Lost

[jlcarey1usa]

Strange situation. Have a company with two physical sites connected via a point-to-point T1. On each end of the T1 are old Cisco 1602R routers. The problem is actually with Exchange servers failing to talk to each other properly. Site A is main office and Site B is branch office. Each site contains 1 Exchange server and the sites are supposed to talk over this p-t-p connection. I cannot use telnet to connect from site A to site B over port 25. I can however, connect from site B to site A over port 25.

Essentially, the communication between these Exchange servers is failing because messages cannot go from site A to site B, but can go from site B to site A.

The interesting thing is that I can use telnet from site A to site B using a different port, say 691 which is also used with Exchange and it works fine.

I can telnet into site B router and establish a telnet session to the Exchange server in site B.

The problem is router A. For some reason, it will not allow requests over port 25 to go through.

Any clue???

[Richard Burts]

James

My best guess is that all of the physical interfaces are shut down. When you boot the router with no config it will put all physical interfaces into shutdown state. when you copy the NY config into the router the NY config does not have any no shutdown commands in it because when you saved the config on NY all the physical interfaces that you are using were in the no shutdown state.

My suggestion is to go into the router, issue no shutdown commands on the interfaces that would be used and see if that brings things up.

Now that I re-read my suggestion there is another aspect that occurs to me. The physical interfaces may be shutdown (my first guess) or they may be in protocol down state. If the physical interfaces are not connected to anything they will be in protocol down state. And that would also prevent any routes from showing up in the routing table.

So do the command show ip interface brief. The output will tell you what state the interfaces are in. If that does not give you enough information to solve this issue then post the output and I will see what other alternative I can come up with.

HTH

Rick

[jlcarey1usa]

Rick,

This is the sh ip route from the new NY router. It is a 2811. I don't understand the one static route in here: C 10.1.2.2/32 is directly connected, Serial0/0/0

10.1.2.2 is the interface on the GA router. Why would that be listed with a /32 mask? Why is it listed as directly connected to this router? Why is the 10.0.0.0 variably subnetted? I don't even understand where 10.0.0.0 came from because our p-t-p connection is 10.1.2.0

NY2811#sh ip route

Codes: C - connected, S - static, R - RIP, M - mobile, B - BGP

D - EIGRP, EX - EIGRP external, O - OSPF, IA - OSPF inter area

N1 - OSPF NSSA external type 1, N2 - OSPF NSSA external type 2

E1 - OSPF external type 1, E2 - OSPF external type 2

i - IS-IS, su - IS-IS summary, L1 - IS-IS level-1, L2 - IS-IS level-2

ia - IS-IS inter area, * - candidate default, U - per-user static route

o - ODR, P - periodic downloaded static route

Gateway of last resort is 192.168.0.2 to network 0.0.0.0

S 192.168.120.0/24 [1/0] via 10.1.2.2

C 192.168.110.0/24 is directly connected, FastEthernet0/1

10.0.0.0/8 is variably subnetted, 2 subnets, 2 masks

C 10.1.2.0/24 is directly connected, Serial0/0/0

C 10.1.2.2/32 is directly connected, Serial0/0/0

C 192.168.0.0/24 is directly connected, FastEthernet0/0

S* 0.0.0.0/0 [1/0] via 192.168.0.2

[jlcarey1usa]

This is the GA router, which we are planning on replacing. The real problem is that we cannot ping anything from NY to GA past the gateway of 120.1.

GA_router#sh ip route

Codes: C - connected, S - static, I - IGRP, R - RIP, M - mobile, B - BGP

D - EIGRP, EX - EIGRP external, O - OSPF, IA - OSPF inter area

N1 - OSPF NSSA external type 1, N2 - OSPF NSSA external type 2

E1 - OSPF external type 1, E2 - OSPF external type 2, E - EGP

i - IS-IS, L1 - IS-IS level-1, L2 - IS-IS level-2, * - candidate default

U - per-user static route, o - ODR

Gateway of last resort is 192.168.120.2 to network 0.0.0.0

S 192.168.110.0/24 [1/0] via 10.1.2.1

10.0.0.0/8 is variably subnetted, 2 subnets, 2 masks

C 10.1.2.1/32 is directly connected, Serial1

C 10.1.2.0/24 is directly connected, Serial1

C 192.168.120.0/24 is directly connected, Ethernet0

S* 0.0.0.0/0 [1/0] via 192.168.120.2

[Richard Burts]

James

First lets take the easy question:

I don't understand the one static route in here: C 10.1.2.2/32 is directly connected, Serial0/0/0

the entry for 10.1.2.2 is a byproduct of the address negotiation done by ppp. It is quite normal.

The 10.0.0.0 is the class A network of which you are using subnets. It shows as variably subnetted because you have configured a /24 subnet and then ppp reports a /32 (host address). So there are both /24 and /32 present in the network.

I am not sure why you can not ping from NY to the GA addresses. I do not believe that it is a router issue. What is configured on the end stations in GA for a default gateway? If the end station sends a response that gets to this router and if the ping is sourced from 192.168.110.0 then the router should send it.

HTH

Rick

[jlcarey1usa]

All default gateways in GA are 120.2, which is the Sonicwall. I believe this is our problem with contacting them via ping. Would you agree?

We set up the new GA router and shipped it down there today overnight. So, once we replace the existing one with the new one, I believe this should finally work.

[Richard Burts]

James

If the end station default gateway is the Sonicwall it does help explain the symptoms. If the ping from NY is coming over the serial link and the response from the end stations is going to the Sonicwall then it is an assymetric path and it might be expected that the Sonicwall would discard the outbound traffic that was not a response to inbound traffic that the firewall had seen. This is a common expectation for stateful inspection by a firewall.

I am not clear what about the replacement router will be different and will make things work from NY to GA.

HTH

Rick

[jlcarey1usa]

The replacement routers will create one gateway for the lan subnets. This makes it much easier to manage and understand. All requests for the other subnet will simply go over the p-t-p connection and all Internet requests will go out their respective sonicwalls. The big change is that it removes two physical gateways from the subnets and leaves one gateway. Ordinarily, I would agree that having two gateways makes sense for redundancy. However, for whatever reason, it does not work on this network. I am confident that this solution will work and should end this ridiculous situation. I have never seen a network so screwed up before.

[Richard Burts]

James

If you make changes along with the new router so that all end stations have their default gateway as the new router, then I agree that it should work ok. I was not aware that this was part of the plan for the new router.

HTH

Rick

[jlcarey1usa]

Rick,

Got it up and running last night. However, I see that there are errors on the serial interfaces. Specifically, the GA T1 sees it as a fractional CSU/DSU. Also, look at all the errors. Is there something we can change here?

NY2811#sh int s0/0/0

Serial0/0/0 is up, line protocol is up

Hardware is GT96K with integrated T1 CSU/DSU

Description: connected to GA via T1

Internet address is 10.1.2.1/24

MTU 1500 bytes, BW 1120 Kbit, DLY 20000 usec,

reliability 250/255, txload 16/255, rxload 72/255

Encapsulation PPP, LCP Open

Open: CDPCP, IPCP, loopback not set

Keepalive set (10 sec)

Last input 00:00:06, output 00:00:22, output hang never

Last clearing of "show interface" counters 16:25:24

Input queue: 0/75/0/0 (size/max/drops/flushes); Total output drops: 0

Queueing strategy: fifo

Output queue: 0/40 (size/max)

5 minute input rate 318000 bits/sec, 57 packets/sec

5 minute output rate 72000 bits/sec, 51 packets/sec

98544 packets input, 40972710 bytes, 0 no buffer

Received 0 broadcasts, 0 runts, 1 giants, 0 throttles

8668 input errors, 8657 CRC, 3741 frame, 1942 overrun, 0 ignored, 7325 abor

t

96055 packets output, 25410865 bytes, 0 underruns

0 output errors, 0 collisions, 944 interface resets

0 output buffer failures, 0 output buffers swapped out

11 carrier transitions

DCD=up DSR=up DTR=up RTS=up CTS=up

GA2611#sh int s0/0

Serial0/0 is up, line protocol is up

Hardware is PQUICC with Fractional T1 CSU/DSU

Description: Connected to NY via T1

Internet address is 10.1.2.2/24

MTU 1500 bytes, BW 1120 Kbit, DLY 20000 usec,

reliability 166/255, txload 77/255, rxload 17/255

Encapsulation PPP, loopback not set

Keepalive set (10 sec)

LCP Open

Open: IPCP, CDPCP

Last input 00:00:00, output 00:00:00, output hang never

Last clearing of "show interface" counters 16:19:51

Queueing strategy: fifo

Output queue 0/40, 0 drops; input queue 0/75, 0 drops

5 minute input rate 76000 bits/sec, 49 packets/sec

5 minute output rate 342000 bits/sec, 56 packets/sec

94845 packets input, 25362803 bytes, 0 no buffer

Received 0 broadcasts, 3 runts, 0 giants, 0 throttles

4460363 input errors, 21952 CRC, 4436092 frame, 0 overrun, 0 ignored, 2318

abort

102541 packets output, 44235706 bytes, 0 underruns

0 output errors, 0 collisions, 785 interface resets

0 output buffer failures, 0 output buffers swapped out

7 carrier transitions

DCD=up DSR=up DTR=up RTS=up CTS=up

[Richard Burts]

James

I do not understand your concern about GA showing it as a fractional CSU/DSU. Here is part of what you posted from NY earlier in the thread. Is this still the configuration of NY?

interface Serial0/0/0

description connected to GA via T1

service-module t1 timeslots 1-20

Assuming that it is still the configuration of NY then it clearly shows that NY is treating the connection as a partial T1 - and GA should treat it as partial also.

There are a fair number of input errors showing up on both sides of the connection. It is hard to know in what is posted how much of these errors were generated in the process of attempting to bring up the link and how many were generated after the link was active. I would suggest that you clear counters on both routers, let them run for a while (at least an hour or two - perhaps longer) and post the output of show interface then. That should clarify whether there is still some issue or not.

HTH

Rick