Is anyone working on Routing and switching Vol 1 task 7.34 of BGP as -set ?

Is anyone working on Routing and switching Vol 1 task 7.34 of BGP as -set ?

 

Question :- R4 will receive 112.0.0.0/8 from BB3 and 112.0.0.0/5 from R1.And sending both os these to routes to R5.

my question is why he is sending both routes to R5 .....he (R4) has ro select best route and send only best route to R5.

i.e. R4 should send only 112.0.0.0/8 to R5.

Please correct my understanding.

 

Regards

Ganpat

 

Comments

  • my question is why he is sending both routes to R5 .....he (R4) has ro select best route and send only best route to R5.

    i.e. R4 should send only 112.0.0.0/8 to R5.

    These two routes are not same so bgp will not run its best path selection algo, it will take both the routes in the bgp table if these routes appear as the best routes in the bgp table, it will advertise them to its peers.

     

     

  • Thanks for your reply!!!

    Can you please let me know in what parameter these routes are not same.

    R4 is getting 112.0.0.0/8 and 112.0.0.0/5

     

    /Ganpat

  • Can you please let me know in what parameter these routes are not same.

    R4 is getting 112.0.0.0/8 and 112.0.0.0/5

    Longest match criteria. One has /8 mask and other has /5

     

  • To BGP these are totally different prefixes due to the different mask (/5 vs /8). If you look into R4's BGP table both routes will be listed as best and both routes will be installed into the routing table:


    Rack1R4(config-router)# do sh ip bgp 112.0.0.0/8

    BGP routing table entry for 112.0.0.0/8, version 20

    Paths: (1 available, best #1, table default)

      Advertised to update-groups:

         1          3

      54

        204.12.1.254 from 204.12.1.254 (31.3.0.1)

          Origin incomplete, localpref 100, valid, external, best



    Rack1R4(config-router)# do sh ip bgp 112.0.0.0/5

    BGP routing table entry for 112.0.0.0/5, version 37

    Paths: (1 available, best #1, table default)

      Advertised to update-groups:

         1

      54, (aggregated by 100 150.1.1.1)

        155.1.146.1 from 155.1.146.1 (150.1.1.1)

          Origin incomplete, metric 0, localpref 100, valid, internal, best




    Rack1R4(config-router)#do sho ip route | i 112

    B     112.0.0.0/5 [200/0] via 155.1.146.1, 00:04:46

    B     112.0.0.0/8 [20/0] via 204.12.1.254, 00:04:46


    Therefore, both routes will be advertised to R5.


    HTH

    Good luck!

  • so as these both routes are BEST routes marked in BGP table they considered as differnent...is it ?

    /Ganpat

     

  • so as these both routes are BEST routes marked in BGP table they considered as differnent...is it ?

    Hi ganpatspatil,

    /5 and /8 are different routes and both are marked as BEST route ( see on

    qqabdal's example) that's why installed into routing table). But when host sends the traffic towards destination, longest match criteria is performed.

    Good Luck..


  • Narayan is correct. To BGP two prefixes with different masks are different routes, for IGPs this is not true. So BGP will advertise all the routes and the far end router will decide which one it will use, the longest match. IGPs on the other hand, it will only advertise the longest match route, because only the longest match route makes its way from the database to the routing table and it is candidate to be advertised.

    HTH

  • IGPs on the other hand, it will only advertise the longest match route, because only the longest match route makes its way from the database to the routing table and it is candidate to be advertised.

    If i am understanding your statement correctly, then i disagree with this statement. Longest match criteria comes into picture when router need to forward a packet at layer3. But if a router has two different routes in its database both will make its way to the routing table and hence are candidate to be advertise.

  •  

    Hi dcancerian,

    I am not sure if I make it clear. But here is what I meant:

    Let's say EIGRP - the router will install only the longest match route into the routing table, the /5 in this case will not make its way to the routing table because it was learned from a different routing protocol. It is true that if the /8 is gone, then it will install the /5, but in the first place only the /8 will be installed into the routing table. Moreover, with EIGRP only the best routes will be installed, so EIGRP will only advertise the route that gets installed into its routing table. Therefore, the /8 will be installed into the RT and consequently advertised on. I meant on a multi routing protocol scenario.

    Hope this clarifies.

    Good luck!

     

  • Hi qabdal ,

    I have following senario

    R1----------R4---------BB2
                     |
                     |
                     |
                   R5 

    In this case R1 is sending one summary route 22.0.0.0/8 to R4 by EIGRP and BB2 is sending one summary route 22.0.0.0/9 to R4.All configured with EIGRP AS 100.

    R1#sh running-config interface f0/0
    Building configuration...

    Current configuration : 146 bytes
    !
    interface FastEthernet0/0
     ip address 14.14.14.1 255.255.255.0
     ip summary-address eigrp 100 22.0.0.0 255.0.0.0 5
     duplex auto
     speed auto
    end

    BB2#sh running-config interface f0/0
    Building configuration...

    Current configuration : 150 bytes
    !
    interface FastEthernet0/0
     ip address 42.42.42.254 255.255.255.0
     ip summary-address eigrp 100 22.0.0.0 255.128.0.0 5
     duplex auto
     speed auto
    end

    On R4 --in routing table both routes are installed.

    R4#sh ip route
         22.0.0.0/8 is variably subnetted, 2 subnets, 2 masks
    D       22.0.0.0/9 [90/409600] via 42.42.42.254, 00:10:24, FastEthernet0/1
    D       22.0.0.0/8 [90/460800] via 14.14.14.1, 00:08:59, FastEthernet0/0

    And R5---in Routing table both routes are installed

    R5#sh ip route
    Codes: C - connected, S - static, R - RIP, M - mobile, B - BGP
    D       22.0.0.0/9 [90/435200] via 45.45.45.4, 00:11:46, FastEthernet0/0
    D       22.0.0.0/8 [90/486400] via 45.45.45.4, 00:09:42, FastEthernet0/0

    SO my conclusion is both routes are advertised by IGP.

    Route selection will be done only when actual traffic need to send.

    Same behavior in IGP and BGP.

    please correct if requried.

    /Ganpat

     

     

     

  • I have following senario

    R1----------R4---------BB2
                     |
                     |
                     |
                   R5 

    In this case R1 is sending one summary route 22.0.0.0/8 to R4 by EIGRP and BB2 is sending one summary route 22.0.0.0/9 to R4.All configured with EIGRP AS 100.

    R1#sh running-config interface f0/0
    Building configuration...

    Current configuration : 146 bytes
    !
    interface FastEthernet0/0
     ip address 14.14.14.1 255.255.255.0
     ip summary-address eigrp 100 22.0.0.0 255.0.0.0 5
     duplex auto
     speed auto
    end

    BB2#sh running-config interface f0/0
    Building configuration...

    Current configuration : 150 bytes
    !
    interface FastEthernet0/0
     ip address 42.42.42.254 255.255.255.0
     ip summary-address eigrp 100 22.0.0.0 255.128.0.0 5
     duplex auto
     speed auto
    end

    On R4 --in routing table both routes are installed.

    R4#sh ip route
         22.0.0.0/8 is variably subnetted, 2 subnets, 2 masks
    D       22.0.0.0/9 [90/409600] via 42.42.42.254, 00:10:24, FastEthernet0/1
    D       22.0.0.0/8 [90/460800] via 14.14.14.1, 00:08:59, FastEthernet0/0

    And R5---in Routing table both routes are installed

    R5#sh ip route
    Codes: C - connected, S - static, R - RIP, M - mobile, B - BGP
    D       22.0.0.0/9 [90/435200] via 45.45.45.4, 00:11:46, FastEthernet0/0
    D       22.0.0.0/8 [90/486400] via 45.45.45.4, 00:09:42, FastEthernet0/0

    SO my conclusion is both routes are advertised by IGP.

    Route selection will be done only when actual traffic need to send.

    Same behavior in IGP and BGP.

    Thats what I meant in my last post.

Sign In or Register to comment.