Re: New Point to Point T-1 Locking up ????
jason philbrook (jp@c6h22o11.midcoast.com)
Sat, 3 May 1997 22:12:24 -0400 (EDT)
>
> I have a new point to Point T-1 with IRX-114s at each end, running ver 3.5
> COMOS with 16 Megs Ram already installed in both. Only ports in use are
> for this new circuit.(S1)
>
> When I set up the circuit everything runs fine for about 10 minutes and
> then traffic stops flowing. If I do a reset S1 at one end the circuit goes
> into a connecting state and will not come back up until I do a hard reboot
> of the TSU. I wanted to blame the TSU BUT I have tried 4 different TSUs 2
> Kentrox Dserv TSUs and 2 Adtran TSU LTs have all failed. It appears that
> the TSUs are locking up. The routers can be reached from the ethernet at
> either end and they seem completely repsponsive to commands, at both ends.
> The telco(PacBell) says "we can loop the NI at both ends but we cant see
> your CSU". So it must be my problem.
>
> After resetting the TSUs the telco can run stress tests to and loop my TSUs
> without problems. Plug in the IRXs set up routes and the TSUs or at least
> one locks up again.
>
> Is it possible that a configuration of the IRX is confusing the TSUs and
> locking them up. ? This is my first point to point compared to Frame Relay.
>
> Is it possible that the IRX is the problem ?
>
> Is it possible that the circuit has a problem that is only found when there
> is live traffic compared to the telco testing ?
>
> Any other suggestions???
>
> Thanks
> Rick
If you have the adtran TSU, you can press the upper left hand button and
view the error information since it was powered up, or during the
reporting interval. If there are ANY errors, code violations, or errored
seconds, mention them to the telco, ask them to open a ticket, perform an
end-to-end test, and dont close the ticket till you have found an answer.
Stress testing a leased line from their test center doesnt test the whole
circuit. We've had telco carrier system problems which locked up ascend
routers such that they had to be powered down to fix the problem, and the
Livingston just said "connecting". When the telco went to put it into
test mode, the automated test program clicked something in the carrier
system which made the problem go away. They had to do a manual test, and
test their equipment on the lowest layers.
In our case, there was a problem with a Nynex T3, and a problem with an
independent's T3 demux/mux thinger referred to only by a number, and by a
bad piece of wire on a pole in the independent's local loop.
--
/*
Jason Philbrook | Midcoast Internet Solutions
jp@midcoast.com | Internet Access, LAN, WAN, and Linux
www.midcoast.com/jp/ | Service and Consulting for Midcoast Maine
*/