OSPF-NBMA-nonBroadcast mode

arsalan921

Hello guys,

i am configuring OSPF in an NBMA topology, as you guys know there are 5 modes in this,
non-broadcast,broadcast,ptp,ptm and ptm non-broadcast.

i have started with non-broadcast, and would like to get some help in its configuration, and would also appreciate if someone can give a link where the configuration of these 5 modes of operation is shown.

so i am using GNS3 simulator for the configuration,
total 4 routers used, using one router as a frame-relay switch, other 3 are for communication
R1= hub
R2 and R3 spokes

dlci values used for R1-R2 , 102 and 201
dlci values used for R1-R3 , 103 and 301.

now configured frame-relay switch like this....
S 1/1 attached to hub, and
S1/2 to R2 and
S1/3 to R3.


interface Serial1/1
no ip address
encapsulation frame-relay
serial restart-delay 0
frame-relay intf-type dce
frame-relay route 102 interface Serial1/2 201
frame-relay route 103 interface Serial1/3 301
!
interface Serial1/2
no ip address
encapsulation frame-relay
serial restart-delay 0
frame-relay route 201 interface Serial1/1 102
!
interface Serial1/3
no ip address
encapsulation frame-relay
serial restart-delay 0
frame-relay intf-type dce
frame-relay route 301 interface Serial1/1 103
!

now i have configured hub (R1) and 2 spokes ( but the links are down ). can anyone please tell me how to configure hub and spoke ?

i have used one ip subnet.
10.1.1.1 for hub
10.1.1.2 for R2
10.1.1.3 for R3

I have used neighbor command on R1 and statically made R2 and R3 neighbors of R1.
how would dlci values be mapped ?
frame-relay interface-dlci command or frame-relay map ip command ?

pls help

thank you

yuriz43

Why are you using a router as the frame relay switch? You can just use the one in GNS3.

BTW:

frame-relay interface-dlci is for mapping a dlci to a sub-interface

frame-relay map is for creating a static DLCI to IP mapping. ( no inverse-arp )

billscott92787

yuriz43 wrote: »

Why are you using a router as the frame relay switch? You can just use the one in GNS3.

BTW:

frame-relay interface-dlci is for mapping a dlci to a sub-interface

frame-relay map is for creating a static DLCI to IP mapping. ( no inverse-arp )

I've done what he is doing in dynamips. There is nothing wrong with doing it.





You said that dlci values used for R1-R2 , 102 and 201
dlci values used for R1-R3 , 103 and 301.

now configured frame-relay switch like this....
S 1/1 attached to hub, and
S1/2 to R2 and
S1/3 to R3.




So on your switch have you configured the frame-relay route command to configure the routes to the DLCIs. If your using a router you'll have to do this. As far as the local DLCI map statements on R1 you would do:


frame-relay map ip x.x.x.x 102 broadcast (IP address of R3)
frame-relay map ip x.x.x.x 103 broadcast (IP address of R3)

and vice versa. On R2 you would make the IP address of R1 with the DLCI 201.

frame-relay map ip x.x.x.x 201 broadcast (IP address of R1)
frame-relay map ip x.x.x.x 201 broadcast (IP address of R3)

R3:

frame-relay map ip x.x.x.x 301 broadcast (IP address of R1)
frame-relay map ip x.x.x.x 301 broadcast (IP address of R2)



On the router acting as the frame-relay switch:

S 1/1: frame-relay route 102 int s1/2 201
frame-relay route 103 int s1/3 301

S1/2 to R2: frame-relay route 201 int s1/1 102

S1/3 to R3. frame-relay route 301 int s1/1 103

ip ospf priority 0. So that they do not participate in the OSPF election and the Hub becomes the DR everytime. hope this helps you out.

arsalan921

thank you both of you for your replies
it has worked.
thanks billscott for a very detail reply, i have been able to configure now.

and
best of luck for your bsci !!

one thing which i would like to ask is that which labbook are you following for bsci exam ? which version of it ?
the one i have is version5.0 (copyright 2006). is there any latest version available for this ? or any other lab book for BSCI ?

arsalan921

i have done the configuration of a non-broadcast mode.

now the hub (R1) is able to see the two spokes as neighbors ( R2 and R3). R2 and R3 cannot see eacother as neighbor ( becz of hub and spoke topology) .

now i have one more question after this configuration

i have created 3 loopback interfaces .

loopback 1 on R1 (HUB)
loopback 2 on R2
loopback 3 on R3

i have advertised these also in ospf configuration

network 1.1.1.1 0.0.0.0 area 0
network 2.2.2.2 0.0.0.0 area 0
network 3.3.3.3 0.0.0.0 area 0


now when i use show ip route commands on spokes , r2 for example here

R2#show ip route ospf
1.0.0.0/24 is subnetted, 1 subnets
O 1.1.1.0 [110/65] via 10.1.1.1, 00:30:29, Serial1/2
3.0.0.0/24 is subnetted, 1 subnets
O 3.3.3.0 [110/65] via 10.1.1.3, 00:30:29, Serial1/2

cost to reach Loopback-1 and Loopback-3 is same = 65
Loopback 3 reachable thru via 10.1.1.3

and when i traceroute 3.3.3.3

R2#traceroute 3.3.3.3

Tracing the route to 3.3.3.3

1 10.1.1.1 244 msec 28 msec 184 msec
2 10.1.1.3 184 msec 200 msec *


it shows that it goes first to R1 and R3 (and this is how it should operate).
but why is cost the same for reaching both loopbacks (65) and why is it shown in routing table that next hop for 3.3.3.3 is 10.1.1.3, it should be 10.1.1.1


please tell what mistake i am making.

billscott92787

arsalan921 wrote: »

i have done the configuration of a non-broadcast mode.

now the hub (R1) is able to see the two spokes as neighbors ( R2 and R3). R2 and R3 cannot see eacother as neighbor ( becz of hub and spoke topology) .

now i have one more question after this configuration

i have created 3 loopback interfaces .

loopback 1 on R1 (HUB)
loopback 2 on R2
loopback 3 on R3

i have advertised these also in ospf configuration

network 1.1.1.1 0.0.0.0 area 0
network 2.2.2.2 0.0.0.0 area 0
network 3.3.3.3 0.0.0.0 area 0


now when i use show ip route commands on spokes , r2 for example here

R2#show ip route ospf
1.0.0.0/24 is subnetted, 1 subnets
O 1.1.1.0 [110/65] via 10.1.1.1, 00:30:29, Serial1/2
3.0.0.0/24 is subnetted, 1 subnets
O 3.3.3.0 [110/65] via 10.1.1.3, 00:30:29, Serial1/2

cost to reach Loopback-1 and Loopback-3 is same = 65
Loopback 3 reachable thru via 10.1.1.3

and when i traceroute 3.3.3.3

R2#traceroute 3.3.3.3

Tracing the route to 3.3.3.3

1 10.1.1.1 244 msec 28 msec 184 msec
2 10.1.1.3 184 msec 200 msec *


it shows that it goes first to R1 and R3 (and this is how it should operate).
but why is cost the same for reaching both loopbacks (65) and why is it shown in routing table that next hop for 3.3.3.3 is 10.1.1.3, it should be 10.1.1.1


please tell what mistake i am making.

Did you do these full mess from R2 to R3? Or does R2 have to go through R1 to get to R3 and R3 has to go through R1 to get to R2. If that's the case, then the cost is the same because it is the cost to get to the next hop.

arsalan921

R2 goes to R3 using R1
R3 goes to R2 using R1

hub-spoke topology

R2 goes to R3's loopback at cost of 65
R2 goes to R1's loopback at cost of 65

this is what i was asking ?
cost of R2 to reach R1's loopback of 65 is understandable.
but i think the cost for R2 to reach R3's loopback should be 129 ( 64 + 64 + 1 ). because 1st the packet would get out of R2's serial interface(64) and then it would get of R1's serial interface(64) and then get out of R3's loopback(1).

this is what i understand.

which lab book have you used for practicing bsci labs ?

thank you

billscott92787

arsalan921 wrote: »

R2 goes to R3 using R1
R3 goes to R2 using R1

hub-spoke topology

R2 goes to R3's loopback at cost of 65
R2 goes to R1's loopback at cost of 65

this is what i was asking ?
cost of R2 to reach R1's loopback of 65 is understandable.
but i think the cost for R2 to reach R3's loopback should be 129 ( 64 + 64 + 1 ). because 1st the packet would get out of R2's serial interface(64) and then it would get of R1's serial interface(64) and then get out of R3's loopback(1).

this is what i understand.

which lab book have you used for practicing bsci labs ?

thank you

I have used the Exam Certification Library. That is the cost because it's the cost to the "next hop" That's the easiest way for me to explain it.