0, label: 101
        remote binding: lsr: 3.3.3.3:0, label: imp-null

3.2.5 查看 LFIB 表

R2#sh mpls forwarding-table
Local      Outgoing   Prefix           Bytes Label   Outgoing   Next Hop
Label      Label      or Tunnel Id     Switched      interface
200        Pop Label  1.1.1.1/32       0             Et0/0      10.1.12.1 ! outgoing 是 Pop 的原因是从邻居收到的 label 是 3 (imp-null)
201        Pop Label  3.3.3.3/32       0             Et0/1      10.1.23.3

3.3 邻居建立过程

Figure 8: 邻居发现

Figure 9: 会话建立

Figure 10: 标签信息交互

pcap file

4 原理

4.1 标签建立过程 ⁴

4.1.1 依赖 IGP 构建 RIB 与 FIB(CEF) ⁵

4.1.2 运行 LDP 以维护 LIB

为路由表中每一条路由前缀 ( BGP 路由除外 ) 分配标签。同时，LSR 也会将本地的标签信息通告给 LDP 邻居。

4.1.3 基于 LIB 和 FIB 维护 LFIB

4.2 数据包传递过程

Figure 14: 数据包传递过程的标签查询与转换

4.3 Penultimate Hop Popping (次末跳优化)

Figure 15: 优化前

由于最后一跳收到的数据是带标签的，因此会先查 FLIB 表，当去掉标签后又需要根据 NH IP 查询 FIB 找到转发接口。
标签在上一条就可以被去掉了。

Figure 16: 优化后

这样最后一跳收到不带标签的数据时，只需要查询 FIB 即可。

4.3.1 imp-null

直连路由 在本地分配到的标签为 imp-null(label:3) ，LDP 邻居收到这个标签后，会将 Outgoing Label 设置为 Pop 。

Figure 17: imp-null

4.3.2 POP 与 UNTAGGED 区别

POP 是表示对端路由器发来 imp-null 的标签，说明对端是有这条路由的，当本地路由器收到数据时，会弹出标签，并以 IP 包的方式传给对端路由器。
UNTAGGED 表示没有收到对端路由器发来关于此条路由的标签，说明对端路由器没有该路由或是不支持 mpls 。当本地路由器收到数据时， 会弹出标签，并查找 FIB 来寻找下一跳 。

4.3.3 exp-null

impl-null 会将二层帧中的 tag 信息全部弹出，包括涉及 QoS 的信息，为了保留 QoS 信息(EXP 字段)，可以使用 exp-null 。

Figure 18: exp-null

R3(config)#mpls ldp explicit-null

R2#sh mpls forwarding-table
Local      Outgoing   Prefix           Bytes Label   Outgoing   Next Hop
Label      Label      or Tunnel Id     Switched      interface
200        Pop Label  1.1.1.1/32       0             Et0/0      10.1.12.1
201        explicit-n 3.3.3.3/32       0             Et0/1      10.1.23.3

PCAP 💾

4.4 TTL propagation

Figure 19: 默认的 TTL 传播过程

默认的 TTL 操作方式在使用 traceroute 的时候暴露内部网络拓扑，可以通过命令关闭 TTL Propagation 。

Figure 20: 关闭 TTL Propagation 后的传播过程

通常在边界设备上配置关闭该特性 : no mpls ip propagate-ttl

5 MPLS 中 BGP 路由黑洞问题研究

5.1 实验一(原理)

Figure 21: 实验拓扑

实验环境特点:
- R1 与 R4 之间建立 IBGP 邻接关系， IBGP 邻接关系建立在物理接口上
- OSPF 中宣告 R1-R5 和 R4-R6 的直连网段
- R5 及 R6 各自在 BGP 进程中宣告自己的 Loopback 路由

初始配置清单

----------------- R1 ----------------------
en
conf t
 hostname R1
 no ip do lo
 line con 0
  exec-timeout 0 0
  logging synchronous
 exit
 int lo0
  ip address 1.1.1.1 255.255.255.255
 exit
 int e0/0
  ip address 10.1.12.1 255.255.255.0
  no sh
 exit
  int e0/1
  ip address 10.1.15.1 255.255.255.0
  no sh
 exit
 router ospf 110
  router-id 1.1.1.1
  network 10.1.12.1 0.0.0.0 area 0
  network 10.1.15.1 0.0.0.0 area 0
  network 1.1.1.1 0.0.0.0 area 0
 exit
 router bgp 1234
  bgp router-id 1.1.1.1
  neighbor 10.1.15.5 remote-as 500
  neighbor 10.1.34.4 remote-as 1234
 exit
end

----------------- R2 ----------------------
en
conf t
 hostname R2
 no ip do lo
 line con 0
  exec-timeout 0 0
  logging synchronous
 exit
 int lo0
  ip address 2.2.2.2 255.255.255.255
 exit
 int e0/0
  ip address 10.1.12.2 255.255.255.0
  no sh
 exit
 int e0/1
  ip address 10.1.23.2 255.255.255.0
  no sh
 exit
 router ospf 110
  router-id 2.2.2.2
  network 10.1.12.2 0.0.0.0 area 0
  network 10.1.23.2 0.0.0.0 area 0
  network 2.2.2.2 0.0.0.0 area 0
 exit
end

----------------- R3 ----------------------
en
conf t
 hostname R3
 no ip do lo
 line con 0
  exec-timeout 0 0
  logging synchronous
 exit
 int lo0
  ip address 3.3.3.3 255.255.255.255
 exit
 int e0/0
  ip address 10.1.23.3 255.255.255.0
  no sh
 exit
 int e0/1
  ip address 10.1.34.3 255.255.255.0
  no sh
 exit
 router ospf 110
  router-id 3.3.3.3
  network 10.1.23.3 0.0.0.0 area 0
  network 10.1.34.3 0.0.0.0 area 0
  network 3.3.3.3 0.0.0.0 area 0
 exit
end

----------------- R4 ----------------------
en
conf t
 hostname R4
 no ip do lo
 line con 0
  exec-timeout 0 0
  logging synchronous
 exit
 int lo0
  ip address 4.4.4.4 255.255.255.255
 exit
 int e0/0
  ip address 10.1.34.4 255.255.255.0
  no sh
 exit
 int e0/1
  ip address 10.1.46.4 255.255.255.0
  no sh
 exit
 router ospf 110
  router-id 4.4.4.4
  network 10.1.34.4 0.0.0.0 area 0
  network 10.1.46.4 0.0.0.0 area 0
  network 4.4.4.4 0.0.0.0 area 0
 exit
 router bgp 1234
  bgp router-id 4.4.4.4
  neighbor 10.1.46.6 remote-as 600
  neighbor 10.1.12.1 remote-as 1234
 exit
end

----------------- R5 ----------------------
en
conf t
 hostname R5
 no ip do lo
 line con 0
  exec-timeout 0 0
  logging synchronous
 exit
 int lo0
  ip address 5.5.5.5 255.255.255.255
 exit
 int e0/0
  ip address 10.1.15.5 255.255.255.0
  no sh
 exit
 ip route 0.0.0.0 0.0.0.0 10.1.15.1
 router bgp 500
  bgp router-id 5.5.5.5
  neighbor 10.1.15.1 remote-as 1234
  network 5.5.5.5 mask 255.255.255.255
 exit
end

----------------- R6 ----------------------
en
conf t
 hostname R6
 no ip do lo
 line con 0
  exec-timeout 0 0
  logging synchronous
 exit
 int lo0
  ip address 6.6.6.6 255.255.255.255
 exit
 int e0/0
  ip address 10.1.46.6 255.255.255.0
  no sh
 exit
 ip route 0.0.0.0 0.0.0.0 10.1.46.4
 router bgp 600
  bgp router-id 6.6.6.6
  neighbor 10.1.46.4 remote-as 1234
  network 6.6.6.6 mask 255.255.255.255
 exit
end

完整配置清单(高亮为 MPLS 相关配置)

----------------- R1 ----------------------
en
conf t
 hostname R1
 no ip do lo
 line con 0
  exec-timeout 0 0
  logging synchronous
 exit
 int lo0
  ip address 1.1.1.1 255.255.255.255
 exit
 int e0/0
  ip address 10.1.12.1 255.255.255.0
  no sh
 exit
  int e0/1
  ip address 10.1.15.1 255.255.255.0
  no sh
 exit
 router ospf 110
  router-id 1.1.1.1
  network 10.1.12.1 0.0.0.0 area 0
  network 10.1.15.1 0.0.0.0 area 0
  network 1.1.1.1 0.0.0.0 area 0
 exit
 router bgp 1234
  bgp router-id 1.1.1.1
  neighbor 10.1.15.5 remote-as 500
  neighbor 10.1.34.4 remote-as 1234
 exit
 
 ip cef
 mpls ldp router-id lo0
 mpls label range 100 199
 int e0/0
  mpls ip
 exit
 
end

----------------- R2 ----------------------
en
conf t
 hostname R2
 no ip do lo
 line con 0
  exec-timeout 0 0
  logging synchronous
 exit
 int lo0
  ip address 2.2.2.2 255.255.255.255
 exit
 int e0/0
  ip address 10.1.12.2 255.255.255.0
  no sh
 exit
 int e0/1
  ip address 10.1.23.2 255.255.255.0
  no sh
 exit
 router ospf 110
  router-id 2.2.2.2
  network 10.1.12.2 0.0.0.0 area 0
  network 10.1.23.2 0.0.0.0 area 0
  network 2.2.2.2 0.0.0.0 area 0
 exit
 
 ip cef
 mpls ldp router-id lo0
 mpls label range 200 299
 int range e0/0,e0/1
  mpls ip
 exit
 
end

----------------- R3 ----------------------
en
conf t
 hostname R3
 no ip do lo
 line con 0
  exec-timeout 0 0
  logging synchronous
 exit
 int lo0
  ip address 3.3.3.3 255.255.255.255
 exit
 int e0/0
  ip address 10.1.23.3 255.255.255.0
  no sh
 exit
 int e0/1
  ip address 10.1.34.3 255.255.255.0
  no sh
 exit
 router ospf 110
  router-id 3.3.3.3
  network 10.1.23.3 0.0.0.0 area 0
  network 10.1.34.3 0.0.0.0 area 0
  network 3.3.3.3 0.0.0.0 area 0
 exit
 
 ip cef
 mpls ldp router-id lo0
 mpls label range 300 399
 int range e0/0,e0/1
  mpls ip
 exit
 
end

----------------- R4 ----------------------
en
conf t
 hostname R4
 no ip do lo
 line con 0
  exec-timeout 0 0
  logging synchronous
 exit
 int lo0
  ip address 4.4.4.4 255.255.255.255
 exit
 int e0/0
  ip address 10.1.34.4 255.255.255.0
  no sh
 exit
 int e0/1
  ip address 10.1.46.4 255.255.255.0
  no sh
 exit
 router ospf 110
  router-id 4.4.4.4
  network 10.1.34.4 0.0.0.0 area 0
  network 10.1.46.4 0.0.0.0 area 0
  network 4.4.4.4 0.0.0.0 area 0
 exit
 router bgp 1234
  bgp router-id 4.4.4.4
  neighbor 10.1.46.6 remote-as 600
  neighbor 10.1.12.1 remote-as 1234
 exit
 
 ip cef
 mpls ldp router-id lo0
 mpls label range 400 499
 int e0/0
  mpls ip
 exit
 
end

----------------- R5 ----------------------
en
conf t
 hostname R5
 no ip do lo
 line con 0
  exec-timeout 0 0
  logging synchronous
 exit
 int lo0
  ip address 5.5.5.5 255.255.255.255
 exit
 int e0/0
  ip address 10.1.15.5 255.255.255.0
  no sh
 exit
 ip route 0.0.0.0 0.0.0.0 10.1.15.1
 router bgp 500
  bgp router-id 5.5.5.5
  neighbor 10.1.15.1 remote-as 1234
  network 5.5.5.5 mask 255.255.255.255
 exit
end

----------------- R6 ----------------------
en
conf t
 hostname R6
 no ip do lo
 line con 0
  exec-timeout 0 0
  logging synchronous
 exit
 int lo0
  ip address 6.6.6.6 255.255.255.255
 exit
 int e0/0
  ip address 10.1.46.6 255.255.255.0
  no sh
 exit
 ip route 0.0.0.0 0.0.0.0 10.1.46.4
 router bgp 600
  bgp router-id 6.6.6.6
  neighbor 10.1.46.4 remote-as 1234
  network 6.6.6.6 mask 255.255.255.255
 exit
end

5.1.1 连通性分析

R6#ping 5.5.5.5 source 6.6.6.6
Type escape sequence to abort.
Sending 5, 100-byte ICMP Echos to 5.5.5.5, timeout is 2 seconds:
Packet sent with a source address of 6.6.6.6
!!!!!
Success rate is 100 percent (5/5), round-trip min/avg/max = 1/1/2 ms

R6 始发的报文是基于 IP 的，到了 R4 ，R4 查 CEF 表，发现目的地:5.5.5.5 的条目，关联了一个 Label: 303，于是 R4 将 IP 包压上标签 303，然后传给下一跳，即 R3 :

R4#sh ip cef 5.5.5.5 detail
5.5.5.5/32, epoch 0, flags [rib only nolabel, rib defined all labels]
  recursive via 10.1.15.5
    recursive via 10.1.15.0/24
      nexthop 10.1.34.3 Ethernet0/0 label 304-(local:404)

为什么 R4 CEF 显示要用 Label:304 呢? 通过查看 R4 的路由表可以发现，去往 5.5.5.5 的路由下一跳为 10.1.15.5 ，那么 R4 就要查看是否有从 LDP 邻居收到的关于 10.1.15.5 的标签。

R4#sh ip route bgp
Codes: L - local, C - connected, S - static, R - RIP, M - mobile, B - BGP
       D - EIGRP, EX - EIGRP external, O - OSPF, IA - OSPF inter area
       N1 - OSPF NSSA external type 1, N2 - OSPF NSSA external type 2
       E1 - OSPF external type 1, E2 - OSPF external type 2
       i - IS-IS, su - IS-IS summary, L1 - IS-IS level-1, L2 - IS-IS level-2
       ia - IS-IS inter area, * - candidate default, U - per-user static route
       o - ODR, P - periodic downloaded static route, H - NHRP, l - LISP
       a - application route
       + - replicated route, % - next hop override, p - overrides from PfR

Gateway of last resort is not set

      5.0.0.0/32 is subnetted, 1 subnets
B        5.5.5.5 [200/0] via 10.1.15.5, 00:40:23
      6.0.0.0/32 is subnetted, 1 subnets
B        6.6.6.6 [20/0] via 10.1.46.6, 00:24:39

再看下 R3 的 FIB 表，发现它为该路由分配的本地标签正是 304 。

R3#sh mpls ldp bindings local-label 304 ! FIB
  lib entry: 10.1.15.0/24, rev 12
        local binding:  label: 304
R3#sh mpls forwarding-table ! LFIB
Local      Outgoing   Prefix           Bytes Label   Outgoing   Next Hop
Label      Label      or Tunnel Id     Switched      interface
300        200        1.1.1.1/32       0             Et0/0      10.1.23.2
301        Pop Label  2.2.2.2/32       0             Et0/0      10.1.23.2
302        Pop Label  4.4.4.4/32       0             Et0/1      10.1.34.4
303        Pop Label  10.1.12.0/24     87330         Et0/0      10.1.23.2
304        203        10.1.15.0/24     4102          Et0/0      10.1.23.2
305        Pop Label  10.1.46.0/24     3502          Et0/1      10.1.34.4

顺便提一句，10.1.15.0/24 是属于 R1 的直连网段，因此 R2 的转发表中将显示 Pop (PHP 机制):

R2# sh mpls forwarding-table
Local      Outgoing   Prefix           Bytes Label   Outgoing   Next Hop
Label      Label      or Tunnel Id     Switched      interface
200        Pop Label  1.1.1.1/32       0             Et0/0      10.1.12.1
201        Pop Label  3.3.3.3/32       0             Et0/1      10.1.23.3
202        302        4.4.4.4/32       0             Et0/1      10.1.23.3
203        Pop Label  10.1.15.0/24     4446          Et0/0      10.1.12.1
204        Pop Label  10.1.34.0/24     88684         Et0/1      10.1.23.3
205        305        10.1.46.0/24     3626          Et0/1      10.1.23.3

也就是说 R1 将收到纯 IP 数据包，而 R1 上存在去往 5.5.5.5 的路由，所以数据包可以正确转发。

5.2 实验二(PHP 的影响)

Figure 22: 实验拓扑

上一个实验中，在 Core OSPF 中宣告了 R1-R5 及 R4-R6 的直连网段，然而在实际合中，往往不会在 Core IGP 中宣告 AS 外的链路。如果不在 OSPF 中宣告这两个直连链路，会有什么现象。

由于 OSPF 没有了这两个直连网段，那么 5.5.5.5 及 6.6.6.6 两条 EBGP 路由就不能被优化，需在 R1/R4 上对 R4/R1 使用 next-hop-self 。

配置清单

----------------- R1 ----------------------
en
conf t
 hostname R1
 no ip do lo
 line con 0
  exec-timeout 0 0
  logging synchronous
 exit
 int lo0
  ip address 1.1.1.1 255.255.255.255
 exit
 int e0/0
  ip address 10.1.12.1 255.255.255.0
  no sh
 exit
  int e0/1
  ip address 10.1.15.1 255.255.255.0
  no sh
 exit
 router ospf 110
  router-id 1.1.1.1
  network 10.1.12.1 0.0.0.0 area 0
  ! network 10.1.15.1 0.0.0.0 area 0
  network 1.1.1.1 0.0.0.0 area 0
 exit
 router bgp 1234
  bgp router-id 1.1.1.1
  neighbor 10.1.15.5 remote-as 500
  neighbor 10.1.34.4 remote-as 1234
  neighbor 10.1.34.4 next-hop-self
 exit
 ip cef
 mpls ldp router-id lo0
 mpls label range 100 199
 int e0/0
  mpls ip
 exit
end

----------------- R2 ----------------------
en
conf t
 hostname R2
 no ip do lo
 line con 0
  exec-timeout 0 0
  logging synchronous
 exit
 int lo0
  ip address 2.2.2.2 255.255.255.255
 exit
 int e0/0
  ip address 10.1.12.2 255.255.255.0
  no sh
 exit
 int e0/1
  ip address 10.1.23.2 255.255.255.0
  no sh
 exit
 router ospf 110
  router-id 2.2.2.2
  network 10.1.12.2 0.0.0.0 area 0
  network 10.1.23.2 0.0.0.0 area 0
  network 2.2.2.2 0.0.0.0 area 0
 exit
 ip cef
 mpls ldp router-id lo0
 mpls label range 200 299
 int range e0/0,e0/1
  mpls ip
 exit
end

----------------- R3 ----------------------
en
conf t
 hostname R3
 no ip do lo
 line con 0
  exec-timeout 0 0
  logging synchronous
 exit
 int lo0
  ip address 3.3.3.3 255.255.255.255
 exit
 int e0/0
  ip address 10.1.23.3 255.255.255.0
  no sh
 exit
 int e0/1
  ip address 10.1.34.3 255.255.255.0
  no sh
 exit
 router ospf 110
  router-id 3.3.3.3
  network 10.1.23.3 0.0.0.0 area 0
  network 10.1.34.3 0.0.0.0 area 0
  network 3.3.3.3 0.0.0.0 area 0
 exit
 ip cef
 mpls ldp router-id lo0
 mpls label range 300 399
 int range e0/0,e0/1
  mpls ip
 exit
end

----------------- R4 ----------------------
en
conf t
 hostname R4
 no ip do lo
 line con 0
  exec-timeout 0 0
  logging synchronous
 exit
 int lo0
  ip address 4.4.4.4 255.255.255.255
 exit
 int e0/0
  ip address 10.1.34.4 255.255.255.0
  no sh
 exit
 int e0/1
  ip address 10.1.46.4 255.255.255.0
  no sh
 exit
 router ospf 110
  router-id 4.4.4.4
  network 10.1.34.4 0.0.0.0 area 0
  ! network 10.1.46.4 0.0.0.0 area 0
  network 4.4.4.4 0.0.0.0 area 0
 exit
 router bgp 1234
  bgp router-id 4.4.4.4
  neighbor 10.1.46.6 remote-as 600
  neighbor 10.1.12.1 remote-as 1234
  neighbor 10.1.12.1 next-hop-self
 exit
 ip cef
 mpls ldp router-id lo0
 mpls label range 400 499
 int e0/0
  mpls ip
 exit
end

----------------- R5 ----------------------
en
conf t
 hostname R5
 no ip do lo
 line con 0
  exec-timeout 0 0
  logging synchronous
 exit
 int lo0
  ip address 5.5.5.5 255.255.255.255
 exit
 int e0/0
  ip address 10.1.15.5 255.255.255.0
  no sh
 exit
 ip route 0.0.0.0 0.0.0.0 10.1.15.1
 router bgp 500
  bgp router-id 5.5.5.5
  neighbor 10.1.15.1 remote-as 1234
  network 5.5.5.5 mask 255.255.255.255
 exit
end

----------------- R6 ----------------------
en
conf t
 hostname R6
 no ip do lo
 line con 0
  exec-timeout 0 0
  logging synchronous
 exit
 int lo0
  ip address 6.6.6.6 255.255.255.255
 exit
 int e0/0
  ip address 10.1.46.6 255.255.255.0
  no sh
 exit
 ip route 0.0.0.0 0.0.0.0 10.1.46.4
 router bgp 600
  bgp router-id 6.6.6.6
  neighbor 10.1.46.4 remote-as 1234
  network 6.6.6.6 mask 255.255.255.255
 exit
end

5.2.1 不连通性分析

R4#sh ip cef 5.5.5.5 detail
5.5.5.5/32, epoch 0, flags [rib only nolabel, rib defined all labels]
  recursive via 10.1.12.1
    recursive via 10.1.12.0/24 ! 本质上要贴上去往 10.1.12.0/24 网段的标签
      nexthop 10.1.34.3 Ethernet0/0 label 303-(local:403)
R4#

而 R3 上针对这个网段的 LFIB 为:

R3#sh mpls forwarding-table
Local      Outgoing   Prefix           Bytes Label   Outgoing   Next Hop
Label      Label      or Tunnel Id     Switched      interface
300        200        1.1.1.1/32       0             Et0/0      10.1.23.2
301        Pop Label  2.2.2.2/32       0             Et0/0      10.1.23.2
303        Pop Label  10.1.12.0/24     95034         Et0/0      10.1.23.2
304        Pop Label  4.4.4.4/32       0             Et0/1      10.1.34.4

因为 10.1.12.0/24 是属于 R2 的直连网段，所以 R2 发出的针对该网段的标签为 imp-null 。这样 R3 向 R2 转发的其实是 IP 数据包，而非标签数据包，由于 BGP 路由黑洞的存在，R2 上并没有去往 5.5.5.5 的路由，因此产生丢包。

解决方案见实验三。

5.3 实验三 (IBGP on Loopback)

Figure 23: 实验环境

配置清单

----------------- R1 ----------------------
en
conf t
 hostname R1
 no ip do lo
 line con 0
  exec-timeout 0 0
  logging synchronous
 exit
 int lo0
  ip address 1.1.1.1 255.255.255.255
 exit
 int e0/0
  ip address 10.1.12.1 255.255.255.0
  no sh
 exit
  int e0/1
  ip address 10.1.15.1 255.255.255.0
  no sh
 exit
 router ospf 110
  router-id 1.1.1.1
  network 10.1.12.1 0.0.0.0 area 0
  network 1.1.1.1 0.0.0.0 area 0
 exit
 router bgp 1234
  bgp router-id 1.1.1.1
  neighbor 10.1.15.5 remote-as 500
  
  neighbor 4.4.4.4 remote-as 1234
  neighbor 4.4.4.4 next-hop-self
  neighbor 4.4.4.4 update-source lo0
  
 exit
 ip cef
 mpls ldp router-id lo0
 mpls label range 100 199
 int e0/0
  mpls ip
 exit
end

----------------- R2 ----------------------
en
conf t
 hostname R2
 no ip do lo
 line con 0
  exec-timeout 0 0
  logging synchronous
 exit
 int lo0
  ip address 2.2.2.2 255.255.255.255
 exit
 int e0/0
  ip address 10.1.12.2 255.255.255.0
  no sh
 exit
 int e0/1
  ip address 10.1.23.2 255.255.255.0
  no sh
 exit
 router ospf 110
  router-id 2.2.2.2
  network 10.1.12.2 0.0.0.0 area 0
  network 10.1.23.2 0.0.0.0 area 0
  network 2.2.2.2 0.0.0.0 area 0
 exit
 ip cef
 mpls ldp router-id lo0
 mpls label range 200 299
 int range e0/0,e0/1
  mpls ip
 exit
end

----------------- R3 ----------------------
en
conf t
 hostname R3
 no ip do lo
 line con 0
  exec-timeout 0 0
  logging synchronous
 exit
 int lo0
  ip address 3.3.3.3 255.255.255.255
 exit
 int e0/0
  ip address 10.1.23.3 255.255.255.0
  no sh
 exit
 int e0/1
  ip address 10.1.34.3 255.255.255.0
  no sh
 exit
 router ospf 110
  router-id 3.3.3.3
  network 10.1.23.3 0.0.0.0 area 0
  network 10.1.34.3 0.0.0.0 area 0
  network 3.3.3.3 0.0.0.0 area 0
 exit
 ip cef
 mpls ldp router-id lo0
 mpls label range 300 399
 int range e0/0,e0/1
  mpls ip
 exit
end

----------------- R4 ----------------------
en
conf t
 hostname R4
 no ip do lo
 line con 0
  exec-timeout 0 0
  logging synchronous
 exit
 int lo0
  ip address 4.4.4.4 255.255.255.255
 exit
 int e0/0
  ip address 10.1.34.4 255.255.255.0
  no sh
 exit
 int e0/1
  ip address 10.1.46.4 255.255.255.0
  no sh
 exit
 router ospf 110
  router-id 4.4.4.4
  network 10.1.34.4 0.0.0.0 area 0
  ! network 10.1.46.4 0.0.0.0 area 0
  network 4.4.4.4 0.0.0.0 area 0
 exit
 router bgp 1234
  bgp router-id 4.4.4.4
  neighbor 10.1.46.6 remote-as 600
  
  neighbor 1.1.1.1 remote-as 1234
  neighbor 1.1.1.1 next-hop-self
  neighbor 1.1.1.1 update-source lo0
  
 exit
 ip cef
 mpls ldp router-id lo0
 mpls label range 400 499
 int e0/0
  mpls ip
 exit
end

----------------- R5 ----------------------
en
conf t
 hostname R5
 no ip do lo
 line con 0
  exec-timeout 0 0
  logging synchronous
 exit
 int lo0
  ip address 5.5.5.5 255.255.255.255
 exit
 int e0/0
  ip address 10.1.15.5 255.255.255.0
  no sh
 exit
 ip route 0.0.0.0 0.0.0.0 10.1.15.1
 router bgp 500
  bgp router-id 5.5.5.5
  neighbor 10.1.15.1 remote-as 1234
  network 5.5.5.5 mask 255.255.255.255
 exit
end

----------------- R6 ----------------------
en
conf t
 hostname R6
 no ip do lo
 line con 0
  exec-timeout 0 0
  logging synchronous
 exit
 int lo0
  ip address 6.6.6.6 255.255.255.255
 exit
 int e0/0
  ip address 10.1.46.6 255.255.255.0
  no sh
 exit
 ip route 0.0.0.0 0.0.0.0 10.1.46.4
 router bgp 600
  bgp router-id 6.6.6.6
  neighbor 10.1.46.4 remote-as 1234
  network 6.6.6.6 mask 255.255.255.255
 exit
end

6 路由汇总对 MPLS 的影响

Figure 24: 路由汇总示例(该实例中数据仍可正常传输，虽然C上发生过重打标签的情况)

路由汇总会将原先完整一段 LSP 分割成两段
问题的本质在于，在汇聚点上由于数据包已被去除 tag 信息，会以纯 IP 包的方式查找 FIB ，如果此时无法找到下一跳，则会发生丢包，这在 MPLS VPN 环境尤为常见

Footnotes:

Ingress and Egress are known as Edge Routers

加入标签信息是因为如果是 Edge LSR ，当收到未加标签的数据时只会查询 FIB ，而这时就可以直接从 FIB 中提取需要压入的标签

Why does LDP "require" a /32 Loopback?

⁴

Label allocation, label imposing, label swapping, and label popping usually happen in the service provider network, not the customer (enterprise) network. Customer routers will never see a label.

⁵

MPLS 只针对 IGP 分发标签，不会为 BGP 路由产生标签

MPLS {Back to Index} Expand All Details on Page

Table of Contents

1 术语

1.1 FEC ( Forwarding Equivalence Class )

1.2 CEF

1.3 LSR

1.3.1 Edge LSR 1

1.4 FIB/LIB/LFIB

2 标签格式

3 LDP

3.1 基本配置

3.2 查看命令

3.2.1 查看邻居

3.2.2 查看是否收到 Hello

3.2.3 查看 FIB 表

3.2.4 查看 LIB 表

3.2.5 查看 LFIB 表

3.3 邻居建立过程

4 原理

4.1 标签建立过程 4

4.1.1 依赖 IGP 构建 RIB 与 FIB(CEF) 5

4.1.2 运行 LDP 以维护 LIB

4.1.3 基于 LIB 和 FIB 维护 LFIB

4.2 数据包传递过程

4.3 Penultimate Hop Popping (次末跳优化)

4.3.1 imp-null

4.3.2 POP 与 UNTAGGED 区别

4.3.3 exp-null

4.4 TTL propagation

5 MPLS 中 BGP 路由黑洞问题研究

5.1 实验一(原理)

5.1.1 连通性分析

5.2 实验二(PHP 的影响)

5.2.1 不连通性分析

5.3 实验三 (IBGP on Loopback)

6 路由汇总对 MPLS 的影响

Footnotes:

MPLS
{Back to Index}

1.3.1 Edge LSR ¹

4.1 标签建立过程 ⁴

4.1.1 依赖 IGP 构建 RIB 与 FIB(CEF) ⁵