dslreports logo
 
    All Forums Hot Topics Gallery
spc
Search similar:


uniqs
2430

R_K
join:2014-04-11
Canada

R_K to Glen1

Member

to Glen1

Re: [Homephone] An odd, recurring interactive "glitch" on a landline

said by Glen1:

When and if you call Bell to report the trouble. . . . describe the symptom and don't try to diagnose the trouble.

I hear you. When it comes to weird little issues of this sort, one can easily end up sounding like some kind of paranoid crackpot, too. The less said the better. I'll keep this in mind.
R_K

1 edit

R_K

Member

At the risk of sounding like a pest, I decided to add a few final posts to this thread for the sake of completeness (or maybe OCD).

Until today I hadn't tried to see what would happen if I caught the glitch in the act twice in a row on the same day. On several occasions I took the phone off-hook during the (roughly) 5:50 PM time slot, but never tried to do this again when the phenomenon "responded" by returning at (roughly) 6 PM.

Today, for the first time I picked up the handset at 5:48 and then also at 6:04 PM, when the quirk predictably came back. After what kyphos had said (that the test head is probably set to retry at fixed intervals), I kinda expected yet another "response" some 15 minutes later. So when the anomaly instead "struck back" again at 6:06 -- i.e. after only 2 minutes -- I was a bit surprised (I almost failed to record it). This is the first time ever that the glitch occurred three times in a single day.

So, another established fact about the algorithm controlling the (hypothetical) test head is that the "retry mechanism" is operating on something other than a simple fixed schedule. As I indicated in the original post, another irregular element of the scheduling involves the dates (and days of the week) at which the anomaly "chooses" to appear. It's impossible to discern any pattern at all; it seems perfectly random.

I know some will (understandably) ask "who cares?". At this point I'm no longer expecting, nor asking for, replies. I'm just trying to fully "document" this little curiosity.
R_K

4 edits

R_K

Member

landline_gli···OICE.mp3
149,211 bytes
An example from Jan. 16, 2014 - with speech
landline_gli···BEEP.mp3
53,080 bytes
An example from March 4, 2014 - with off-hook warning tone
Unusual Examples of the Glitch

At the top of this post I enclose two more MP3s with examples of the glitch. Some of you might like to check them out because they stand out from the many dozens of same-sounding recordings I've made. Unlike the others, these two samples feature not only the usual clicks, but also some extra sounds at the beginning.

I'm having trouble uploading attachments, so if they don't show up at the top of this post, try the links below.

- An example from Jan. 16, 2014 - with speech:

»www.hidrive.strato.com/l ··· IcUrt3Gm

- An example from March 4, 2014 - with off-hook warning tone:

»www.hidrive.strato.com/l ··· lEUrtabR

The example from Jan. 16th begins with a brief snippet of speech. Although the male voice is truncated, it's fairly easy to make out what it's saying (you might have to replay the file a few times, though). The recording from March 4th, on the other hand, begins with a brief electronic beep. This tone too is easily recognizable: it's a single burst of the off-hook warning tone that the line (the CO) puts out when the phone has been left off-hook too long, after the other party had already hung up. (I believe this tone is sometimes also used to indicate that a call is being transferred to another party.)

This thread is slowly getting old (and the issue is more of a curiosity than anything serious -- at least so far), but I'm wondering if anyone here knows if these additional sounds I've described (and enclosed) are completely consistent with the faulty-test-head theory. The example with the voice is especially puzzling to me. Doesn't it, by definition, indicate some sort of momentary crosstalk-type phenomenon? (To be clear, the snippet of speech was suddenly "injected" on the line while the line was idle, during the usual 5:48 PM time slot.)

Among the more than 70 instances of the glitch I've captured, there is only one recording featuring a voice, and only 2 samples with the off-hook warning beep.

Don't worry, I'm pretty much done with this thread. At this point I can't think of any other info pertaining to this topic that's worth mentioning...

bmccoy
join:2013-03-18
Port Orchard, WA

bmccoy

Member

I still think this thread is very interesting. Keep us updated!

Anyway, I think that this issue might still be caused by a bad line card. Instead of testing at a better time, its scheduling mechanism might be broken and the test tones aren't properly injected. So, it might be a combination of different bugs with your line card that should be checked out. Hearing another person talking on your line is kind of creepy, and that alone should make Bell want to check it out.
HeadSpinning
MNSi Internet
join:2005-05-29
Windsor, ON

HeadSpinning

Member

I agree. Bad line card. The line card sits on a digital bus that contains the multiplexed TDM voice and control signals. If the card is broken, it may be demuxing the wrong TDM timeslot (hence the snippets of someone else's conversation, and tones).

R_K
join:2014-04-11
Canada

R_K

Member

said by bmccoy:

this issue might still be caused by a bad line card. Instead of testing at a better time, its scheduling mechanism might be broken and the test tones aren't properly injected

said by HeadSpinning:

The line card sits on a digital bus that contains the multiplexed TDM voice and control signals. If the card is broken, it may be demuxing the wrong TDM timeslot (hence the snippets of someone else's conversation, and tones)

I appreciate the helpful comments, especially the more detailed suggestions. Once I've got a chance to get Bell to look at this thing, I might post about the results in this thread (assuming the admins will let me).

DKS
Damn Kidney Stones

join:2001-03-22
Owen Sound, ON

DKS to Thane_Bitter

to Thane_Bitter
said by Thane_Bitter:

Did Bell ever offer an explanation as to what the cause of that was? I am sure your local police force was thrilled once the false calls stopped.

No. Near as I can figure out (and Bell engineers denied) it was a flaw in the programming of the DMS100 switch. Once Ident-a-Call was removed, the problem stopped. It stayed away when the features were all added.
DKS

DKS to btech805

to btech805
said by btech805:

Normally the new process is the 911 service will actually alert us to any call issues, whether it be a disconnection during a real call or noise or false calls like this. Usually false calls are a problem within the central office

I hope that was one outcome of my experience. it was a slice having cruisers show up at your home several times a day and Bell denying there was a problem.
btech805
join:2013-08-01
Canada

btech805

Member

I wouldn't exactly say it is Bell denying any issues, it's more the call centre agents you speak to have no idea what you are talking about because it isn't on their call scripts. And even if they did know what you were talking about by some stroke of luck, they are penalized (aka fired if they're in the Phillipines or India) for deviating from the script. They're paid to talk not think.