知识：
Docker容器（打包环境加应用）
静态镜像运行container
docker run（start）形成拓扑
目的：switch1获取switch2的网络统计数据
start.sh启动拓扑，配置Ip地址，Pin通，每台交换机收集本机信息
test 用sonic telementary取得对端交换机信息，并写入数据库
stop
load image
pull ubuntu（系统）
sonic.p4(镜像）
docke.file
需求支持，在原有基础上新增内容
sudo docker ps 正在运行的...
sudo docker run 运行 -> 静态文件变成操作系统
add bridge 加端口
add point
构造拓扑配置命名为switch1
ethenet0 统计信息 arp广播
reads数据库（交换机收集的信息存到数据库，否则其他交换机无法配置）
test.ping 首先联通 -c5 ping 包发5个
moliterlization -> grp（收集数据）
conters数据库收集1，0口，1交换机，2交换机
sudo ./start sh.实验 目的ip
交换机收集数据->流量转发，数据收集->为了网络配置，更改方案
实验
我们有以下拓扑：主机1（Ubuntu 14:04，192.168.1.2/24）<-->交换机1（Sonic）<-->交换机2（Sonic）<-->主机2（Ubuntu 14:04，192.168.2.2/24）
1）我们执行命令：./install_requirement.sh 通过这种方式，我们安装了Docker、Open-VSSwitch和Bridge-Utils。
2)我们执行命令：./load_image.sh 从Sonic-P4中加载图像并构建ESA和监控客户端
3)我们执行命令：./start.sh 建立和准备环境。一旦执行，我们应该运行5个容器。我们可以通过运行ps docker命令来检查它。
4)我们等待~ 3分钟，以便正确配置一切...我们执行命令：./test.sh 这样，主机1将ping到主机2，反之亦然，我们将测量GNMI Client的遥测。
5)我们最终执行了命令：./stop.sh 这将删除使用Docker和OVS（Open Virtual Switch）创建的容器和桥。我们还删除了我们创建的Docker“管理”网络。
持续集成服务Travis
Dockerfile_golang1
Dokerfile_sonic
install_requirements.sh
load_image.sh
start.sh （建立拓扑）
#!/bin/bashsudo docker run --net=none --privileged --entrypoint /bin/bash --name switch1 -it -d -v $PWD/switch1:/sonic docker-sonic-p4:latestsudo docker run --net=none --privileged --entrypoint /bin/bash --name switch2 -it -d -v $PWD/switch2:/sonic docker-sonic-p4:latestsudo docker run --net=none --privileged --entrypoint /bin/bash --name host1 -it -d ubuntu:14.04sudo docker run --net=none --privileged --entrypoint /bin/bash --name host2 -it -d ubuntu:14.04sudo ovs-vsctl add-br switch1_switch2sudo ovs-docker add-port switch1_switch2 sw_port0 switch1sudo ovs-docker add-port switch1_switch2 sw_port0 switch2sudo ovs-vsctl add-br host1_switch1sudo ovs-docker add-port host1_switch1 sw_port1 switch1sudo ovs-docker add-port host1_switch1 eth1 host1sudo ovs-vsctl add-br host2_switch2sudo ovs-docker add-port host2_switch2 sw_port1 switch2sudo ovs-docker add-port host2_switch2 eth1 host2sudo docker exec -d host1 sysctl net.ipv6.conf.eth0.disable_ipv6=1sudo docker exec -d host1 sysctl net.ipv6.conf.eth1.disable_ipv6=1sudo docker exec -d host2 sysctl net.ipv6.conf.eth0.disable_ipv6=1sudo docker exec -d host2 sysctl net.ipv6.conf.eth1.disable_ipv6=1sudo docker exec -d host1 ifconfig eth1 192.168.1.2/24 mtu 1400sudo docker exec -d host1 ip route replace default via 192.168.1.1sudo docker exec -d host2 ifconfig eth1 192.168.2.2/24 mtu 1400sudo docker exec -d host2 ip route replace default via 192.168.2.1sudo docker exec -d switch1 ip netns add sw_netsudo docker exec -d switch1 ip link set dev sw_port0 netns sw_netsudo docker exec -d switch1 ip netns exec sw_net sysctl net.ipv6.conf.sw_port0.disable_ipv6=1sudo docker exec -d switch1 ip netns exec sw_net ip link set sw_port0 upsudo docker exec -d switch1 ip link set dev sw_port1 netns sw_netsudo docker exec -d switch1 ip netns exec sw_net sysctl net.ipv6.conf.sw_port1.disable_ipv6=1sudo docker exec -d switch1 ip netns exec sw_net ip link set sw_port1 upsudo docker exec -d switch2 ip netns add sw_netsudo docker exec -d switch2 ip link set dev sw_port0 netns sw_netsudo docker exec -d switch2 ip netns exec sw_net sysctl net.ipv6.conf.sw_port0.disable_ipv6=1sudo docker exec -d switch2 ip netns exec sw_net ip link set sw_port0 upsudo docker exec -d switch2 ip link set dev sw_port1 netns sw_netsudo docker exec -d switch2 ip netns exec sw_net sysctl net.ipv6.conf.sw_port1.disable_ipv6=1sudo docker exec -d switch2 ip netns exec sw_net ip link set sw_port1 up##########################################################################################################Bridge de gestion de contenedoressudo docker network create \ --driver bridge \ --subnet=192.18.0.0/24 \ --gateway=192.18.0.1 \ --opt "com.docker.network.bridge.name"="gestion" \ gestion#Creacion de contenedor con gnmi_getsudo docker run --privileged --entrypoint /bin/sh --name gnmicli -it -d gnmi_clientsudo ./iftobridge add-link mgmt1 switch1 gestion --sip="192.18.0.11/24"sudo ./iftobridge add-link mgmt1 switch2 gestion --sip="192.18.0.12/24"sudo ./iftobridge add-link mgmt1 gnmicli gestion --sip="192.18.0.10/24"#Actualizacion de redisDBsudo docker exec -d switch1 sh /sonic/scripts/update_redisDB.sh &sudo docker exec -d switch2 sh /sonic/scripts/update_redisDB.sh &#Activamos la telemetría en los switches:sudo docker exec -d switch1 ./sonic/telemetry --port 8080 --insecure --logtostderr --allow_no_client_auth &sudo docker exec -d switch2 ./sonic/telemetry --port 8080 --insecure --logtostderr --allow_no_client_auth &#########################################################################################################echo "Booting switches, please wait ~3 minutes for switches to load"sudo docker exec -d switch1 sh /sonic/scripts/startup.shsudo docker exec -d switch2 sh /sonic/scripts/startup.shsleep 180docker ps
stop.sh
#!/bin/bashsudo docker rm -f switch1sudo docker rm -f switch2sudo docker rm -f host1sudo docker rm -f host2sudo docker rm -f gnmiclisudo ovs-vsctl del-br switch1_switch2sudo ovs-vsctl del-br host1_switch1sudo ovs-vsctl del-br host2_switch2sudo docker network prune -f
test.sh
#!/bin/bash#Ping desde host1 a switch1sudo docker exec -it host1 ping 192.168.1.1 -c5#Ping desde switch1 a host1sudo docker exec -it switch1 ping 192.168.1.2 -c5#Ping desde host1 a host2sudo docker exec -it host1 ping 192.168.2.2 -c5#Ping desde host2 a switch2sudo docker exec -it host2 ping 192.168.2.1 -c5#Ping desde switch2 a host2sudo docker exec -it switch2 ping 192.168.2.2 -c5#Ping desde host2 a host1sudo docker exec -it host2 ping 192.168.1.2 -c5sleep 10#Monitorización switch1sudo echo "Paquetes recibidos por Ethernet 1 en switch1"sudo docker exec -it gnmicli ./bin/gnmi_get -xpath_target COUNTERS_DB -xpath interface:Ethernet1/in-pkts -target_addr 192.18.0.11:8080 -target_name switch1 -insecure -logtostderrsudo echo "Paquetes recibidos por Ethernet 0 en switch1"sudo docker exec -it gnmicli ./bin/gnmi_get -xpath_target COUNTERS_DB -xpath interface:Ethernet0/in-pkts -target_addr 192.18.0.11:8080 -target_name switch1 -insecure -logtostderr#Monitorización switch2sudo echo "Paquetes recibidos por Ethernet 1 en switch2"sudo docker exec -it gnmicli ./bin/gnmi_get -xpath_target COUNTERS_DB -xpath interface:Ethernet1/in-pkts -target_addr 192.18.0.12:8080 -target_name switch2 -insecure -logtostderrsudo echo "Paquetes recibidos por Ethernet 0 en switch2"sudo docker exec -it gnmicli ./bin/gnmi_get -xpath_target COUNTERS_DB -xpath interface:Ethernet0/in-pkts -target_addr 192.18.0.12:8080 -target_name switch2 -insecure -logtostderr
iftobridge
#!/bin/bash# Copyright (C) 2014 Nicira, Inc.## Licensed under the Apache License, Version 2.0 (the "License");# you may not use this file except in compliance with the License.# You may obtain a copy of the License at:## http://www.apache.org/licenses/LICENSE-2.0## Unless required by applicable law or agreed to in writing, software# distributed under the License is distributed on an "AS IS" BASIS,# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.# See the License for the specific language governing permissions and# limitations under the License.# Check for programs we'll need.search_path () { save_IFS=$IFS IFS=: for dir in $PATH; do IFS=$save_IFS if test -x "$dir/$1"; then return 0 fi done IFS=$save_IFS echo >&2 "$0: $1 not found in \$PATH, please install and try again" exit 1}ovs_vsctl () { ovs-vsctl --timeout=60 "$@"}create_netns_link () { mkdir -p /var/run/netns if [ ! -e /var/run/netns/"$SPID" ]; then ln -s /proc/"$SPID"/ns/net /var/run/netns/"$SPID" trap 'delete_netns_link_source' 0 for signal in 1 2 3 13 14 15; do trap 'delete_netns_link_source; trap - $signal; kill -$signal $$' $signal done fi}delete_netns_link_source () { rm -f /var/run/netns/"$SPID"}add_port () { SINT="$1" SCNT="$2" BRIDGE="$3" if [ -z "$SINT" ] || [ -z "$SCNT" ]; then echo >&2 "$UTIL add-link: not enough arguments (use --help for help)" exit 1 fi shift 3 while [ $# -ne 0 ]; do case $1 in --sip=*) SADDR=`expr X"$1" : 'X[^=]*=\(.*\)'` shift ;; *) echo >&2 "$UTIL add-link: unknown option \"$1\"" exit 1 ;; esac done if [ -z "$SADDR" ]; then echo >&2 "$UTIL add-link: not enough arguments (use --help for help)" exit 1 fi # Commented we assume it already exists #if brctl addbr "$BRIDGE" ; then :; else # echo >&2 "$UTIL: Failed to add bridge $BRIDGE" # exit 1 #fi if SPID=`docker inspect -f '{{.State.Pid}}' "$SCNT"`; then :; else echo >&2 "$UTIL: Failed to get the PID of the container" exit 1 fi create_netns_link # Create a veth pair. asp="${SCNT}_${SINT}" asp2=`echo "$asp" | md5sum | cut -f1 -d" "` SPORTNAME=${asp2:0:12} ip link add "${SPORTNAME}_l" type veth peer name "${SPORTNAME}_c" # Add one end of veth to OVS bridge. if brctl addif "$BRIDGE" "${SPORTNAME}_l"; then :; else echo >&2 "$UTIL: Failed to add "${SPORTNAME}_l" port to bridge $BRIDGE" ip link delete "${SPORTNAME}_l" exit 1 fi ip link set "${SPORTNAME}_l" up # Move "${PORTNAME}_c" inside the container and changes its name. ip link set "${SPORTNAME}_c" netns "$SPID" ip netns exec "$SPID" ip link set dev "${SPORTNAME}_c" name "$SINT" ip netns exec "$SPID" ip link set "$SINT" up if [ -n "$SADDR" ]; then ip netns exec "$SPID" ip addr add "$SADDR" dev "$SINT" fi # This is "just in case..." ifconfig "$BRIDGE" up # This is also "just in case..." sudo iptables -A FORWARD -p all -i "$BRIDGE" -j ACCEPT}del_port () { SINT="$1" SCNT="$2" if [ "$#" -lt 2 ]; then usage exit 1 fi asp="${SCNT}_${SINT}" asp2=`echo "$asp" | md5sum | cut -f1 -d" "` SPORT=${asp2:0:12} ip link delete "${SPORT}_l"}usage() { cat << EOF${UTIL}: Performs integration of Open vSwitch with Docker.usage: ${UTIL} COMMANDCommands: add-link SINT SCNT BRIDGE --sip="ADDRESS" del-link SINT SCNTOptions: -h, --help display this help message.EOF}UTIL=$(basename $0)search_path brctlsearch_path dockersearch_path uuidgenif (ip netns) > /dev/null 2>&1; then :; else echo >&2 "$UTIL: ip utility not found (or it does not support netns),"\ "cannot proceed" exit 1fiif [ $# -eq 0 ]; then usage exit 0ficase $1 in "add-link") shift add_port "$@" exit 0 ;; "del-link") shift del_port "$@" exit 0 ;; -h | --help) usage exit 0 ;; *) echo >&2 "$UTIL: unknown command \"$1\" (use --help for help)" exit 1 ;;esac
实验内容
问题1:在虚拟机导入文件
解决方案：使用U盘导入
问题2：权限
解决方案: sudo passwd输入新密码
su root
增加权限：sudo chmod -R 777
步骤1：./install_requirements.sh
问题3：docker容器无法访问
sudo apt-get install docker.io
