cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
1166
Views
0
Helpful
9
Replies

CSU/DSU

amitsonik
Level 1
Level 1

I have Cisco 4000 in my office and I have Cisco 4500 at our colo. I have tested the configuration back-to-back using DTE and DCE cables and the Cisco's seems to be working fine. They also work fine with having the 2 Larscom Split-T CSU back-to back. But when I put the Larscom Split-T csu on either ends of the T1 loop it doesn't work. I once thought that there is something to do with the clock of the loop. Then I have also tested having Normal Clock at one end and Inverse clock on the other, then too it doesn't work.

Please advice what else can be done to solve this problem.

Thanks,

Amit

9 Replies 9

svermill
Level 4
Level 4

Based on your other postings, it seems like you may have already fixed this and moved on. Are you still having trouble? You need to match the framing and coding of the T1 loop on your CSUs and you need to know the clock scheme. One of two clock cofings, depending on the T1 setup: a) both CSUs set for recovered/line timing. b) One CSU set for internal, other set for recovered/line. Don't mess with inverted clock unless you are compensating for counterdirectional timing issues on long copper cable runs.

When I connect the two Larscom's on the T1 loop, the Line indicator lights up. Which means that 2 csu's are talking to each other. But the Line Protocol on the Router doesn't come-up. Though when i connected the 2 Larscom csu's back-to-back the Line protocol comes up. To troubleshoot this problem I have even tried looping the two Larscom csu's and they get looped fine and the Line protocol comes up on one of the Routers depending on the direction of the loop. I will check this today that you have sent.

Thanks once again.

I bet that it's the Telco's wiring...isn't it ALWAYS their fault!? :)

I've been to sooo many customers in the past to do router installs, and I had re-punch the T-1s myself...I can't even count. I tell you it teaches you troubleshooting anf wiring very quickly! A lot of time, the Tx and Rx will be reversed. Anyway it might not be the problem in your case, but it's worth taking a look at it anyway. If you have an "extended demarc" make sure the pin-out (1-2-4-5)is correct all the way from the NIU to you computer room .

Good luck!

You will want to verify this against your Larscom documentation, but...

The "Line" indicator light does not necessarily mean that CSU A sees CSU B and vice versa. Usually all that is required is for the CSU to see valid framing and maybe valid coding as well. You can probably turn one of the CSUs completely off and still have the line indicator all lit up on the other end. So your T1 may not even be provisioned (correctly).

Since it doesn't sound as if you have T1 test equipment to verify the circuit yourself, you should definately open a ticket with your LEC. Depending on your version of firmware in the Larscom, you can enable a function (IEEE T1.403) that allows the LEC to remotely test to (and control) your CSU without your even needing to touch it.

I have tested the T1 loop, since there is a option in the Larscom unit that you can Loop the T1-loop, so when I did a loop from my colo to my office the Line Protocol on my Cisco 4500 came up. Then I have also tried the doing a loop from my office to the Colo csu and the Line Protocol on my Cisco 4000 comes up.

Yes, we have t1403 been set on the Larscom.

Here I am working on both the sides of the network. I mean in the colo and the office and we have just got this loop from Verizon.

Thanks

When i disconnect the other csu the light on the line becomes red(alarm) on the CSU. Doyou still feel that T1 is not provisioned correctly.

No, I doubt that your T1 is misconfigured based on your latest info. Sounds as if you problem is getting one Cisco to talk to the other. I would focus on the config of those serial interfaces (encapsulation, etc).

Yes, It needs to check the two serial ports of the router. If its encapulation is HDLC and the keep alive time on both port is not same, it could be caused the same problem. Please check the keep alive time.

I have tested it with HDLC, and have also used debug serial interface command to check the keepalive's. The sequence is not correct. They don't match. I have also set the keepalive 10, command on the interface it doesn't seem to help. But when i connect the 2 routers and csu's back-to-back with a T1 cross connect, they work fine. But it is only when I put them on the T1-loop and connect the routers' again to the csu's on either ends that I don't get the Line Protocol up on the Cisco's.

Getting Started

Find answers to your questions by entering keywords or phrases in the Search bar above. New here? Use these resources to familiarize yourself with the community:

Innovations in Cisco Full Stack Observability - A new webinar from Cisco