Erratic U/S SNRM and resulting DLM interventions

[#1]

*books the day off*

[#2]

Okay, this getting silly. Overnight February 5th-6th, DLM once again intervened, and set connection speed back down to 7866kbps/downstream SNR back up to ~14.5dB, which it has now held up to now. An incoming phone call this morning gave rise to an erratic d/s margin, but did not drop the connection, and the margin returned to 'normal' when the call was finished.

Checking to graph just prior to posting this shows a two minute period when the d/s snr dropped to around 6dB, then shot back up to 14.8, but the u/s trace is now slightly erratic, having previously relatively stable.

Below are the graphs for the 5th and 6th.

[#3]

Hi @jab1

Sorry to hear that this is playing up again - if you raise a fault, we'll ask our suppliers to take a look.

Best wishes

Dave

[#4]

@RandallFlagg Thanks for your response. I am a little concerned, however, that if I raise a fault, OR will come out, do their usual tests, ignore data they are shown, report NFF, and I will be left with a £65.00 charge and no resolution - not an acceptable outcome, I'm afraid.

For information, I will post my two available RSL plots for today, with the following explanations:- The unstable u/s (green) trace from ~09.29 to ~10.32 was during two outgoing calls, the loss of plot at 10.54 was due to RouterStats 'losing it', but as I was watching, a quick restart fixed that and then I finally had an unrequested re-sync at ~13.09 - you can see the result!

[#5]

Hi @jab1

I understand your concerns, but given your knowledge and experience as a forum veteran, I'm sure you appreciate that there's little we can do regarding this that hasn't already been done, particularly as the issue appears to be related to the correlation between the physical phone and broadband aspects of the service.

Did you have an alternative resolution in mind outside of the engineer appointment?

Best wishes

Dave

[#6]

Dave, do I detect a bit of exasperation in your last reply?

Sadly, the only other thing I can try from my end is to install the spare TG582n router I have (which I now intend to do over the weekend).

My reluctance with regard to the BT engineers is solely down to my past experience with them - show them the hard evidence and they say 'not interested/irrelevant' go and fiddle elsewhere and the problem comes back almost as soon as they've left my short road.

[#7]

Thanks for coming back to me, @jab1 - not exasperated, just trying to be realistic in terms of the next steps.

The TG Router is a good shout though - let us know how you get on once you've managed to get it hooked up over the weekend.

All the best

Dave

[#8]

No problem, Dave - my comment was meant in jest(note the ). I understand about setting expectations, but as I said, a 'NFF' report and a £65.00 bill is not something I relish.

[#9]

Interesting that you say that phone calls cause the broadband rate to change. As I know you know, phone calls and broadband are on different frequencies, so one should not affect the other.

So as one is affecting the other, all I know that can cause this is:

1. Microfilter problems, so if you can change this it must be worth trying
2. Non linearities on the line. This is quite unusual. I had it one when the cover came off a BT box up a pole, and contacts inside corroded in the weather.

Now as phone can mess up broadband, you would also expect the opposite, so if you start a download (say a you tube video), what does the quiet line test do then?

If it is a fault like this, the engineer can pick it up easily, they have kit that measures the distance to changes in impedance, or maybe where cross talk happens. This means that They can usually tell the distance from the socket to the problem. In my case they detected a 40 year old connection in my undercroft (which they changed) and at something like 30m the problem with the box that had become corroded.

Anyway cross talk between phone and broadband is good, if only because its a very clear sign that there is something badly wrong.

[#10]

Hi, @summers nice to see you are still around.

To answer you, and thanks for the input, I will say this:-

I know the BB and phone are on different frequencies (and shouldn't cause the effects they do) for me.

(1) Changing microfilters - three times - didn't resolve anything, and the replacement of my NTE by the last engineer by a 5C has not cured it either.

(2) I don't think the last guy went up my local pole or down into the chamber my wires come out of, but he did say he had run external checks prior to knocking on my door, but I don't know what they were.

The phone messing the BB is not every time - i.e. a call about 20 minutes ago merely caused the u/s trace to fall and didn't affect d/s at all, but with my current d/s SNR I'm not too surprised. NOTE: there was no noise on the line during the conversation at all, but that is normal - the only time I do get noise during a call is if the re-sync occurs during the call.

I don't know if you've read through the thread, but the RSL plots I've provided tell a sorry tale.

I'll try a download/QLT test tomorrow sometime - good call.

As I've said earlier in the thread, my final personal equipment change will be to swap out my current 582N for the spare PN sent me in error about ten back, but to be honest I don't think that is going to do much apart from bug me with changing al the devices to the new password &c.

I will add tonights' RSL plot later.

Edited to to add RSL plot.

[#11]

Hi @jab1,

I'm sorry that you're still experiencing problems with your connection and for the inconvenience caused.

I can see from your connection log that there have been a high number of drops, so there definitely looks to be something out of place:

However, testing your line isn't picking up the cause of this:

xDSL Status Check
Circuit ID:	CBUKXXXXXXXXX	Service ID:	BBEUXXXXXXXXX
Telephone NO.:	NA	Test Executed On:	09-02-2019 09:59:53

xDSL Status Test Summary
Sync Status:		Circuit In Sync
General Information
NTE Status:		NTE Power Status:	Unknown	Bypass Status:

	Upstream DSL Link Information	Downstream DSL Link Information
Loop Loss:	15.9	32.5
SNR Margin:	6.3	14.9
Errored Seconds:	0	0
HEC Errors:	0
Cell Count:	4437	25380
Speed:	804	7884

Maximum Stable Rate (KBPS):	11360	Fault Threshold Rate (KBPS):	9088
Mean Time Between Retrains (Seconds):	21575	Mean Time Between Errors Upstream (Seconds):	669
Indicative Line Quality:	A	Mean Time Between Errors Downstream (Seconds):	355
Custom Thresholds
MTBR_RED:	MTBE_RED:
MTBR_GREEN:	MTBE_GREEN:

As the previous fault ticket has closed, please can you report the issue here and let us know when you've completed it. We'll then be able to progress this further for you.

[#12]

Morning, @EmilyD - thanks for the RADIUS plot and KBD check - useful.

As I said a little earlierr in this thread, I intend to swap out my old router for  a spare this weekend to see if if that has any effect, so that will effectively mean everything my side of the network is new. If this doesn't solve the problem, then I will raise a fault, as I will then be reasonably certain the errors aren't my side of the circuit.

Quick edit - I notice max d/s stable rate is quoted as 11360KBPS on the KBD report - RouterStats is telling me my connection speed is 12165 - does this give any clue?

[#13]

Hi @jab1,

Thank you for trying a new router - we'll investigate this further for you if the new router doesn't resolve the connection issues.

[#14]

@EmilyD, @RandallFlagg, @Gandalf Replacement router installed around 1440 today.

[#15]

Right - this is helpful. Router just did an unasked for re-sync, a the RSL plot below is the result!