T1 Serial Problems

Discussion in 'Cisco' started by John, Sep 18, 2009.

  1. John

    John Guest

    Thanks again Andrey. I'll see if my guy can try the loopback test tomorrow
    and I'll post the results. However, I am a little concerned about the
    answers.

    If the line isn't the problem, and it's a low probability that it's a cisco
    hardware problem, then the problem probably lies with the telco. If that's
    the case, what could be causing the problem at the telco? Just curious.
    Thanks again for all the help.
     
    John, Sep 24, 2009
    #21
    1. Advertisements

  2. From the highest to lowest probability -

    1. Cable run between smartjack and RouterB
    2. Telco
    3. RouterB CSU/DSU
    4. Something else

    Check the cable first, worry about telco later :)

    Regards,
    Andrey.
     
    Andrey Tarasov, Sep 24, 2009
    #22
    1. Advertisements

  3. If the LCVs and PCVs are increasing, thats your biggest problem.
    In a straight-up point-to-point T1 circuit, the telco doesn't provide
    clock. The clocking comes from RouterA's 'clock internal'
    statement. In your setup you need one side to provide clocking, and
    one side to sync to it. (ie. clock internal vs clock line).
    That shouldn't do anything different. Clocking issues will just get
    you slips and if the slips get bad enough, and occasional line
    protocol down.

    If you are incrementing LCVs and PCVs, then you have something lower
    than worry about the clocking. Clocking doesn't affect these at *all*.
    One of your routers needs to originate clock, one needs to sync to
    that clock. If you do both set to line, they'll try to sync to
    whatever the other is doing, but you will have slips eventually. Clock
    slips mostly show up on voice circuits with odd digital noises and things.
    In data circuits, if it continues on enough, you can take a line
    protocol hit from time to time, but most times will work fine.
    In my experience, this sort of issue is going to be something bad in
    the CO, a bad smartjack card, or bad wiring. At the small end of
    probability is a bad CSU/DSU on your T1 interface. At zero probability
    is the router. The CSU/DSU is much more likely a culpret
    than the router, but this looks more like a cable or CO
    issue. Replacing the router while retaining the CSU/DSU will most likely
    do nothing for you.
    I don't think so, other than replacing the cable may bypass bad wiring.
    The ABAM cable won't be so much better, it'll just be different copper
    pairs, which you could probably do by switching to different house pairs.
    Unlikely. More likely is a CO or cabling issue.
     
    Doug McIntyre, Sep 24, 2009
    #23
  4. John

    John Guest

    Thanks. What is the difference between this cable and a
    standard cat5? Is this one basically a cat3 with less
    twists that a cat5?
     
    John, Sep 24, 2009
    #24
  5. The main difference with the ABAM cabling is that each pair is
    individually shielded, and it uses 22AWG copper inside (which can be a
    problem if you are punching this down into anything, 66-blocks will
    take the 22AWG, but patch panels, 110's or Krone/BIX won't take the
    22AWG wire, it'll be too big, it was designed for screw terminals).

    A real of Beldon 7838A is around $460.

    I also don't know of any that have Plenum jackets, which could very well
    be an issue, 7838A is PVC, and the stuff I've seen in the field is as well.
    You may run into fire codes preventing its use.
     
    Doug McIntyre, Sep 24, 2009
    #25
  6. John

    John Guest

    Weird, my newsgroup access has been down for a couple of days. Anyway, we
    replaced some hardware and nothing changed.

    Andrey, I'm going to take your advice and perform the loopback testing. I
    probably won't be able to test the problem router for a couple of days, but
    will post the results when I do. Right now, I need to read up on how to do
    loopback testing. :cool:

    What loopback tests should I perform? If I unplug RouterB and plugin the
    loopback there, and run an extended ping from RouterA through the loopback,
    and everything works fine, this verifies that my problem is either my long
    cable or some type of hardware failure with my cisco device (either router
    but more likely csu/dsu card) right?

    If that loopback test fails, this means the problem exist with the Telco
    wiring right?

    What are the extended ping commands that I should use for this test and what
    am I looking for? Thanks again for all your help!


     
    John, Sep 29, 2009
    #26
  7. Wiring, configuration, repeaters, etc.
    Ping with 0x0000, 0xFFFF, 0x4040 and 0x8000 patterns. Tell us if any of
    them fail.

    Very good source of information about T1 circuits is T1 A Survival Guide
    by Matthew S. Gast.

    Regards,
    Andrey.
     
    Andrey Tarasov, Sep 29, 2009
    #27
  8. John

    John Guest

    Thanks. I'll post the results. I also just ordered the book you suggested.
    I do have one last (I hope the last) question though:

    If I plug in my loopback wire into the smartjack where routerB is located,
    and perform the ping test from RouterA, and everything works fine, then the
    problem isn't with my Telco.

    From here, I would like to verify that the problem is with the long cat5
    cable and not with RouterB. To test this, I was thinking I could plug in a
    short cable from the smartjack to RouterB, configure RouterB to have a
    software loopback, and then run the ping test again from RouterA. If all of
    that works fine, I'll do the same software loopback test from RouterA, but
    with the long cable from the RouterB smartjack to RouterB. If there are
    errors, then it's definately the line.

    Anyway, I have a question about configuring RouterB for the loopback. If I
    go into my serial interface configuration:
    RouterB(config-if)#

    There is a loopback command:
    RouterB(config-if)#loopback

    From here, there are several options:
    RouterB(config-if)#loopback dte
    RouterB(config-if)#loopback line
    RouterB(config-if)#loopback remote

    I think the command that I want to just test the circuit from RouterA to
    RouterB (without going through the CSU/DSU card [which will just test the
    external line]) is:
    RouterB(config-if)#loopback line payload

    Is this correct? If it is, then can I just run the same test you suggested
    earlier to test the line?

    Thanks again. Huge help.
     
    John, Sep 29, 2009
    #28
  9. Well, if you can drag RouterB close enough to smartjack to bypass your
    Cat5 cable, you can just run those patterns between the routers. No need
    for loopback.

    Regards,
    Andrey.
     
    Andrey Tarasov, Sep 30, 2009
    #29
  10. John

    John Guest

    Hey guys,

    Thanks for all of the help on this problem. We were finally able to
    coordinate with the ISP correctly and they determined that the problem was
    on their line back to the CO.

    Thanks again.
    John



     
    John, Oct 27, 2009
    #30
    1. Advertisements

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.