Unresolved support ticket open for more than six months - is this a record?

[#1]

Careful jelv, if we follow that logic to an extreme, we'll be saying how good DLM is  I don't think.

[#2]

Well, the saga continues: a BT engineer returned to my house this afternoon, but I was at work (they hadn't previously advised me they would need access to my property); fortunately, my mother had just popped over to check on something else for me and she was able to let the engineer in to do further line tests, after which he left a note for me which read:
"I can find no fault with your broadband today. 448/5280kbps - no errors. I ran a test on your copper wiring back to the exchange. This test needs an earth contact at this end, I had to use your heating pipe for this because when I used your mains at the socket I had a warning pop up on my meter that suggests you may have a possible earth problem with your mains. It may be worth having it checked out. (It may / may not affect your broadband. I'm not an electrician, I'm afraid)."
But no news on whether the "lift and shift" at the exchange was done...
I've run another BT speed test (at 17:36 this evening) and this shows an extremely low throughput rate and IP profile:-
"Test1 comprises of Best Effort Test: -provides background information.
Your DSL connection rate: 6176 kbps(DOWN-STREAM), 448 kbps(UP-STREAM)
IP profile for your line is - 135 kbps
Actual IP throughput achieved during the test was - 103 kbps"
A very sick line indeed, and worse than when BT started tinkering with it on Tuesday...
Before I go to the expense of getting an electrician to come out and do earth continuity and integrity tests, does anyone know whether this is likely to affect the broadband performance at all as the BT engineer didn't seem sure.

[#3]

Quote from: Catweazle

Your DSL connection rate: 6176 kbps(DOWN-STREAM), 448 kbps(UP-STREAM) IP profile for your line is - 135 kbps

I would count that as significant progress TBH.
You now have a sync rate that is in the right "ball-park".
The IP Profile should adjust relatively quickly if you stay connected at that speed. Don't disconnect too often though.
Your poor earth MAY indirectly add to background electical interference but I wouldn't expect it to directly affect ADSL in a major way.

[#4]

Quote

A very sick line indeed, and worse than when BT started tinkering with it on Tuesday.

I'm not so sure about that ! At least your synch speed is back up to something sensible. The profile could well be shot by the number of disconnects due to the lift and shift. Let's see if the synch stays that way , if it does then then profile should follow pretty rapidly...
HPSauce types faster than me obviously..

[#5]

I'd like to think that the improved synch rate is indeed progress, but I've been disappointed several times before i.e. BT engineer visits, resynchs at 6000+kbps and then a few hours later the synch rate drops down again. Indeed, as I was typing this, the synch rate has dropped down to 2880kbps (noise margin stayed roughly the same at 9Db, apart from the drop to 0Db when the connection was dropped).

[#6]

The noise margin will be the same because after a resync it will be at or near the target.
Can you run Routerstats against it to log the noise margin with it sat at the lower sync speed, that might give some clues.

[#7]

I've been running Routerstats since I got home (just after 5pm) but the graphs don't seem to show anything that interesting; but I've posted the Rx noise margin and sync speed graphs for info.

[#8]

Something went badly wrong about 18:20 but as you say nothing showed up on the RX noise.
Can you start the graphs for all functions and see if anything shows up.
I am particularly interested in the error rates up and down both Rx and Tx
Also very worrying is that the noise on your line jumped by a heck of a lot (noise margin didn't change but the speed did which is the same effect)
Is there anything in your locality which came on at that time as it was a step change effect not a gradual change

[#9]

That's just bizarre.
What are the DMT tone graphs like?

[#10]

Quote from: Oldjim

Is there anything in your locality which came on at that time as it was a step change effect not a gradual change

I'm not aware of any electrical appliances that came on at that time - I don't have anything on a timer apart from the gas central heating. But I think the c/h is ok (unless it has a very intermittent fault) as I haven't noticed the sync rate and noise margin being affected when the boiler starts up (it's a combi boiler, by the way, so no h/w cylinder to keep topped up with heat).
I've attached a couple of graphs from 31st Jan as this also shows quite a bit of activity after I rebooted the router at 10:05 (because the sync rate was very low). Again, I wasn't aware of any other electrical activity and I remember doing a BT quiet line test as well - no noise discernible on the line. I've not been using Routerstats long and I can't see how to switch on the other graphs that you mentioned - could you give me a pointer, please.

Quote from: HPsauce

What are the DMT tone graphs like?

I haven't set up DMT yet, but I did manage to get the Netgear DG834 v4 router back late last night and downloaded the latest version of Routerstats as well. I need to check the Routerstats instructions to see if anything else needs configuring for the DG834 (or whether the default settings are fine), but please let me know if I should look for anything in particular.

[#11]

Now those graphs are REALLY INTERESTING. Especially around 11:15 to 11:20! 
The sync speed was constant but the Noise Margin moved down by 9dB!
That's serious chunk of interference and explains everything. Shame you couldn't show that to the BT person.
Now you have to find it!

[#12]

Just to clarify my thinking - others may have different ideas of course.
You need to get your router sync'ed at a long-term stable speed - this will be low at this stage.
Then you need to monitor it constantly with Routerstats in order to identify when the interference occurs.
You may also be able to use DMT to:
a) force the sync rate to what you need.
b) Look at the tone graphs for clues on frequencies.
It may need a specialist REIN investigation.
You should also get your earthing sorted out ASAP, that may be at the heart of this!
(not to mention a safety risk)

[#13]

Quote from: HPsauce

Now those graphs are REALLY INTERESTING. Especially around 11:15 to 11:20!

Well, I did show these graphs and some of my other stats to the BT engineers on Tuesday, but they seem to be very blinkered when it comes to anything that doesn't come out of a BT-branded box i.e. "thanks for showing us, but our equipment says you don't have a problem at the moment" (or words to that effect). Infuriating...
I've also noticed that Routerstats has some options specifically for v4 of the DG834 so I'll tinker with these to see what comes out. I've managed to get a Telnet Router tab to appear and this has lots of sub-tabs, one of which shows Bits/Tone - is this the tone graph you mentioned? Or does DMT do something similar? Can't see how to capture the Routerstats tone graph just yet, but will do some more digging later.
Re. the earthing - yes, safety considerations should be paramount but, to be honest, I was a little sceptical that there was actually a problem as there have been so many red herrings to date and I don't have a lot of confidence in the BT engineers, I'm afraid. I've spent a lot of time and money on this problem so far, so paying more for a potential "poisson rouge" isn't my number one priority at the moment. But I guess the nagging safety aspects will force me to consult Sparky soon...

[#14]

Just got DMT Tools up and running and attached a snapshot for info...

[#15]

By comparison look at mine - The main difference you will see is that my bits start to cut off at over 170 (that is mostly due to but swapping) but I never get over 200 even immediately after a resync at 15dB default noise margin because I am a lot further from the exchange.
The other thing to note is that your upstream noise margin is lower than it should be. I have exactly that problem only worse where the upstream sync speed drops to as low as 196kbps at 6dB noise margin (this coincides with noise on the line but it is so intermittent I have given up on it.)
Looking at the general shape of the graph your lower frequencies are running at a lower noise margin than the upper frequencies which, together with the upstream noise margin would lead me to believe that there is a source of interference in those frequencies (600 - 680 kHz for the downstream) which is exactly in the medium wave radio frequency band. http://www.mediumwave.de/#list
Now why your connection is sensitive to this I have no idea.