4

Buffing up on Frame-Relay for testing EIGRP nuances over FR for CCNP - topology and relevant show command output here:

https://i.sstatic.net/YUhcq.jpg

Relevant PVC is between DLCI 103/301 (R1/R4)

Why does R4 send echo replies to 192.168.123.1, when I ping from R1?? I know that if I source the ping from s0/0.103 it is successful, but I don't understand this conceptually. I figured when R4 ARP'd for 192.168.123.4 it would receive the DLCI 103 and the combination of the map in R2 and route in R4 would get it to where it needed to go, but I'm clearly missing something. Thanks in advance!

Improve this question

1 Answer 1

Reset to default
5

Your configuration is a bit strange. You have the same subnet on two multipoint interfaces. When you don't specify a source from R1 it chooses 192.168.123.1 as the source IP of the packet. R4 does not know how to encapsulate the frame to 192.168.123.1. This is how it looks from R1:

R1#ping 192.168.123.3

Type escape sequence to abort.
Sending 5, 100-byte ICMP Echos to 192.168.123.3, timeout is 2 seconds:

IP: tableid=0, s=192.168.123.1 (local), d=192.168.123.3 (Serial0/0.103), routed via FIB
IP: s=192.168.123.1 (local), d=192.168.123.3 (Serial0/0.103), len 100, sending
Serial0/0.103(o): dlci 103(0x1871), pkt type 0x800(IP), datagramsize 104.

If you look at CEF you can see that R1 has two entries for 192.168.123.0/24.

R1#sh ip cef 192.168.123.0 255.255.255.0
192.168.123.0/24, version 37, epoch 0, attached, connected, per-destination sharing
0 packets, 0 bytes
  via Serial0/0.102, 0 dependencies
    traffic share 1
    valid glean adjacency
  via Serial0/0.103, 0 dependencies
    traffic share 1
    valid glean adjacency
  0 packets, 0 bytes switched through the prefix
  tmstats: external 0 packets, 0 bytes
           internal 0 packets, 0 bytes

R4 does not know how to encapsulate the frame.

Serial0/0(i): dlci 301(0x48D1), pkt type 0x800, datagramsize 104
ICMP: echo reply sent, src 192.168.123.3, dst 192.168.123.1
Serial0/0.301:Encaps failed--no map entry link 7(IP)

You can add a frame map to resolve it.

R4(config)#int s0/0.301
R4(config-subif)#frame map ip 192.168.123.1 301

R1#ping 192.168.123.3

Type escape sequence to abort.
Sending 5, 100-byte ICMP Echos to 192.168.123.3, timeout is 2 seconds:
!!!!!
Success rate is 100 percent (5/5), round-trip min/avg/max = 36/49/60 ms

ICMP: echo reply sent, src 192.168.123.3, dst 192.168.123.1
Serial0/0.301(o): dlci 301(0x48D1), pkt type 0x800(IP), datagramsize 104
Serial0/0(i): dlci 301(0x48D1), pkt type 0x800, datagramsize 104
Improve this answer
1
  • Hi Daniel, thanks so much for your thorough answer! I spent more time going over it, and I eventually came to a similar conclusion, that I shouldn't have the same subnet over two different multi-point interfaces. Once I collapsed both interfaces into a single one, everything worked perfectly. I don't know about CEF at all, but it's looks very interesting. Onward and upward, thank you again for your time and consideration. Commented Aug 8, 2013 at 12:27

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.