BT Line Fault (a perspective on Plusnet Customer Service)

[#1]

So far my router has had one dropped connection about an hour after restarting, following the line repair.  Could this be the result of Ben bumping the Plusnet IP Profile?  It happened at about that time.  No problems overnight, so this morning I asked my router to disconnect and reconnect.  No change to the sync rate and noise margin but my BT IP Profile has risen to 4500 kbps.  Again the Plusnet profile seems to be lagging behind this as the measured download speed was about half that.  As it's the weekend with nobody from Plusnet to intervene, I'll see how long it takes for Plusnet's automated systems to catch up.

[#2]

It's now 24 hours later.  My router re-synced very close to the stroke of midnight and my measured download speed is now pushing 4000 kbps (everything else the same).  I presume this means:

• Normal adjustments to Plusnet IP Profiles happen at midnight


• The change forces a disconnection

[#3]

I have been advised - because I asked the question - that when Plusnet change their version of the IP profile it causes a disconnect from the internet but not from the broadband connection.
This is why my router log shows LCP up but Routerstats thinks I have been connected continuously as it seems to monitor the connection to the exchange not the connection to the ISP
The adjustments happen at random times looking at my router log

[#4]

Another update:
Voice line fault repaired last Friday.  No problems since except that my router hung a couple of times over the weekend (old age?).  Now on 85 hours of a stable connection.  My S/N ratio ("noise margin") is stuck at 15 dB whereas it used to be 6 dB so my download speeds are about two thirds of what they were.  All this is due to four days of a bad connection on the line - nothing whatsoever to do with Plusnet.
I think as my download speed is now acceptably high I will bear with this for a while and not ask Plusnet to intervene.  I want to see if the BT DLM does it's job and eventually reduces the noise margin.    

[#5]

Quote from: ReedRichards

I want to see if the BT DLM does it's job and eventually reduces the noise margin.

You may have a long wait, based on my experience with the DLM system.

[#6]

The worst case that I'm expecting is about 10 days to bring the noise margin down to 12 dB, another 10 days then 9 dB and another 10 days then 6 dB.  So 30 days in total!!!  Can it really be that bad? 

[#7]

I think the issue here is that the line needs to be constantly stable through all that time for the SNR to drop, although I may be wrong.

[#8]

you may be looking more at 90 days, in my experience.
Of course, a quick ask for a retraining restart means you'd be back to a 6db overnight, and the line would retrain over the next 10 days so it makes sense to ask for it.

[#9]

My line is very stable, except when it goes wrong! 
Point taken Barry, but as current speeds do not represent a major inconvenience I plan to wait out at least the first ten days to see what happens.  If I get fed-up of waiting I will ask Plusnet to intervene.

[#10]

Okay, it's now 16 days since the voice line fault was fixed.  My broadband connection has been completely solid, no dropped connections.  In the first couple of days I had some problems when my router hung and I had to power it down then back on.  Since then I've disconnected every 100 hours or so to see if there has been any change.  Today I disconnected the router for about 9 hours and when I reconnected the S/N margin had dropped from 15 dB to 12 dB and my sync rate had increased to 6176 kbps.  So that's 16 days for one 3 dB drop, two more adjustments to go before I get back to the norm. of 6 dB.  All as a consequence of four days of a bad connection.       

[#11]

I think I'd ask for a retrain!

[#12]

Yes, asking for a retrain makes sense on a personal level.  But only a tiny minority of broadband users know that they can do this, most just have to take whatever the system throws at them so I'm curious to know if the (DLM) system will eventually right itself, given enough time.  And although you get people on this forum with a huge range of broadband speeds but all complaining if their speed drops, I actually think that for 99% of the things I do online my current speed is quite adequate.
So no retrain yet, thanks.  I'll wait it out a bit longer.

[#13]

Since my SNR/"noise margin" managed to drop from 15 dB to 12 dB on 8th August it hasn't moved at all.  It always used to be 6 dB before I had the line fault in July and I would really like to understand the mechanisms by which it can get stuck like this.  Are there any tell-tale signs that something is wrong?  Below are all the stats my Netgear router will produce.  I switched it off the other day when a thunderstorm was threatening so the uptime is relatively short but other than that is there anything unusual?  Sorry about all the detail, I'm not sure which parts can be ignored:

adsl: ADSL driver and PHY status
Status: ShowtimeRetrain Reason: 0
Channel: INTR, Upstream rate = 448 Kbps, Downstream rate = 6208 Kbps
Link Power State: L0
Mode:                  G.DMT
Channel:                Interleave
Trellis:                OFF
Line Status:            No Defect
Training Status:        Showtime
                Down            Up
SNR (dB):      11.7            24.0
Attn(dB):      37.0            20.0
Pwr(dBm):      19.8            11.9
Max(Kbps):      6720            1100
Rate (Kbps):    6208            448
                        G.dmt framing
K:              195(0)          15
R:              14              16
S:              1              8
😧              16              2
                        ADSL2 framing
MSGc:          1              1
B:              195            15
M:              1              8
T:              1              1
R:              14              16
S:              1.0714          9.0000
L:              1568            128
😧              16              2
                        Counters
SF:            5780406        5780337
SFErr:          70              0
RS:            393067654              49132864
RSCorr:        3777457        36
RSUnCorr:      734            0
HEC:            60              0
OCD:            4              0
LCD:            0              0
Total Cells:    1438775295              0
Data Cells:    5989428        0
Drop Cells:    0
Bit Errors:    0              0
ES:            34              0
SES:            0              0
UAS:            46              0
AS:            98267
INP:            0.57            1.00
PER:            1.87            1.96
delay:          4.28            4.50
OR:            29.86          28.44
Bitswap:        0              0
Total time = 1 days 3 hours 19 min 4 sec
SF  = 5780406
CRC = 70
LOS = 0
LOF = 0
ES  = 34
Latest 1 day time = 3 hours 19 min 4 sec
SF  = 702572
CRC = 12
LOS = 0
LOF = 0
ES  = 5
Latest 15 minutes time = 4 min 4 sec
SF  = 14337
CRC = 0
LOS = 0
LOF = 0
ES  = 0
Previous 15 minutes time = 15 min 0 sec
SF  = 52923
CRC = 0
LOS = 0
LOF = 0
ES  = 0
Previous 1 day time = 24 hours 0 sec
SF  = 5077834
CRC = 58
LOS = 0
LOF = 0
ES  = 29
15 minutes interval [-30 min to -15 min] time = 15 min 0 sec
SF  = 52982
CRC = 7
LOS = 0
LOF = 0
ES  = 1
15 minutes interval [-45 min to -30 min] time = 15 min 0 sec
SF  = 52923
CRC = 0
LOS = 0
LOF = 0
ES  = 0
15 minutes interval [-60 min to -45 min] time = 15 min 0 sec
SF  = 52923
CRC = 1
LOS = 0
LOF = 0
ES  = 1
Showtime Drop Reason:  0
Last Retrain Reason:    0

[#14]

I don't know but from experience I would suspect that this may be the reason - why the corrected number should have any effect is beyond me
RS:            393067654              49132864
RSCorr:        3777457        36
RSUnCorr:      734            0

[#15]

Sorry, Oldjim, you're going to have to help me with the terminology and thought processes.
I take it RS stands for Reed Solomon, which is a type of error connection. 
Thus, in the downstream direction it seems that about 1 in every 100 packets (? is that the right unit?) require correction but are successfully corrected and one in every 500,000 packets (?) is not successfully corrected.  And you think the 1 in 100 ratio may be too high if used as a kind of line quality indicator?