Need help with OSPF over hub/spoke frame-relay

polevector · July 2011

Please help!
I'm currently practicing with lab
So here's the topology screen:

http://www.polevector.com/screen1.jpg

The problem is that I can't get pings going between spokes, meaning r3 can't ping r4 or r4 can't ping r5. However pings between hub (r2) and spokes (r3/r4/r5) are successfull.

here's some useful output:

R2#sh ip ospf nei
Neighbor ID Pri State Dead Time Address Interface
3.3.3.3 0 FULL/DROTHER 00:01:49 172.12.30.3 Serial0/1
4.4.4.4 0 FULL/DROTHER 00:01:52 172.12.30.4 Serial0/1
5.5.5.5 0 FULL/DROTHER 00:01:42 172.12.30.5 Serial0/1
192.168.1.1 0 FULL/ - 00:00:30 192.168.1.1 Serial0/0
R2#

R5#sh ip ospf nei
Neighbor ID Pri State Dead Time Address Interface
2.2.2.2 1 FULL/DR 00:01:34 172.12.30.2 Serial0/0
R5#

R5#sh ip route ospf
2.0.0.0/32 is subnetted, 1 subnets
O IA 2.2.2.2 [110/65] via 172.12.30.2, 00:09:02, Serial0/0
3.0.0.0/32 is subnetted, 1 subnets
O IA 3.3.3.3 [110/65] via 172.12.30.3, 00:09:02, Serial0/0
4.0.0.0/32 is subnetted, 1 subnets
O IA 4.4.4.4 [110/65] via 172.12.30.4, 00:09:02, Serial0/0
192.168.1.0/30 is subnetted, 1 subnets
O IA 192.168.1.0 [110/128] via 172.12.30.2, 00:09:02, Serial0/0
O*E2 0.0.0.0/0 [110/1] via 172.12.30.2, 00:09:02, Serial0/0
R5#

Here's related configs:

R2:

!-ommited-!
interface Serial0/1
ip address 172.12.30.2 255.255.255.0
encapsulation frame-relay
ip split-horizon
serial restart-delay 0
frame-relay map ip 172.12.30.3 201 broadcast
frame-relay map ip 172.12.30.4 202 broadcast
frame-relay map ip 172.12.30.5 203 broadcast
no frame-relay inverse-arp
!
!-ommited-!
!
router ospf 1
log-adjacency-changes
network 2.2.2.2 0.0.0.0 area 20
network 172.12.30.0 0.0.0.255 area 0
network 192.168.1.0 0.0.0.3 area 10
neighbor 172.12.30.3
neighbor 172.12.30.4
neighbor 172.12.30.5
default-information originate
!
!-ommited-!

R3:

!-ommited-!
interface Serial0/0
ip address 172.12.30.3 255.255.255.0
encapsulation frame-relay
ip ospf priority 0
serial restart-delay 0
frame-relay map ip 172.12.30.2 301 broadcast
no frame-relay inverse-arp
!
!-ommited-!
!
router ospf 1
log-adjacency-changes
network 3.3.3.3 0.0.0.0 area 30
network 172.12.30.0 0.0.0.255 area 0
!
!-ommited-!

R4:

!-ommited-!
!
interface Serial0/0
ip address 172.12.30.4 255.255.255.0
encapsulation frame-relay
ip ospf priority 0
serial restart-delay 0
frame-relay map ip 172.12.30.2 401 broadcast
no frame-relay inverse-arp
!
!-ommited-!
!
router ospf 1
log-adjacency-changes
network 4.4.4.4 0.0.0.0 area 40
network 172.12.30.0 0.0.0.255 area 0
!
!-ommited-!

R5:

!-ommited-!
!
interface Serial0/0
ip address 172.12.30.5 255.255.255.0
encapsulation frame-relay
ip ospf priority 0
serial restart-delay 0
frame-relay map ip 172.12.30.2 501 broadcast
no frame-relay inverse-arp
!
!-ommited-!
!
router ospf 1
log-adjacency-changes
network 5.5.5.5 0.0.0.0 area 50
network 172.12.30.0 0.0.0.255 area 0
!
!-ommited-!

networker050184 · July 2011

What address is R5 going to use when it pings 3.3.3.3?

How is it going to resolve that address?

polevector · July 2011

networker050184 wrote: »

What address is R5 going to use when it pings 3.3.3.3?

How is it going to resolve that address?

Here's debug output when r5's is trying to ping 3.3.3.3

R5#debug ip packet detail
IP packet debugging is on (detailed)
R5#ping 3.3.3.3
Type escape sequence to abort.
Sending 5, 100-byte ICMP Echos to 3.3.3.3, timeout is 2 seconds:
*Mar 1 01:35:40.491: IP: s=172.12.30.5 (local), d=172.12.30.2 (Serial0/0), len 80, sending, proto=89
*Mar 1 01:35:40.787: IP: tableid=0, s=172.12.30.5 (local), d=3.3.3.3 (Serial0/0), routed via RIB
*Mar 1 01:35:40.791: IP: s=172.12.30.5 (local), d=3.3.3.3 (Serial0/0), len 100, sending
*Mar 1 01:35:40.795: ICMP type=8, code=0
*Mar 1 01:35:40.795: IP: s=172.12.30.5 (local), d=3.3.3.3 (Serial0/0), len 100, encapsulation failed
*Mar 1 01:35:40.799: ICMP type=8, code=0.
*Mar 1 01:35:42.787: IP: tableid=0, s=172.12.30.5 (local), d=3.3.3.3 (Serial0/0), routed via RIB
*Mar 1 01:35:42.791: IP: s=172.12.30.5 (local), d=3.3.3.3 (Serial0/0), len 100, sending
*Mar 1 01:35:42.795: ICMP type=8, code=0
*Mar 1 01:35:42.795: IP: s=172.12.30.5 (local), d=3.3.3.3 (Serial0/0), len 100, encapsulation failed
*Mar 1 01:35:42.799: ICMP type=8, code=0.
*Mar 1 01:35:44.787: IP: tableid=0, s=172.12.30.5 (local), d=3.3.3.3 (Serial0/0), routed via RIB
*Mar 1 01:35:44.787: IP: s=172.12.30.5 (local), d=3.3.3.3 (Serial0/0), len 100, sending
*Mar 1 01:35:44.787: ICMP type=8, code=0
*Mar 1 01:35:44.787: IP: s=172.12.30.5 (local), d=3.3.3.3 (Serial0/0), len 100, encapsulation failed
*Mar 1 01:35:44.787: ICMP type=8, code=0.
*Mar 1 01:35:46.787: IP: tableid=0, s=172.12.30.5 (local), d=3.3.3.3 (Serial0/0), routed via RIB
*Mar 1 01:35:46.791: IP: s=172.12.30.5 (local), d=3.3.3.3 (Serial0/0), len 100, sending
*Mar 1 01:35:46.795: ICMP type=8, code=0
*Mar 1 01:35:46.795: IP: s=172.12.30.5 (local), d=3.3.3.3 (Serial0/0), len 100, encapsulation failed
*Mar 1 01:35:46.799: ICMP type=8, code=0.
*Mar 1 01:35:48.787: IP: tableid=0, s=172.12.30.5 (local), d=3.3.3.3 (Serial0/0), routed via RIB
*Mar 1 01:35:48.787: IP: s=172.12.30.5 (local), d=3.3.3.3 (Serial0/0), len 100, sending
*Mar 1 01:35:48.787: ICMP type=8, code=0
*Mar 1 01:35:48.787: IP: s=172.12.30.5 (local), d=3.3.3.3 (Serial0/0), len 100, encapsulation failed
*Mar 1 01:35:48.787: ICMP type=8, code=0.
Success rate is 0 percent (0/5)

Hopefully there's an answer to your question.

Chris_ · July 2011

networker050184 wrote: »

What address is R5 going to use when it pings 3.3.3.3?

How is it going to resolve that address?

Networker is pointing you in the right direction, just another nudge from me

O IA 3.3.3.3 [110/65] via 172.12.30.3, 00:09:02, Serial0/0

how does R5 get to 172.12.30.3 over the frame relay network - can it get there?????

These are clues fro you not questions that we want answering!

polevector · July 2011

Chris_ wrote: »

Networker is pointing you in the right direction, just another nudge from me

O IA 3.3.3.3 [110/65] via 172.12.30.3, 00:09:02, Serial0/0

how does R5 get to 172.12.30.3 over the frame relay network - can it get there?????

These are clues fro you not questions that we want answering!

I'm sorry I'm a total noob, I understand that routing table is messed up.
pings from r5 to 3.3.3.3 should go via 172.12.30.2 since it's a hub. But I don't know how to fix it via OSPF.

Forsaken_GA · July 2011

*Mar 1 01:35:48.787: IP: s=172.12.30.5 (local), d=3.3.3.3 (Serial0/0), len 100, encapsulation failed

The encapsulation failed message is a big hint. It means something is wrong with your layer 2 to layer 3 mappings.

Take what Chris pointed out.

Then look at your DLCI mappings on R5.

Chris_ · July 2011

polevector wrote: »

I'm sorry I'm a total noob, I understand that routing table is messed up.
pings from r5 to 3.3.3.3 should go via 172.12.30.2 since it's a hub. But I don't know how to fix it via OSPF.

nothing wrong with the routing table, R5 is seeing it's next hop to R3 as R3 (172.12.30.3) as it is on the same subnet.

However, take a look at your frame Relay config on R5, does R5 have a DLCI mapped to the IP 172.12.30.3?

and then for R3 to respond, does it have a DLCI mapped to 172.12.30.5?

Forsaken_GA · July 2011

polevector wrote: »

I'm sorry I'm a total noob, I understand that routing table is messed up.
pings from r5 to 3.3.3.3 should go via 172.12.30.2 since it's a hub. But I don't know how to fix it via OSPF.

No they shouldn't. Look at what type of network you have OSPF configured as. All of the routers interfaces are in the same subnet, all members of the same /24. That means that R5 thinks it should be able to talk to R3 directly over the frame relay network, without a need to go through R2.

*Edited to remove incorrect information

polevector · July 2011

Chris_ wrote: »

nothing wrong with the routing table, R5 is seeing it's next hop to R3 as R3 (172.12.30.3) as it is on the same subnet.

However, take a look at your frame Relay config on R5, does R5 have a DLCI mapped to the IP 172.12.30.3?

and then for R3 to respond, does it have a DLCI mapped to 172.12.30.5?

That would make a direct link between r3 and r5

So then I'm wrong with the whole hub and spoke setup? It should be configured on separate subinterfaces and on different subnets?

Chris_ · July 2011

polevector wrote: »

That would make a direct link between r3 and r5

So then I'm wrong with the whole hub and spoke setup? It should be configured on separate subinterfaces and on different subnets?

You can map multiple IPs to the same DLCI, it just means that the router knows where to send them.

Chris_ · July 2011

polevector wrote: »

That would make a direct link between r3 and r5

So then I'm wrong with the whole hub and spoke setup? It should be configured on separate subinterfaces and on different subnets?

Nothing needs to change in your frame relay switch and it remains a broadcast network type on a hub and spoke setup

You are just saying, in your spoke router's config, that anything for the other spokes needs to be sent to the DLCI for the hub (R2)

-HOPE THIS MAKES SENSE.

Forsaken_GA · July 2011

Chris_ wrote: »

Nothing needs to change in your frame relay switch and it remains a broadcast network type on a hub and spoke setup

You are just saying, in your spoke router's config, that anything for the other spokes needs to be sent to the DLCI for the hub (R2)

-HOPE THIS MAKES SENSE.

Ah crap, you're right, if he shoves it down the same DLCI, it'll transit R2 before it goes out to R3.

Apparently I need to go refresh myself on frame relay!

Chris_ · July 2011

Forsaken_GA wrote: »

Ah crap, you're right, if he shoves it down the same DLCI, it'll transit R2 before it goes out to R3.

Apparently I need to go refresh myself on frame relay!

I hadn't realised you'd said anything incorrect!!!

polevector · July 2011

Chris_ wrote: »

Nothing needs to change in your frame relay switch and it remains a broadcast network type on a hub and spoke setup

You are just saying, in your spoke router's config, that anything for the other spokes needs to be sent to the DLCI for the hub (R2)

-HOPE THIS MAKES SENSE.

That does make sense! Although a weird work load detected! So for each spoke I decide to add later I'll have to not only to edit the hub's config but also each spoke which it needs to have connectivity to.

R5#ping 3.3.3.3
Type escape sequence to abort.
Sending 5, 100-byte ICMP Echos to 3.3.3.3, timeout is 2 seconds:
!!!!!
Success rate is 100 percent (5/5), round-trip min/avg/max = 32/56/84 ms
R5#traceroute 3.3.3.3
Type escape sequence to abort.
Tracing the route to n003-000-000-000.static.ge.com (3.3.3.3)
1 172.12.30.2 60 msec 44 msec 24 msec
2 172.12.30.3 84 msec * 76 msec
R5#

polevector · July 2011

Thanks for your help guys! I really appreciate it!

Forsaken_GA · July 2011

Chris_ wrote: »

I hadn't realised you'd said anything incorrect!!!

I was on crack and thinking that if you wanted to force transit to R2, you'd have to convert to p2p links instead, and I know better hehe

Chris_ · July 2011

Forsaken_GA wrote: »

I was on crack and thinking that if you wanted to force transit to R2, you'd have to convert to p2p links instead, and I know better hehe

Stay off that stuff

- However, my preference would be P2P links anyway.

That would solve the OPs question about workload with static mappings.

I've actually never seen a multipoint production network, just done it for studying.

capitanuionut · July 2011

I tried also OSPF over Frame Relay and until i didn't issued the command ip ospf network point-to-point on each interface of the spoke routers it didn't worked....try also ip ospf network point-to-multipoint on the hub router

Need help with OSPF over hub/spoke frame-relay

Comments