From: Bit Gossip (bit.gossip@chello.nl)
Date: Tue Jul 24 2007 - 16:23:46 ART
I have labbed your setup and in both cases I have R2 sending directly to R3
because R1 set R3 as forward-address in its update to R2.
So no intervention of ICMP in any case.
r2#show ip ospf database external
OSPF Router with ID (10.10.4.2) (Process ID 1)
Type-5 AS External Link States
Routing Bit Set on this LSA
LS age: 57
Options: (No TOS-capability, DC)
LS Type: AS External Link
Link State ID: 10.1.3.0 (External Network Number )
Advertising Router: 10.1.1.1
LS Seq Number: 80000001
Checksum: 0x7AFB
Length: 36
Network Mask: /24
Metric Type: 2 (Larger than any link state path)
TOS: 0
Metric: 20
Forward Address: 10.1.1.3
External Route Tag: 0
The only way I can think to force R1 not to set the forward address is to
change the OSPF network type to P2P or P2M on R1 and R2
r2#show ip route ospf
10.0.0.0/24 is subnetted, 2 subnets
O E2 10.1.3.0 [110/20] via 10.1.1.1, 00:00:13, FastEthernet0/0
r2#show ip ospf database external
OSPF Router with ID (10.10.4.2) (Process ID 1)
Type-5 AS External Link States
Routing Bit Set on this LSA
LS age: 54
Options: (No TOS-capability, DC)
LS Type: AS External Link
Link State ID: 10.1.3.0 (External Network Number )
Advertising Router: 10.1.1.1
LS Seq Number: 80000002
Checksum: 0xC6BD
Length: 36
Network Mask: /24
Metric Type: 2 (Larger than any link state path)
TOS: 0
Metric: 20
Forward Address: 0.0.0.0
External Route Tag: 0
In this case indeed ICMP comes into play
r2#ping 10.1.3.3
Type escape sequence to abort.
Sending 5, 100-byte ICMP Echos to 10.1.3.3, timeout is 2 seconds:
!!!!!
Success rate is 100 percent (5/5), round-trip min/avg/max = 1/2/4 ms
r2#
*Jul 24 14:33:38.271: ICMP: redirect rcvd from 10.1.1.1- for 10.1.3.3 use gw
10.1.1.3
*Jul 24 14:33:38.275: ICMP: echo reply rcvd, src 10.1.3.3, dst 10.1.1.2
----- Original Message -----
From: "Ben" <bmunyao@gmail.com>
To: "Cisco certification" <ccielab@groupstudy.com>
Sent: Saturday, July 21, 2007 3:16 PM
Subject: IP redirect with OSPF IGP
> Hi
>
> Sometime back, I came across this interesting piece of information on the
> use of OSPF on a multiaccess segment (Ethernet):
>
> R2
> |
> | 10.1.1.0/24
> ---------------------------------------------
> | |
> | |
> R1 R3
> |
> --------------
> 10.1.3.0/24
>
> R1,R2 running OSPF, R1,R3 running RIP. if you use "net 10.1.1.1 0.0.0.0
> area
> 0" to enable ospf on R1, then R2 will forward traffic for 10.1.3.0/24 to
> R1,
> R1 will send an icmp redirect to R2, informing it that R3 is metrically
> closer to the destination.
>
> What I didn't know was that if instead you use "net 10.1.1.0 0.0.0.255
> area
> 0" on R1, R2 will no longer send traffic to 10.1.3.0 via R1, and will
> instead learn dynamically the metrically closer next-hop of 10.1.1.3
> through
> OSPF, thus saving one hop.
>
> Does anyone have variations of this scenario I could try out, to test this
> concept.
>
> TIA
> Ben
>
> _______________________________________________________________________
> Subscription information may be found at:
> http://www.groupstudy.com/list/CCIELab.html
This archive was generated by hypermail 2.1.4 : Sat Aug 18 2007 - 08:17:41 ART