1、路由配置
路由是互联网络的核心，没有路由的网络如同一座孤岛，掌握路由的配置是IT人员的必备技能。
例如：现在有三台主机需要通信，其中A和B在同一网段，C在另一网段，这两个网段有三个路由相隔，如何实现他们之间的通信呢？
主机A:IP=192.168.1.100/24
主机B:IP=192.168.1.63/24
主机C:IP=10.2.110.100/16
R1的接口0:IP=192.168.1.1/24,接口1:IP=110.1.24.10/24
R2的接口0:IP=110.1.24.20/24,接口1:IP=72.98.2.10/16
R3的接口0:IP=72.98.70.20/16,接口1:IP=10.2.0.1/16
通过分析上面的网络环境，可以得到R1,R2和R3的路由信息，这里我们指定每一个路由的静态路由表
R1：路由表
网段 网关接口
192.168.1.0/24 0.0.0.0/0 eth0
110.1.24.10/24 0.0.0.0/0 eth1
72.98.0.0/16 110.1.24.20 eth1
10.2.0.0/16 110.1.24.20 eth1
0.0.0.0/0 110.1.24.20 eth1
R2：路由表
网段 网关接口
192.168.1.0/24 110.1.24.10 eth0
110.1.24.10/24 0.0.0.0/0 eth0
72.98.0.0/16 0.0.0.0/0 eth1
10.2.0.0/1672.98.70.20 eth1
0.0.0.0/0 外网IP(这里不写)
R3：路由表
网段 网关接口
192.168.1.0/24 72.98.2.10 eth0
110.1.24.10/24 72.98.2.10 eth0
72.98.0.0/16 0.0.0.0/0 eth0
10.2.0.0/160.0.0.0/0 eth1
0.0.0.0/0 72.98.2.10 eth0
这里用3台centos系统作为路由
用node1主机来做route1
[root@node1~]#ipaddrshowdeveth13:eth1: BROADCAST,MULTICAST,UP,LOWER_UP mtu1500qdiscpfifo_faststateUPqlen1000link/ether00:0c:29:e2:96:7cbrdff:ff:ff:ff:ff:ffinet192.168.1.1/24scopeglobaleth1inet6fe80::20c:29ff:fee2:967c/64scopelinkvalid_lftforeverpreferred_lftforever[root@node1~]#ipaddrshowdeveth24:eth2: BROADCAST,MULTICAST,UP,LOWER_UP mtu1500qdiscpfifo_faststateUPqlen1000link/ether00:0c:29:e2:96:86brdff:ff:ff:ff:ff:ffinet110.1.24.10/24scopeglobaleth2inet6fe80::20c:29ff:fee2:9686/64scopelinkvalid_lftforeverpreferred_lftforever
[root@node1~]#routeadd-net10.2.0.0/16gw110.1.24.20deveth2[root@node1~]#route-nKernelIProutingtableDestinationGatewayGenmaskFlagsMetricRefUseIface192.168.1.00.0.0.0255.255.255.0U000eth1110.1.24.00.0.0.0255.255.255.0U000eth210.2.0.0110.1.24.20255.255.0.0UG000eth272.98.0.0110.1.24.20255.255.0.0UG000eth210.1.0.00.0.0.0255.255.0.0U000eth0169.254.0.00.0.0.0255.255.0.0U10020
[root@node1~]#echo1 /proc/sys/net/ipv4/ip_forward
note2用来做route2
[root@node2~]#ipaddradd110.1.24.20/24deveth1[root@node2~]#ipaddradd72.98.2.10/16deveth2[root@node2~]#ipaddrshowdeveth13:eth1: BROADCAST,MULTICAST,UP,LOWER_UP mtu1500qdiscpfifo_faststateUPqlen1000link/ether00:0c:29:00:90:24brdff:ff:ff:ff:ff:ffinet110.1.24.20/24scopeglobaleth1inet6fe80::20c:29ff:fe00:9024/64scopelinkvalid_lftforeverpreferred_lftforever[root@node2~]#ipaddrshowdeveth24:eth2: BROADCAST,MULTICAST,UP,LOWER_UP mtu1500qdiscpfifo_faststateUPqlen1000link/ether00:0c:29:00:90:2ebrdff:ff:ff:ff:ff:ffinet72.98.2.10/16scopeglobaleth2inet6fe80::20c:29ff:fe00:902e/64scopelinkvalid_lftforeverpreferred_lftforever
[root@node2~]#route-nKernelIProutingtableDestinationGatewayGenmaskFlagsMetricRefUseIface110.1.24.00.0.0.0255.255.255.0U000eth172.98.0.00.0.0.0255.255.0.0U000eth210.1.0.00.0.0.0255.255.0.0U000eth0169.254.0.00.0.0.0255.255.0.0U100200eth0169.254.0.00.0.0.0255.255.0.0U100300eth1[root@node2~]#routeadd-net192.168.1.0/24gw110.1.24.10deveth1[root@node2~]#routeadd-net10.2.0.0/16gw72.98.70.20deveth2[root@node2~]#route-nKernelIProutingtableDestinationGatewayGenmaskFlagsMetricRefUseIface192.168.1.0110.1.24.10255.255.255.0UG000eth1110.1.24.00.0.0.0255.255.255.0U000eth110.2.0.072.98.70.20255.255.0.0UG000eth272.98.0.00.0.0.0255.255.0.0U000eth210.1.0.00.0.0.0255.255.0.0U000eth0169.254.0.00.0.0.0255.255.0.0U100200eth0169.254.0.00.0.0.0255.255.0.0U100300eth1
[root@node2~]#echo1 /proc/sys/net/ipv4/ip_forward
note3用来做route3
[root@node3~]#ipaddradd72.98.70.20/16deveth1[root@node3~]#ipaddradd10.2.0.1/16deveth2[root@node3~]#ipaddrshowdeveth13:eth1: BROADCAST,MULTICAST,UP,LOWER_UP mtu1500qdiscpfifo_faststateUPqlen1000link/ether00:0c:29:47:d8:e1brdff:ff:ff:ff:ff:ffinet72.98.70.20/16scopeglobaleth1inet6fe80::20c:29ff:fe47:d8e1/64scopelinkvalid_lftforeverpreferred_lftforever[root@node3~]#ipaddrshowdeveth24:eth2: BROADCAST,MULTICAST,UP,LOWER_UP mtu1500qdiscpfifo_faststateUPqlen1000link/ether00:0c:29:47:d8:ebbrdff:ff:ff:ff:ff:ffinet10.2.0.1/16scopeglobaleth2inet6fe80::20c:29ff:fe47:d8eb/64scopelinkvalid_lftforeverpreferred_lftforever
[root@node3~]#routeadd-net110.1.24.0/24gw72.98.2.10deveth1[root@node3~]#route-nKernelIProutingtableDestinationGatewayGenmaskFlagsMetricRefUseIface192.168.1.072.98.2.10255.255.255.0UG000eth1110.1.24.072.98.2.10255.255.255.0UG000eth110.2.0.00.0.0.0255.255.0.0U000eth272.98.0.00.0.0.0255.255.0.0U000eth110.1.0.00.0.0.0255.255.0.0U000eth0169.254.0.00.0.0.0255.255.0.0U100200eth0169.254.0.00.0.0.0255.255.0.0U100300eth1
[root@node3~]#echo1 /proc/sys/net/ipv4/ip_forward
主机A:
[root@host1~]#ipaddradd192.168.1.100/24deveno33554984[root@host1~]#iprouteadddefaultvia192.168.1.1[root@host1~]#ipaddrshowdeveno335549843:eno33554984: BROADCAST,MULTICAST,UP,LOWER_UP mtu1500qdiscpfifo_faststateUPqlen1000link/ether00:0c:29:2b:82:a6brdff:ff:ff:ff:ff:ffinet192.168.1.100/24scopeglobaleno33554984valid_lftforeverpreferred_lftforeverinet6fe80::20c:29ff:fe2b:82a6/64scopelinkvalid_lftforeverpreferred_lftforever[root@host1~]#route-n-bash:route:commandnotfound[root@host1~]#iprouteshow10.1.0.0/16deveno16777736protokernelscopelinksrc10.1.70.171metric100192.168.1.0/24deveno33554984protokernelscopelinksrc192.168.1.1000.0.0.0via192.168.1.1deveno33554984
主机B:
[root@host2~]#ipaddrshowdeveno335549843:eno33554984: BROADCAST,MULTICAST,UP,LOWER_UP mtu1500qdiscpfifo_faststateUPqlen1000link/ether00:0c:29:aa:22:47brdff:ff:ff:ff:ff:ffinet192.168.1.63/24scopeglobaleno33554984valid_lftforeverpreferred_lftforeverinet6fe80::20c:29ff:feaa:2247/64scopelinkvalid_lftforeverpreferred_lftforever[root@host2~]#route-nKernelIProutingtableDestinationGatewayGenmaskFlagsMetricRefUseIface10.1.0.00.0.0.0255.255.0.0U10000eno16777736192.168.1.00.0.0.0255.255.255.0U000eno335549840.0.0.0192.168.1.1255.255.255.255UGH000eno33554984
主机C:
root@debian:~#ipaddrshowdeveth13:eth1: BROADCAST,MULTICAST,UP,LOWER_UP mtu1500qdiscpfifo_faststateUPgroupdefaultqlen1000link/ether00:0c:29:f1:04:08brdff:ff:ff:ff:ff:ffinet10.2.110.100/16scopeglobaleth1valid_lftforeverpreferred_lftforeverroot@debian:~#route-nKernelIProutingtableDestinationGatewayGenmaskFlagsMetricRefUseIface10.1.0.00.0.0.0255.255.0.0U000eth010.2.0.00.0.0.0255.255.0.0U000eth10.0.0.010.2.0.1255.255.255.255UGH000eth1root@debian:~#
至此所有配置已经结束,关闭所有主机的网关和selinux
测试:
在主机C上:
root@debian:~#ping-Ieth1192.168.1.1PING192.168.1.1(192.168.1.1)from10.2.110.100eth1:56(84)bytesofdata.64bytesfrom192.168.1.1:icmp_seq=1ttl=62time=0.691ms64bytesfrom192.168.1.1:icmp_seq=2ttl=62time=1.17ms---192.168.1.1pingstatistics---2packetstransmitted,2received,0%packetloss,time1000msrttmin/avg/max/mdev=0.691/0.931/1.171/0.240msroot@debian:~#ping-Ieth1192.168.1.63PING192.168.1.63(192.168.1.63)from10.2.110.100eth1:56(84)bytesofdata.64bytesfrom192.168.1.63:icmp_seq=1ttl=61time=1.22ms64bytesfrom192.168.1.63:icmp_seq=2ttl=61time=0.927ms---192.168.1.63pingstatistics---2packetstransmitted,2received,0%packetloss,time1001msrttmin/avg/max/mdev=0.927/1.074/1.221/0.147msroot@debian:~#ping-Ieth1192.168.1.100PING192.168.1.100(192.168.1.100)from10.2.110.100eth1:56(84)bytesofdata.64bytesfrom192.168.1.100:icmp_seq=1ttl=61time=1.21ms64bytesfrom192.168.1.100:icmp_seq=2ttl=61time=1.78ms---192.168.1.100pingstatistics---2packetstransmitted,2received,0%packetloss,time1001msrttmin/avg/max/mdev=1.214/1.497/1.780/0.283msroot@debian:~#
在主机A上:
[root@host1~]#ping-Ieno3355498410.2.110.100PING10.2.110.100(10.2.110.100)from192.168.1.100eno33554984:56(84)bytesofdata.64bytesfrom10.2.110.100:icmp_seq=1ttl=61time=0.985ms64bytesfrom10.2.110.100:icmp_seq=2ttl=61time=1.09ms64bytesfrom10.2.110.100:icmp_seq=3ttl=61time=1.89ms64bytesfrom10.2.110.100:icmp_seq=4ttl=61time=2.00ms---10.2.110.100pingstatistics---4packetstransmitted,4received,0%packetloss,time3005msrttmin/avg/max/mdev=0.985/1.493/2.008/0.459ms[root@host1~]#
在主机B上:
[root@host2~]#ping-Ieno3355498410.2.110.100PING10.2.110.100(10.2.110.100)from192.168.1.63eno33554984:56(84)bytesofdata.64bytesfrom10.2.110.100:icmp_seq=1ttl=61time=1.15ms64bytesfrom10.2.110.100:icmp_seq=2ttl=61time=1.93ms64bytesfrom10.2.110.100:icmp_seq=3ttl=61time=0.979ms---10.2.110.100pingstatistics---3packetstransmitted,3received,0%packetloss,time2003msrttmin/avg/max/mdev=0.979/1.355/1.930/0.412ms
[root@host2~]#ping-Ieno3355498472.98.70.20PING72.98.70.20(72.98.70.20)from192.168.1.63eno33554984:56(84)bytesofdata.64bytesfrom72.98.70.20:icmp_seq=1ttl=62time=0.751ms64bytesfrom72.98.70.20:icmp_seq=2ttl=62time=0.807ms64bytesfrom72.98.70.20:icmp_seq=3ttl=62time=1.33ms---72.98.70.20pingstatistics---3packetstransmitted,3received,0%packetloss,time2000msrttmin/avg/max/mdev=0.751/0.964/1.335/0.264ms[root@host2~]#ping-Ieno3355498472.98.70.10###不知道为啥ping不通PING72.98.70.10(72.98.70.10)from192.168.1.63eno33554984:56(84)bytesofdata.From110.1.24.20icmp_seq=1DestinationHostUnreachableFrom110.1.24.20icmp_seq=2DestinationHostUnreachableFrom110.1.24.20icmp_seq=3DestinationHostUnreachable---72.98.70.10pingstatistics---5packetstransmitted,0received,+3errors,100%packetloss,time4002mspipe4[root@host2~]#ping-Ieno33554984110.1.24.20PING110.1.24.20(110.1.24.20)from192.168.1.63eno33554984:56(84)bytesofdata.64bytesfrom110.1.24.20:icmp_seq=1ttl=63time=0.556ms64bytesfrom110.1.24.20:icmp_seq=2ttl=63time=2.15ms64bytesfrom110.1.24.20:icmp_seq=3ttl=63time=0.972ms---110.1.24.20pingstatistics---3packetstransmitted,3received,0%packetloss,time2002msrttmin/avg/max/mdev=0.556/1.228/2.157/0.678ms[root@host2~]#ping-Ieno33554984110.1.24.10PING110.1.24.10(110.1.24.10)from192.168.1.63eno33554984:56(84)bytesofdata.64bytesfrom110.1.24.10:icmp_seq=1ttl=64time=0.282ms64bytesfrom110.1.24.10:icmp_seq=2ttl=64time=0.598ms64bytesfrom110.1.24.10:icmp_seq=3ttl=64time=0.367ms---110.1.24.10pingstatistics---3packetstransmitted,3received,0%packetloss,time2000msrttmin/avg/max/mdev=0.282/0.415/0.598/0.135ms[root@host2~]#
2.centos7的网络组实现
网络组类似于centos6的bond,都是多个网卡使用一个IP,是增强网络健壮性的一个手段
网络组：是将多个网卡聚合在一起方法，从而实现冗错和提高吞吐量
网络组不同于旧版中bonding技术，提供更好的性能和扩展性
网络组由内核驱动和teamd守护进程实现.包名是teamd
启动网络组接口不会自动启动网络组中的port接口启动网络组接口中的port接口不会自动启动网络组接口禁用网络组接口会自动禁用网络组中的port接口没有port接口的网络组接口可以启动静态IP连接启用DHCP连接时，没有port接口的网络组会等待port接口的加入
具体的runner方式可以查看man 5 teamd.conf帮助
创建网络组接口:
[root@linux~]#nmcliconaddtypeteamcon-nametestifnameteam0config'{"runner":{"name":"activebackup"}}'Connection'test'(5a3bfb26-993f-45ad-add6-246ff419e7bd)successfullyadded.
此时在网络配置目录下生成了一个文件
[root@linux~]#ls/etc/sysconfig/network-scripts/ifcfg-test/etc/sysconfig/network-scripts/ifcfg-test
[root@linux~]#nmclidevshowteam0GENERAL.DEVICE:team0GENERAL.TYPE:teamGENERAL.HWADDR:82:D0:69:2C:48:6EGENERAL.MTU:1500GENERAL.STATE:70(connecting(gettingIPconfiguration))GENERAL.CONNECTION:testGENERAL.CON-PATH:/org/freedesktop/NetworkManager/ActiveConnection/3[root@linux~]#nmcliconshowNAMEUUIDTYPEDEVICEeno33554984fb67dbad-ec81-39b4-42b1-ebf975c3ff13802-3-etherneteno33554984eno16777736d329fbf7-4423-4a10-b097-20b266c26768802-3-etherneteno16777736eno50332208d2665055-8e83-58f1-e9e3-49a5fb133641802-3-etherneteno50332208test5a3bfb26-993f-45ad-add6-246ff419e7bdteamteam0
给team0设置静态IP和开机自启动
[root@linux~]#nmcliconmodtestipv4.methodmanualipv4.addresses"10.1.70.24/16"connection.autoconnectyes[root@linux~]#cat/etc/sysconfig/network-scripts/ifcfg-testDEVICE=team0TEAM_CONFIG="{\"runner\":{\"name\":\"activebackup\"}}"DEVICETYPE=TeamBOOTPROTO=noneDEFROUTE=yesIPV4_FAILURE_FATAL=noIPV6INIT=yesIPV6_AUTOCONF=yesIPV6_DEFROUTE=yesIPV6_FAILURE_FATAL=noNAME=testUUID=5a3bfb26-993f-45ad-add6-246ff419e7bdONBOOT=yesIPADDR=10.1.70.24PREFIX=16IPV6_PEERDNS=yesIPV6_PEERROUTES=yes[root@linux~]#
创建两个port接口
[root@linux~]#nmcliconaddtypeteam-slavecon-nametest-1ifnameeno33554984masterteam0Connection'test-1'(234c3e91-d90d-421c-ae88-133deddfce94)successfullyadded.[root@linux~]#nmcliconaddtypeteam-slavecon-nametest-2ifnameeno50332208masterteam0Connection'test-2'(116ef596-d983-456c-a6ae-a74a4f8c03dc)successfullyadded.[root@linux~]#
[root@linux~]#cat/etc/sysconfig/network-scripts/ifcfg-test-1NAME=test-1UUID=234c3e91-d90d-421c-ae88-133deddfce94DEVICE=eno33554984ONBOOT=yesTEAM_MASTER=team0DEVICETYPE=TeamPort[root@linux~]#cat/etc/sysconfig/network-scripts/ifcfg-test-2NAME=test-2UUID=116ef596-d983-456c-a6ae-a74a4f8c03dcDEVICE=eno50332208ONBOOT=yesTEAM_MASTER=team0DEVICETYPE=TeamPort
查看网络组状态:
[root@linux~]#teamdctlteam0statsetup:runner:activebackuprunner:activeport:
发现port端口均没有开启
开启port端口
[root@linux~]#nmcliconuptest-1Connectionsuccessfullyactivated(D-Busactivepath:/org/freedesktop/NetworkManager/ActiveConnection/5)[root@linux~]#nmcliconuptest-2Connectionsuccessfullyactivated(D-Busactivepath:/org/freedesktop/NetworkManager/ActiveConnection/7)
[root@linux~]#teamdctlteam0statsetup:runner:activebackupports:eno33554984linkwatches:linksummary:upinstance[link_watch_0]:name:ethtoollink:updowncount:0eno50332208linkwatches:linksummary:upinstance[link_watch_0]:name:ethtoollink:updowncount:0runner:activeport:eno33554984
可以看到端口开启成功
[root@linux~]#ping-Iteam010.1.70.172PING10.1.70.172(10.1.70.172)from10.1.70.24team0:56(84)bytesofdata.64bytesfrom10.1.70.172:icmp_seq=1ttl=64time=0.500ms64bytesfrom10.1.70.172:icmp_seq=2ttl=64time=0.804ms---10.1.70.172pingstatistics---2packetstransmitted,2received,0%packetloss,time1001msrttmin/avg/max/mdev=0.500/0.652/0.804/0.152ms[root@linux~]#
配置成功,可以看到当前活动的是eno33554984,测试禁用后能否成功
[root@linux~]#nmclidevicedisconnecteno33554984Device'eno33554984'successfullydisconnected.[root@linux~]#ping-Iteam010.1.70.172PING10.1.70.172(10.1.70.172)from10.1.70.24team0:56(84)bytesofdata.测试不成功,通过查找资料了解到当使用activebackup的runner时,必须加上一个参数
[root@linux~]#nmcliconmodifytestteam.config'{"runner":{"name":"activebackup","hwaddr_policy":"by_active"}}'[root@linux~]#cat/etc/sysconfig/network-scripts/ifcfg-testDEVICE=team0TEAM_CONFIG="{\"runner\":{\"name\":\"activebackup\",\"hwaddr_policy\":\"by_active\"}}"DEVICETYPE=TeamBOOTPROTO=noneDEFROUTE=yesIPV4_FAILURE_FATAL=noIPV6INIT=yesIPV6_AUTOCONF=yesIPV6_DEFROUTE=yesIPV6_FAILURE_FATAL=noNAME=testUUID=5a3bfb26-993f-45ad-add6-246ff419e7bdONBOOT=yesIPADDR=10.1.70.24PREFIX=16IPV6_PEERDNS=yesIPV6_PEERROUTES=yes
详情参考:http://183530300.blog.51cto.com/894387/1851388
『 不可能 』只存在於蠢人的字典里
