SP V1 3.3 Intra AS vpnv4 解法对么？能给个V1的3.3 Intra AS vpnv4 的正确解法么？
本帖最后由 百花开的 于 2015-3-24 00:12 编辑能给个V1的3.3 Intra AS vpnv4 的解法么？
我的解法如下：
R14#show run | s r b
router bgp 112
bgp router-id 172.9.0.3
bgp log-neighbor-changes
no bgp default ipv4-unicast
neighbor 172.9.143.3 remote-as 9
!
address-family ipv4
redistribute ospf 112
neighbor 172.9.143.3 activate
neighbor 172.9.143.3 send-label
exit-address-family
R14#show run | s r ospf
router ospf 112
router-id 172.9.0.14
redistribute bgp 112 subnets
network 172.9.0.14 0.0.0.0 area 0
network 172.9.145.14 0.0.0.0 area 0
distance ospf intra-area 10
R15# show run | s r ospf
router ospf 112
router-id 172.9.0.15
network 172.9.0.15 0.0.0.0 area 0
network 172.9.145.15 0.0.0.0 area 0
network 172.9.157.15 0.0.0.0 area 0
R7#show run | s r b
router bgp 9
bgp log-neighbor-changes
no bgp default ipv4-unicast
!
address-family ipv4 vrf ABC
network 172.9.0.7 mask 255.255.255.255
redistribute ospf 112
exit-address-family
R7#
R7#show run | s r ospf
router ospf 112 vrf ABC
router-id 172.9.0.7
domain-tag 9
redistribute bgp 9 subnets
network 172.9.0.7 0.0.0.0 area 0
network 172.9.157.7 0.0.0.0 area 0
RP/0/0/CPU0:XR3#show running-config router bgp
Mon Mar 23 15:41:07.258 UTC
router bgp 9
vrf ABC
rd 9:9
address-family ipv4 unicast
network 172.9.0.3/32
allocate-label all
!
address-family ipv6 unicast
network 2002:172:9::3/128
!
neighbor 172.9.143.14
remote-as 112
address-family ipv4 unicast
route-policy R14R3ipv4 in
route-policy R3R14ipv4 out
as-override
!
address-family ipv4 labeled-unicast
route-policy R14R3label in
route-policy R3R14label out
as-override
!
!
!
!
RP/0/0/CPU0:XR4(config)#show running-config router bgp
Mon Mar 23 15:42:00.775 UTC
router bgp 9
vrf ABC
rd 9:9
address-family ipv4 unicast
network 172.9.0.4/32
allocate-label all
!
neighbor 172.9.114.11
remote-as 112
address-family ipv4 unicast
route-policy R11R4ipv4 in
route-policy R4R11ipv4 out
as-override
!
address-family ipv4 labeled-unicast
route-policy R11R4label in
route-policy R4R11label out
as-override
!
!
!
!
R8#show run | s r b
router bgp 9
bgp log-neighbor-changes
no bgp default ipv4-unicast
address-family ipv4 vrf ABC
network 172.9.0.8 mask 255.255.255.255
neighbor 172.9.128.12 remote-as 112
neighbor 172.9.128.12 activate
neighbor 172.9.128.12 as-override
neighbor 172.9.128.12 send-label
exit-address-family
!
R11#show run | s r b
router bgp 112
bgp log-neighbor-changes
no bgp default ipv4-unicast
neighbor 172.9.114.4 remote-as 9
!
address-family ipv4
network 172.9.114.0 mask 255.255.255.0
redistribute ospf 112
neighbor 172.9.114.4 activate
neighbor 172.9.114.4 send-label
exit-address-family
R11#
R11#show run | s r ospf
router ospf 112
router-id 172.9.0.11
redistribute bgp 112 subnets
network 172.9.0.11 0.0.0.0 area 0
network 172.9.112.11 0.0.0.0 area 0
distance ospf intra-area 10
R12#show run | s r b
router bgp 112
bgp log-neighbor-changes
neighbor 172.9.128.8 remote-as 9
!
address-family ipv4
redistribute ospf 112
neighbor 172.9.128.8 activate
neighbor 172.9.128.8 send-label
exit-address-family
R12#
R12#show run | s r ospf
router ospf 112
router-id 172.9.0.12
redistribute bgp 112 subnets
network 172.9.0.12 0.0.0.0 area 0
network 172.9.112.12 0.0.0.0 area 0
distance ospf intra-area 10
我有如下问题：
1为什么只有BGP中宣告了R4 R11的这个网段 network 172.9.114.0 mask 255.255.255.0，而R12 R8 与R3 R14的直连网段没有在BGP中通告！
2 R4和R8 上为什么没有用SOO来防止从一个CER11学到的路由通过SP又通告给了CER12？
3 site2的service provider 明确说了是R3为什么在R7上面确做了双向重分布？
4，根据这个解法，发现有大量的次优路由产生
5，请教一个正确的解法！
帮顶，关注楼下回复。 这个解法是site2 上做双点双向重分布，理论上更加好，双出口，但是现在流行只在R7上做单点双向.你说的次优路径应该是在R7那边吧。
还有bgp network的时候呢，都是宣告换回口的，不要把直连接口给通告。 本帖最后由 百花开的 于 2015-3-24 10:02 编辑
白'开水℡ 发表于 2015-3-24 09:36
这个解法是site2 上做双点双向重分布，理论上更加好，双出口，但是现在流行只在R7上做单点双向.你说的次优 ...
你的意思所有直连到SP的网段都不通告？
因为R7上面打了domain tag 9R15访问site1是个负载均衡的情况！R15选择了R14通住site1的话这就是次优路由！site1 也没有通过SOO属性来防环，这种从一个CE学到的路由又能传到了本自治系统内的另一个CE？
R7的domain tag是为了防止R7收到R15过来的ospf路由覆盖掉原来的bgp路由，你既然做了R7的双向重分布的话你应该是R15选择R7才对，或者你的双向重分布没有成功，试试在R7上重分布禁ospf的时候把mertic-type 改成1 白'开水℡ 发表于 2015-3-24 10:05
R7的domain tag是为了防止R7收到R15过来的ospf路由覆盖掉原来的bgp路由，你既然做了R7的双向重分布的话你应 ...
R14#show run | s r ospf
router ospf 112
router-id 172.9.0.14
redistribute bgp 112 subnets
network 172.9.0.14 0.0.0.0 area 0
network 172.9.145.14 0.0.0.0 area 0
distance ospf intra-area 10
R7#show run | s r ospf
router ospf 112 vrf ABC
router-id 172.9.0.7
domain-tag 9
redistribute bgp 9 subnets
network 172.9.0.7 0.0.0.0 area 0
network 172.9.157.7 0.0.0.0 area 0
R14和R7同时将BGP重分布进入OSPF ，R15学到就是个负载均衡的情况！如果R7上改成OE1肯定会选择R7做为出口了，OE1》OE2，但是总感觉这部分解法和版本的需求不相适合！解法太让人纠结了，虽然也是过人的！ 路过了解一下 百花开的 发表于 2015-3-24 10:20
R14#show run | s r ospf
router ospf 112
router-id 172.9.0.14
所以现在解法流行只在R14上单点双向啊 白'开水℡ 发表于 2015-3-24 10:25
所以现在解法流行只在R14上单点双向啊
只有R14上面做重分布这个能过人的吧！:loveliness: 也有点纠结这个，不过第一条问题是考题要求，直连路径不能被宣告。 关于次优路径的问题，我测试了下
R14#show run | s r ospf
router ospf 112
router-id 172.9.0.14
redistribute bgp 112 subnets metric 10000
network 172.9.0.14 0.0.0.0 area 0
network 172.9.145.14 0.0.0.0 area 0
这样不会负载均衡，但是不确定这样做对不对，因为去往R3最优化应该能负载均衡。
考试求过吧，不敢随心敲。 谢谢分享！顶起！
页:
[1]