"The Horror! The Horror!"

... A night without the internet.

(Until mid 2015, I depended on a dial-up connection.)

Yesterday afternoon, I logged on and managed to get a lot done, including making a last-minute just-in-time online payment of my AT&T telephone bill.

That evening (the Perversity Of The Universe at work here? :-) I started getting interruptions and disconnects from my phone connection. When I did manage to get an internet connection (before losing it again) it was painfully slow (maybe around 12 kB/s showing on the connection widget, and only a small fraction of that on the browser).

Attempts to phone anyone usually succeeded, but with a lot of noise and static on the line.

I called one of the '1-800' repair numbers for AT&T and got a guy who could also clearly hear the static. After I confirmed his suspicion that there was heavy rain in my area, he asked me to hang up while he did a test of the line, saying that he would call back in a minute or two.

He did, telling me that he had located the problem in an outside line, and that someone would be on it in the morning. As it was outside, they didn't even need access to my apartment.

A huge chunk of Houston's telephone service network is underground (during Hurricane Ike, in September 2008, the telephone landlines were the most robust of the utilities here, still functioning in areas where cellphone service was dead), but some parts of it are aboveground on utility poles.

Because of the rain, I'm gonna take a wild guess that tree branches come into this somewhere.

This morning (about 0930) I logged on. Back to whatever passes for "normal".

While I had heard of something like this before with communication systems,  I still thought it so cool that he found the location of the problem from whatever center he was operating from.

I'd love to hear from anyone who may know just how he did that (I didn't ask him as I felt he probably had a lot on his plate at that time).

I'm guessing some way of sending a signal on the line and measuring response time and signal strength, or maybe various gadgets on the line to locate signal problems.

Why don't I just look it up on the internet myself?

I'm trying, but another real horror story is trying to conjure up the correct incantation phrasing of words to get the results I'm after (rather than countless items on using the telephone for remote diagnosis of other systems).

I'm SEVENTY YEARS OLD!!! I'm not sure I have enough years left to find that information through an internet search.

HELP!!! :-)

Addendum - Same day, 1352 CDT - The info I'm after may even be based on some pretty ancient technology. I have a vague recollection of reading somewhere that when the trans-Atlantic telegraph cables were laid, they had  already worked out a method of knowing just where to look for a break in the cable.  I'll try to find out some more when I get back from work this evening.

Update - 2340 CDT, Tuesday, 10 Jul 2012 - Commentor Anonymous pretty well answered my question of how it was done (see first comment). My recollection about the trans-Atlantic cables was off; they do use the technique, but the first cables were laid long before it was developed.

You may have gathered that I consider AT&T's techies, charged with helping to solve your problems, first-rate and world class.  But, their others are a different story.

So, for the absolutely perfect cherry with which to top off this experience ...

As I was about to leave for work, I picked up my cellphone and noticed a voice-mail waiting for me. It was from the repair group asking me to call them and let them know if the issue had been satisfactorily taken care off.

Calling that '1-800' number gets you an automated menu that, if you patiently wait out their options, will eventually get you a human being to talk to (as it did last night).

This time, when I thought that was about to happen, what I got instead was
  "Thank you, for calling A-T-&-T."
  "Good Byyyeee!" 
   (click)
   (dialtone).
-

1045 CDT, 12 Jul 2012 - I referred above to a comment by Anonymous that answered my question. Most people reading this post are probably viewing it on the main page, in which case they will never see that comment unless they click on the "comment" link. As few ever do, I've removed that comment from the comments section and am here placing it within the body of the post ...
--------------------------------------------------------------------------
Anonymous said...

One of the most common ways to do this is called Time Domain Reflectometry (TDR). It works kinda like radar. A pulse is sent down the line and is reflected by changes in the cable. The time it takes for the pulse to reflect back is computed to find distance. The shape (actually phase shift) of the reflection can tell if it is a short, open or change in capacitance (degraded insulation etc). Pretty cool tech and its been around for about 70 years or so. Used to be you needed to be an engineer to figure out the reflections but now it is all computerized and the machine spits out the distance and probable fault.

Comment posted: July 10, 2012 7:44 PM
--------------------------------------------------------------------------
-