前言

环境信息：

操作系统：Ubuntu 18.04
VPP：v21.01

VPP 的 L2TP 实现的是剥除了 IPv6 和隧道封装，并留下内部二层头，该模型基于 RFC 4719 中的 PW termination 模型，实现了 LCCE 功能，而非实现了 LAC 或 LNS 功能。可见，VPP 目前不具备、也不区分 LAC 和 LNS 的角色，即：VPP1、2 互为 user-to-net/net-to-user。

组网拓扑

VPP1 配置

Linux Host 执行

创建 veth pair 虚拟网络设备，其中 vpp1out 作为 VPP1 的 Host Interface。

$ ip link add name vpp1out type veth peer name vpp1host
$ ip link set dev vpp1out up
$ ip link set dev vpp1host up
$ ip addr add 10.1.1.1/24 dev vpp1host
$ ip l
6: vpp1host@vpp1out: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP mode DEFAULT group default qlen 1000link/ether aa:ac:58:9d:9b:33 brd ff:ff:ff:ff:ff:ff
7: vpp1out@vpp1host: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP mode DEFAULT group default qlen 1000link/ether 8e:a6:98:fe:01:d7 brd ff:ff:ff:ff:ff:ff

VPP CLI 执行

配置开启 L2TP plugin，配置 L2TP 使用 lookup-session-id 作为查询条件。

$ vi /etc/vpp/startup.conf
...
plugins {...plugin l2tp_plugin.so { enable }plugin l2tp_test_plugin.so { enable }
}
...
l2tp {lookup-session-id
}

NOTE 1：The vpp stack defaults to “lookup-v6-dst.” In the l2-interface-to-tunnel direction, the vpp stack uses the l2 sub-interface as the session lookup key.

配置 Physical Interface IPv6 地址。

vpp# set ip6 address GigabitEthernet0/7/0 fd00::1/8
vpp# set int state GigabitEthernet0/7/0 up

创建 L2TPv3 Tunnel 和 Session，同时会自动创建一个虚拟的路由子接口 l2tpv3_tunnel0。

create l2tpv3 tunnel client <ip6-addr> our <ip6-addr>[local-cookie <%llx>] [remote-cookie <%llx>] [local-session-id <dec-num>] [remote-session-id <dec-num>][l2-sublayer-present]vpp# create l2tpv3 tunnel our fd00::1 client fd00::2 local-session-id 10 remote-session-id 20
l2tpv3_tunnel0

NOTE 2：Local-cookie and remote-cookie default to FFFFFFFFFFFFFFFF. Local-session-id and remote-session-id default to 1.

NOTE 3：Since the l2 sublayer is used to transport sequence information – not present in static l2tpv3 softwires – it seems that the Linux implementation is incorrect.

When interoperating with the current Linux l2tpv3-in-v6 implementation, you’ll need to specify the l2-sublayer-present property. Conversely.
When interoperating with IOS-XR, do not specify l2-sublayer-present.

因为 l2tpv3_tunnel0 是一个虚拟接口，需要与 bridge-domain 进行关联。

vpp# set interface l2 bridge l2tpv3_tunnel0 1
vpp# set interface state l2tpv3_tunnel0 up

NOTE 4：The vpp stack strips the ip6 and tunnel encaps and tracks down the inner l2 header. Vlan tags can be inserted using subinterface VLAN tag-rewrite commands。

Set interface ip6 l2tpv3：In the tunnel-to-l2-interface direction, decapsulation must be explicitly enabled on the input interface。

vpp# set interface ip6 l2tpv3 GigabitEthernet0/7/0
# This can be deconfigured as well:
#set interface ip6 l2tpv3 GigabitEthernet0/7/0 del

创建 Host Interface 并关联到与 L2TP Virtual Interface 的同一个 bridge-domain 上，作为 “外部路由子接口”。

vpp# create host-interface name vpp1out
host-vpp1outvpp# set int state host-vpp1out up
vpp# set int l2 bridge host-vpp1out 1

检查 bridge-domain 1 和 l2tpv3 tunnel 的详情。

vpp# show bridge-domain 1 detailBD-ID   Index   BSN  Age(min)  Learning  U-Forwrd   UU-Flood   Flooding  ARP-Term  arp-ufwd   BVI-Intf1       1       off        on        on       flood        on       off       off        N/A
span-l2-input l2-input-classify l2-input-feat-arc l2-policer-classify l2-input-acl vpath-input-l2 l2-ip-qos-record l2-input-vtr l2-gbp-lpm-classify gbp-src-classify gbp-null-classify l2-gbp-lpm-anon-classify gbp-learn-l2 l2-emulation l2-learn l2-rw l2-fwd gbp-fwd l2-flood l2-flood l2-outputInterface           If-idx ISN  SHG  BVI  TxFlood        VLAN-Tag-Rewritel2tpv3_tunnel0           2     1    0    -      *                 nonehost-vpp1out            3     1    0    -      *                 nonevpp# show l2tpv3 verbose
1 l2tp sessions...
L2tp session lookup on session id
[0] fd00::1 (our) fd00::2 (client) l2tpv3_tunnel0 (sw_if_index 2)local cookies ffffffffffffffff 0000000000000000 remote cookie fffffffffffffffflocal session-id 10 remote session-id 20l2 specific sublayer absentuser-to-net: 2353 pkts 378164 bytesnet-to-user: 2353 pkts 222866 bytes

另外，必要时，还可以修改 L2TPv3 Rollover Cookie：

set l2tpv3 tunnel cookie <interface> [local <64bit hex value>] [remote <64bit hex value>]

NOTE 5：If the local cookie value is specified, both the existing cookie value and the specified rollover value will be accepted. If the remote cookie value is specified, it will immediately replace the existing remote cookie value.

VPP2 配置

Linux Host 执行

$ ip link add name vpp2out type veth peer name vpp2host
$ ip link set dev vpp2out up
$ ip link set dev vpp2host up
$ ip addr add 10.1.1.2/24 dev vpp2host
$ ip l
6: vpp2host@vpp2out: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP mode DEFAULT group default qlen 1000link/ether 2a:82:d4:20:32:7a brd ff:ff:ff:ff:ff:ff
7: vpp2out@vpp2host: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP mode DEFAULT group default qlen 1000link/ether 4e:29:e9:9f:84:f0 brd ff:ff:ff:ff:ff:ff

VPP 执行

set ip6 address GigabitEthernet0/4/0 fd00::2/8
set int state GigabitEthernet0/4/0 upcreate l2tpv3 tunnel our fd00::2 client fd00::1 local-session-id 20 remote-session-id 10set interface l2 bridge l2tpv3_tunnel0 1
set interface state l2tpv3_tunnel0 upset interface ip6 l2tpv3 GigabitEthernet0/4/0create host-interface name vpp2out
set int state host-vpp2out up
set int l2 bridge host-vpp2out 1

测试验证

Host1 ping Host2

$ ping 10.1.1.2
PING 10.1.1.2 (10.1.1.2) 56(84) bytes of data.
64 bytes from 10.1.1.2: icmp_seq=1 ttl=64 time=0.241 ms

检查 VPP1 的 L2 MAC 学习表。

vpp# show ip neighborTime                       IP                    Flags      Ethernet              Interface188.1579                 fd00::2                   D    fa:16:3e:61:6b:4e GigabitEthernet0/7/0

抓 GigabitEthernet0/7/0 的包。

pcap trace tx rx intfc GigabitEthernet0/7/0 max 10000 file vpp1-phy.pcap
pcap trace tx rx status
pcap trace off

4. 抓 Host Interface 的包。

数据报文跟踪。

vpp# trace add dpdk-input 10vpp# show trace
------------------- Start of thread 0 vpp_main -------------------
Packet 100:11:22:393674: dpdk-inputGigabitEthernet0/7/0 rx queue 0buffer 0x9b861: current data 0, length 164, buffer-pool 0, ref-count 1, totlen-nifb 0, trace handle 0x0ext-hdr-validl4-cksum-computed l4-cksum-correctPKT MBUF: port 0, nb_segs 1, pkt_len 164buf_len 2176, data_len 164, ol_flags 0x0, data_off 128, phys_addr 0xaa2e18c0packet_type 0x0 l2_len 0 l3_len 0 outer_l2_len 0 outer_l3_len 0rss 0x0 fdir.hi 0x0 fdir.lo 0x0IP6: fa:16:3e:61:6b:4e -> fa:16:3e:65:42:61L2TP: fd00::2 -> fd00::1tos 0x00, flow label 0x0, hop limit 254, payload length 110
00:11:22:393696: ethernet-inputframe: flags 0x1, hw-if-index 1, sw-if-index 1IP6: fa:16:3e:61:6b:4e -> fa:16:3e:65:42:61
00:11:22:393714: ip6-inputL2TP: fd00::2 -> fd00::1tos 0x00, flow label 0x0, hop limit 254, payload length 110
00:11:22:393718: l2tp-decapL2T: fd00::2 (client) -> fd00::1 (our) session 0
00:11:22:393724: l2-inputl2-input: sw_if_index 2 dst aa:ac:58:9d:9b:33 src 2a:82:d4:20:32:7a [l2-learn l2-fwd l2-flood l2-flood ]
00:11:22:393727: l2-learnl2-learn: sw_if_index 2 dst aa:ac:58:9d:9b:33 src 2a:82:d4:20:32:7a bd_index 1
00:11:22:393728: l2-fwdl2-fwd:   sw_if_index 2 dst aa:ac:58:9d:9b:33 src 2a:82:d4:20:32:7a bd_index 1 result [0x1030000000003, 3] none
00:11:22:393731: l2-outputl2-output: sw_if_index 3 dst aa:ac:58:9d:9b:33 src 2a:82:d4:20:32:7a data 08 00 45 00 00 54 30 9a 00 00 40 01
00:11:22:393732: host-vpp1out-outputhost-vpp1outIP4: 2a:82:d4:20:32:7a -> aa:ac:58:9d:9b:33ICMP: 10.1.1.2 -> 10.1.1.1tos 0x00, ttl 64, length 84, checksum 0x340b dscp CS0 ecn NON_ECNfragment id 0x309aICMP echo_reply checksum 0xfdb4 id 13965vpp# clear tracevpp# trace add af-packet-input 10vpp# show trace
------------------- Start of thread 0 vpp_main -------------------
Packet 100:12:42:388572: af-packet-inputaf_packet: hw_if_index 3 next-index 4tpacket2_hdr:status 0x20000001 len 98 snaplen 98 mac 66 net 80sec 0x60aa1b01 nsec 0x2876c9d6 vlan 0 vlan_tpid 0
00:12:42:388592: ethernet-inputIP4: aa:ac:58:9d:9b:33 -> 2a:82:d4:20:32:7a
00:12:42:388601: l2-inputl2-input: sw_if_index 3 dst 2a:82:d4:20:32:7a src aa:ac:58:9d:9b:33 [l2-learn l2-fwd l2-flood l2-flood ]
00:12:42:388604: l2-learnl2-learn: sw_if_index 3 dst 2a:82:d4:20:32:7a src aa:ac:58:9d:9b:33 bd_index 1
00:12:42:388605: l2-fwdl2-fwd:   sw_if_index 3 dst 2a:82:d4:20:32:7a src aa:ac:58:9d:9b:33 bd_index 1 result [0x1030000000002, 2] none
00:12:42:388608: l2-outputl2-output: sw_if_index 2 dst 2a:82:d4:20:32:7a src aa:ac:58:9d:9b:33 data 08 00 45 00 00 54 78 b1 40 00 40 01
00:12:42:388609: l2tp-encapL2T: fd00::1 (our) -> fd00::2 (client) session 0)
00:12:42:388612: ip6-lookupfib 0 dpo-idx 3 flow hash: 0x00000000L2TP: fd00::1 -> fd00::2tos 0x00, flow label 0x0, hop limit 255, payload length 110
00:12:42:388615: ip6-rewritetx_sw_if_index 1 adj-idx 3 : ipv6 via fd00::2 GigabitEthernet0/7/0: mtu:9000 next:3 fa163e616b4efa163e65426186dd flow hash: 0x0000000000000000: fa163e616b4efa163e65426186dd60000000006e73fefd00000000000000000000000020: 000000000001fd00000000000000000000000000000200000014ffffffffffff00000040: ffff2a82d420327aaaac589d9b3308004500005478b140004001abf30a01010100000060: 0a01010208007dc6368e0001011baa6000000000cf5b0a0000000000
00:12:42:388618: GigabitEthernet0/7/0-outputGigabitEthernet0/7/0IP6: fa:16:3e:65:42:61 -> fa:16:3e:61:6b:4eL2TP: fd00::1 -> fd00::2tos 0x00, flow label 0x0, hop limit 254, payload length 110
00:12:42:388620: GigabitEthernet0/7/0-txGigabitEthernet0/7/0 tx queue 0buffer 0x97e4e: current data -66, length 164, buffer-pool 0, ref-count 1, totlen-nifb 0, trace handle 0x0l2-hdr-offset 0 l3-hdr-offset 14PKT MBUF: port 65535, nb_segs 1, pkt_len 164buf_len 2176, data_len 164, ol_flags 0x0, data_off 62, phys_addr 0xac7f9400packet_type 0x0 l2_len 0 l3_len 0 outer_l2_len 0 outer_l3_len 0rss 0x0 fdir.hi 0x0 fdir.lo 0x0IP6: fa:16:3e:65:42:61 -> fa:16:3e:61:6b:4eL2TP: fd00::1 -> fd00::2tos 0x00, flow label 0x0, hop limit 254, payload length 110vpp# clear trace

FD.io/VPP — L2TP相关推荐

FD.io/VPP — 常用指令集合
目录文章目录目录前言 System Level Interface Add NIC into VPP as Interface Interface State Hardware Interfac ...
FD.io/VPP — Overview
目录文章目录目录 FD.io VPP FD.io 官网:https://fd.io FD.io(Fast data – Input/Output)是 Linux 基金会下属的一个开源项目,成立于 ...
FD.io VPP 20.09版本正式发布：往期VPP文章回顾+下载地址+相关链接
目录下载RPM/DEB包往期文章回顾与推荐 FD.io是一些项目和库的集合,基于DPDK并逐渐演化,支持在通用硬件平台上部署灵活可变的业务.FD.io为软件定义基础设施的开发者提供了一个通用平台, ...
FD.io VPP：探究分段场景下vlib_buf在收发包的处理（dpdk_plugin.so）、rte_mbuf与vlib_buf 关系
Table of Contents rte_mbuf.vlib_buf 关系及内存分布使用dpdk-收包接口函数使用dpdk 发包接口函数总结参考阅读在使用vpp老版本copy报文的时候,经 ...
FD.io VPP：vlib buffer pool（vlib_buffer）内存初始化
Table of Contents vlib buffer创建过程 vlib_buffer相关内存初始化 1.函数一开始就查询numa的个数 2.遍历numa节点来初始化 3.查询系统大页大小. 4. ...
关于FD.io VPP的最新消息
Table of Contents CuVPP:软件数据平面中基于过滤器的最长前缀匹配快速数据项目的矢量包处理器(VPP)版本20.05 2020年打破神话的DPDK 在商品硬件上使用Calico ...
FD.io VPP环境下运行用户应用程序教程
FD.io VPP环境下运行用户应用程序教程 RToax 2020年9月相关文章:<FD.io VPP利用iperf3进行UDP灌包测试-英特尔X520万兆网卡> 1. VPP简介 VP ...
FD.io VPP用户文档：会话层架构与VPP应用
FD.io VPP:用户文档 HostStack RToax 2020年9月 HostStack 1. 描述 VPP的主机协议栈是利用VPP传输,会话和应用程序层协议的用户空间实现.它大致包括四个主要 ...
FD.io VPP：用户文档：VPP RPM包的构建与离线安装
VPP RPM包的构建与离线安装 RToax 2020年9月要开始使用VPP进行开发,您需要获取所需的VPP源代码,然后构建软件包. 1. 设置代理如果你的服务器本地可以连接互联网,则跳过设置代理 ...

FD.io/VPP — L2TP

目录

文章目录

前言

组网拓扑

VPP1 配置

Linux Host 执行

VPP CLI 执行

VPP2 配置

Linux Host 执行

VPP 执行

测试验证

FD.io/VPP — L2TP相关推荐

最新文章

热门文章