利用BGP虚拟下一跳技术实现IP骨干网流量负载均衡VIP免费

配置说明：【R1】interfaceLoopback0ipaddress192.168.0.1255.255.255.255interfaceSerial1/0noshipaddress172.16.15.1255.255.255.0interfaceSerial1/1noshipaddress172.16.12.1255.255.255.0interfaceSerial1/2noshipaddress172.16.16.1255.255.255.0interfaceSerial1/3noshipaddress172.16.13.1255.255.255.0iproute10.0.0.1255.255.255.255172.16.13.3【虚拟下一跳地址指向真正下一跳，并由本域内IGP(OSPF)路由引入，使本域内所有邻居都能学习到虚拟下一跳地址的路由，从而使邻居从自己学习到的路由能最优化】iproute192.168.0.3255.255.255.255172.16.13.3ipprefix-listISP-Bseq5permit172.16.34.0/24【做前缀列表匹配需要改变下一跳的前缀】route-mapset-nexthoppermit10【做route-map，改相应前缀下一跳性】matchipaddressprefix-listISP-Bsetipnext-hop10.0.0.1route-mapset-nexthoppermit20routerospf1router-id192.168.0.1noauto-costredistributestaticmetric10metric-type1subnets【引入静态路由，使域内所有IBGP邻居能学到虚拟下一跳地址，从面使邻居从自己学习到的路由能最优化】network172.16.12.00.0.0.255area0network172.16.15.00.0.0.255area0network172.16.16.00.0.0.255area0network192.168.0.00.0.0.255area0routerbgp100nosynchronizationbgprouter-id192.168.0.1bgplog-neighbor-changesneighborinternalpeer-groupneighborinternalremote-as100neighborinternalupdate-sourceLoopback0neighborinternalroute-mapset-nexthopout【做出方向的ruout-map改变发向邻居路由的下一跳】neighbor192.168.0.2peer-groupinternalneighbor192.168.0.3remote-as200neighbor192.168.0.3ebgp-multihop255neighbor192.168.0.3update-sourceLoopback0neighbor192.168.0.3soft-reconfigurationinboundneighbor192.168.0.5peer-groupinternalneighbor192.168.0.6peer-groupinternalnoauto-summary【R2】interfaceLoopback0ipaddress192.168.0.2255.255.255.255interfaceSerial1/0noshipaddress172.16.26.2255.255.255.0interfaceSerial1/1noshipaddress172.16.12.2255.255.255.0interfaceSerial1/2noshipaddress172.16.25.2255.255.255.0interfaceSerial1/3noshipaddress172.16.24.2255.255.255.0iproute10.0.0.1255.255.255.255172.16.24.4【虚拟下一跳地址指向真正下一跳，并由本域内IGP(OSPF)路由引入，使本域内所有邻居都能学习到虚拟下一跳地址的路由，从而使邻居从自己学习到的路由能最优化】iproute192.168.0.4255.255.255.255172.16.24.4ipprefix-listISP-Bseq5permit172.16.34.0/24【做前缀列表匹配需要改变下一跳的前缀】route-mapset-nexthoppermit10【做route-map，改相应前缀下一跳性】matchipaddressprefix-listISP-Bsetipnext-hop10.0.0.1route-mapset-nexthoppermit20routerospf1router-id192.168.0.2log-adjacency-changesnoauto-costredistributestaticmetric10metric-type1subnets【引入静态路由，使域内所有IBGP邻居能学到虚拟下一跳地址，从面使邻居从自己学习到的路由能最优化】network172.16.12.00.0.0.255area0network172.16.25.00.0.0.255area0network172.16.26.00.0.0.255area0network192.168.0.00.0.0.255area0routerbgp100nosynchronizationbgprouter-id192.168.0.2bgplog-neighbor-changesneighborinternalpeer-groupneighborinternalremote-as100neighborinternalupdate-sourceLoopback0neighborinternalroute-mapset-nexthopout【做出方向的ruout-map改变发向邻居路由的下一跳】neighbor192.168.0.1peer-groupinternalneighbor192.168.0.4remote-as200neighbor192.168.0.4ebgp-multihop255neighbor192.168.0.4update-sourceLoopback0neighbor192.168.0.5peer-groupinternalneighbor192.168.0.6peer-groupinternalnoauto-summary【R3】由于R3、R4域内没有其它IBGP邻居，下挂设备都运行IGP所以不需要以虚拟下一跳的方式建立EBGP邻居，可以用正常的方式建立EBGP邻居即可interfaceLoopback0ipaddress192.168.0.3255.255.255.255interfaceSerial1/...