A nice bit of trivia for anyone who loves Frame Relay.

tunerX

I have a single frame relay PVC. Inverse-arp is running correctly, and the PVC is active. Here is the question.

What can be done by RouterA, that cannot be done by RouterB?

RouterA---DLCI 401---Cloud---DLCI 400---RouterB

RouterA----
interface Serial0
ip address 10.10.13.2 255.255.255.240
encapsulation frame-relay
fair-queue
frame-relay map ip 10.10.13.2 401
frame-relay interface-dlci 401

---

RouterB----
interface Serial0
ip address 10.10.13.5 255.255.255.240
encapsulation frame-relay
fair-queue
frame-relay interface-dlci 400

---

pad10029

hate to ask a question in response to a question.... but what is the purpose of RouterA having a static statement pointing to itself?

tunerX

That should lead you to the answer

Nocturnal

Router A will see its own mapping on a show frame-relay pvc query, whereas Router B won't?

tunerX

You are on a the right track; what does that imply for RouterA as opposed to RouterB?

Nocturnal

Router A has a mapping available of its IP address to its DLCI, whereas Router B does not; this implies that Router A has a configured interface but Router B does not. Router A up/up, and Router B up/down?

tunerX

3810_2#show frame-relay pvc 400

PVC Statistics for interface Serial0 (Frame Relay DTE)

DLCI = 400, DLCI USAGE = LOCAL, PVC STATUS = ACTIVE, INTERFACE = Serial0

input pkts 99 output pkts 55 in bytes 12846
out bytes 8005 dropped pkts 0 in pkts dropped 0
out pkts dropped 0 out bytes dropped 0
in FECN pkts 0 in BECN pkts 0 out FECN pkts 0
out BECN pkts 0 in DE pkts 0 out DE pkts 0
out bcast pkts 16 out bcast bytes 4646
5 minute input rate 0 bits/sec, 0 packets/sec
5 minute output rate 0 bits/sec, 0 packets/sec
pvc create time 00:13:32, last time pvc status changed 00:13:32
3810_2#show frame-relay map
Serial0 (up): ip 10.10.5.2 dlci 400(0x190,0x6400), static,
CISCO, status defined, active
Serial0 (up): ip 10.10.5.3 dlci 401(0x191,0x6410), static,
CISCO, status defined, active
3810_2#show ip int brief
Interface IP-Address OK? Method Status Prot
ocol
Ethernet0 192.168.1.1 YES NVRAM up up

Loopback0 1.1.0.1 YES NVRAM up up

Serial0 10.10.5.1 YES NVRAM up up

Serial1 unassigned YES NVRAM administratively down down

Virtual-Access1 unassigned YES unset up up

3810_2#ping 10.10.5.1

Type escape sequence to abort.
Sending 5, 100-byte ICMP Echos to 10.10.5.1, timeout is 2 seconds:
.....
Success rate is 0 percent (0/5)
3810_2#conf t
Enter configuration commands, one per line. End with CNTL/Z.
3810_2(config)#int serial 0
3810_2(config-if)#frame-relay map ip 10.10.5.1 400
3810_2(config-if)#end
3810_2#
00:14:48: %SYS-5-CONFIG_I: Configured from console by console
3810_2#ping 10.10.5.1

Type escape sequence to abort.
Sending 5, 100-byte ICMP Echos to 10.10.5.1, timeout is 2 seconds:
!!!!!
Success rate is 100 percent (5/5), round-trip min/avg/max = 20/21/24 ms
3810_2#show frame-relay map
Serial0 (up): ip 10.10.5.1 dlci 400(0x190,0x6400), static,
CISCO, status defined, active
Serial0 (up): ip 10.10.5.2 dlci 400(0x190,0x6400), static,
CISCO, status defined, active
Serial0 (up): ip 10.10.5.3 dlci 401(0x191,0x6410), static,
CISCO, status defined, active
3810_2#conf t
Enter configuration commands, one per line. End with CNTL/Z.
3810_2(config)#int serial 0
3810_2(config-if)#no frame-relay map ip 10.10.5.1 400
3810_2(config-if)#end
3810_2#
00:15:29: %SYS-5-CONFIG_I: Configured from console by console
3810_2#show frame-relay map
Serial0 (up): ip 10.10.5.2 dlci 400(0x190,0x6400), static,
CISCO, status defined, active
Serial0 (up): ip 10.10.5.3 dlci 401(0x191,0x6410), static,
CISCO, status defined, active
3810_2#ping 10.10.5.1

Type escape sequence to abort.
Sending 5, 100-byte ICMP Echos to 10.10.5.1, timeout is 2 seconds:
.....
Success rate is 0 percent (0/5)
3810_2#

pad10029

is this like using a loopback cable to test an int?

Nocturnal

I can see from the show frame-relay map commands how x.x.x.1 disappears from the output when the mapping is removed. I can also see that when the mapping was put back in there were two mappings to DLCI 400, x.x.x.1 and x.x.x.2.

Would you be able to ping x.x.x.2 the same way? I didn't see anything to distinguish between the two except for the fact that x.x.x.1 is the first valid IP address in the subnet. Is that significant?

Thanks for the exercise!

tunerX

From the terminal snippet that I included, you can see that the router cannot pings it's own interface unless there is a mapping for it.

So RouterA would be able to pings its serial interface and RouterB would not.

Nocturnal

3810_2#show frame-relay map
Serial0 (up): ip 10.10.5.1 dlci 400(0x190,0x6400), static,
CISCO, status defined, active
Serial0 (up): ip 10.10.5.2 dlci 400(0x190,0x6400), static,
CISCO, status defined, active
Serial0 (up): ip 10.10.5.3 dlci 401(0x191,0x6410), static,
CISCO, status defined, active


I see two Serial interfaces' IPs mapped to one DLCI. This looks like two routers at one end of a PVC. They are all static mappings, there is no sub-interface. What am I looking at? :

tunerX

3810_2-->DLCI 400-->FR<--DLCI 500<--2610_2

3810_2-->DLCI 401-->FR<--DLCI 600<--3640_1

2610_2-->DLCI 501-->FR<--DLCI 601<--3640_1

It is a multipoint triangle.

Serial 0 on 3810_2 has an IP of 10.10.5.1. When I am on 3810_2 I cannot ping my serial 0 IP address unless I have a static map to a local DLCI.

Nocturnal

2610_2 IP address: 10.10.5.2
3640_1 IP address 10.10.5.3

3810_2 talks to 2610_2 on DLCI 400, which responds on DLCI 500


To configure a loopback this way on the 2610:

2610_2(config)#int serial 0
2610_2(config-if)#frame-relay map ip 10.10.5.2 500
2610_2(config-if)#end

and on the 3640

3640_1(config)#int serial 0
3640_1(config-if)#frame-relay map ip 10.10.5.3 600
3640_1(config-if)#end


In each of the three cases, one could have chosen the "x00" or the "x01" DLCI to accomplish this. True?

tunerX

Yeah, you can specify any local DLCI. Your ping has to travel all the way to the distant router (with that DLCI) and then back.

Pings to your local interface will be almost twice as long as a ping to the remote interface.