I can't explain these stats

Last night, while I was out, we had a power cut of half an hour or so. See snapshot BQM.

I didn't remember to check my sync until about an hour ago and was completely baffled.

# xdslcmd info --show
xdslcmd: ADSL driver and PHY status
Status: Showtime
Retrain Reason: 0
Last initialization procedure status: 0
Max: Upstream rate = 17930 Kbps, Downstream rate = 58924 Kbps
Bearer: 0, Upstream rate = 16999 Kbps, Downstream rate = 64514 Kbps

Link Power State: L0
Mode: VDSL2 Annex B
VDSL2 Profile: Profile 17a
TPS-TC: PTM Mode(0x0)
Trellis: U:ON /D:ON
Line Status: No Defect
Training Status: Showtime
Down Up
SNR (dB): 3.0 4.7
Attn(dB): 19.4 0.0
Pwr(dBm): 13.6 7.5
VDSL2 framing
Bearer 0
MSGc: 18 26
B: 239 237
M: 1 1
T: 22 39
R: 0 16
S: 0.1184 0.4447
L: 16216 4569
D: 1 1
I: 240 127
N: 240 254
Counters
Bearer 0
OHF: 40357782 2777766
OHFErr: 83869 30
RS: 0 3233242
RSCorr: 0 2398
RSUnCorr: 0 0

Bearer 0
HEC: 29505 0
OCD: 652 0
LCD: 652 0
Total Cells: 1225131928 0
Data Cells: 9143908 0
Drop Cells: 0
Bit Errors: 0 0

ES: 4246 26
SES: 165 0
UAS: 25 25
AS: 79153

Bearer 0
INP: 0.00 0.00
INPRein: 0.00 0.00
delay: 0 0
PER: 1.96 4.35
OR: 97.89 58.80
AgR: 64611.61 17058.12

Bitswap: 66571/66571 1576/1577

#

I did three wired speed tests:-

Download speed achieved during the test was - 55.01 Mbps
For your connection, the acceptable range of speeds is 40 Mbps-62.45 Mbps .
Additional Information:
IP Profile for your line is - 62.45 Mbps

Upload speed achieved during the test was - 14.22Mbps
Additional Information:
Upstream Rate IP profile on your line is - 20 Mbps

thinkbroadband test 1
thinkbroadband test 2

For bearer to be so far above Max, and the noise margin 3dB now, (3.1dB an hour ago when the tests were done), is really weird. Unfortunately I haven't yet set up monitoring software on my spare machine and don't want it on this one.

The lousy first tbb test and dodgy second one may be a Plusnet problem or may not be. I've never seen anything like either of them on my connection before.

Anyone any ideas? The sync's in my sig are about the highest I normally have, and were correct before the power cut.

Just reboot your router.....when your and everybody else on your cab's power was restored; there was no noise/crosstalk on the line so it could connect at a very high speed....because of the proliferation of BTO modems with very similar training times this is exacerbated.

Ian

I had something very similar happen on my line, though no power cut.

I've got monitoring software on my line and you can see it initially syncs with an 11db margin and a few seconds later it's down at 3db which went down to 2.6db.

I kept the line at that sync rate for 2 days, but the errors on the line were threatening to DLM me into oblivion so I rebooted the modem... It's back to normal sync speed now. Still have no idea why it decided 3db was a good idea!

If I were you I'd monitor those error rates and reboot now or tomorrow morning if you think it can wait.

Edited to correct poor grammer!

Edited by timl (Fri 13-Feb-15 20:53:59)

Reboot it ASAP. Your ES count might already be high enough to trip DLM overnight.

More details shortly....

As Ian said, all the modems resynced in parallel.

That means they all did a quiet line noise check, just listening for noise. Because no modems were squawking, they all saw less crosstalk, so less noise. They all had high SNR levels.

Thus all modems start up, in parallel, with higher than normal bit loading. The sudden crosstalk from all these modems impacts your line - everyone's line really - in the normal way as extra noise, so reducing SNRs. Some modems will lose sync at this point, and retrain, while others hold on.

Those that hold on will be receiving the analogue signal at too low an SNR to get through cleanly, leading to errors and a high CRC rate. That, in turn, leads to a high ES count over each 24 period, which DLM monitors.

Plusnet uses the speediest profile for DLM, which uses a threshold of (we believe) 2,880 ES's per 24 hours. Above that, DLM will intervene.. and you've reached 4226 in 21 hours, although some of that will be for yesterday.

Your saving grace might be that DLM detects the outage as a wide-area event, and ignores the error counts on all the lines in the cab. It depends how widespread the power cut was.

Bah !

After Ian's explanation of how it could occur, and your warning, I agreed with both of you so just rebooted.

I'm not sure it would have caused a DLM problem. Counts immediately before the reboot below, with copy values (when different) from the OP in red. Just to make the increases easier for you all to see. New stats at the end.

Thanks for the help. It was fun while it lasted - I shall always wish I'd risked letting it run.

Counters
Bearer 0
OHF: 40357782 47453230 2777766 2117862 ???? Lower?
OHFErr: 83869 119236 3040
RS: 0 3233242 3838930
RSCorr: 0 2398 2603
RSUnCorr: 0 0

Bearer 0
HEC: 29505 37211 0
OCD: 652 849 0
LCD: 652 849 0
Total Cells: 1225131928 2950757305 0
Data Cells: 9692905 0
Drop Cells: 0
Bit Errors: 0 0

ES: 4246 6272 26 36
SES: 165 223 0
UAS: 25 25
AS: 79153 93070

Bearer 0
INP: 0.00 0.00
INPRein: 0.00 0.00
delay: 0 0
PER: 1.96 4.35
OR: 97.89 58.80
AgR: 64611.61 17058.12

Bitswap: 66571/66571 1576/1577
Bitswap: 77982/77982 2501/2502
************************************************************************
New stats:-

# xdslcmd info --show
xdslcmd: ADSL driver and PHY status
Status: Showtime
Retrain Reason: 0
Last initialization procedure status: 0
Max: Upstream rate = 15183 Kbps, Downstream rate = 56296 Kbps
Bearer: 0, Upstream rate = 15194 Kbps, Downstream rate = 56299 Kbps

Link Power State: L0
Mode: VDSL2 Annex B
VDSL2 Profile: Profile 17a
TPS-TC: PTM Mode(0x0)
Trellis: U:ON /D:ON
Line Status: No Defect
Training Status: Showtime
Down Up
SNR (dB): 6.2 6.1
Attn(dB): 19.5 0.0
Pwr(dBm): 13.5 7.5
VDSL2 framing
Bearer 0
MSGc: 18 25
B: 239 237
M: 1 1
T: 64 64
R: 0 16
S: 0.1357 0.4983
L: 14152 4078
D: 1 1
I: 240 127
N: 240 254
Counters
Bearer 0
OHF: 22485 6160
OHFErr: 0 0
RS: 0 394045
RSCorr: 0 0
RSUnCorr: 0 0

Bearer 0
HEC: 0 0
OCD: 0 0
LCD: 0 0
Total Cells: 5274903 0
Data Cells: 540 0
Drop Cells: 0
Bit Errors: 0 0

ES: 0 0
SES: 0 0
UAS: 23 23
AS: 50

Bearer 0
INP: 0.00 0.00
INPRein: 0.00 0.00
delay: 0 0
PER: 2.17 8.00
OR: 88.10 30.98
AgR: 56387.73 15224.99

Bitswap: 24/24 1/1

Edited by RobertoS (Fri 13-Feb-15 23:17:37)

That explanation makes complete sense for a power cut were all the lines all go out at the same time. No crosstalk so much higher nose margin, but it doesn't explain why my HG612 decided to reboot itself at 3pm and sync at 3db, when all the other lines retained their sync.

It's as if the HG612 was ignoring crosstalk for the start of it's re-sync. Would that be possible?

Did your modem reboot, or just resync?

A resync triggered by a reset or reboot of the DSLAM would look similar, I guess. Harder to say what is happening to a modem that spontaneously reboots.

One way to check is to use an unlocked modem, and one of the programs that monitors statistics. With that you can see the QLN graph; resyncs with low noise make this graph look very different.

HG612 Modem Stats is active on the line. It logged the following in the modem_stats.log file. I wonder if it was perhaps someone in the cabinet briefly disturbed some wires.... which triggered this... but then my other FTTC line connected to the same cabinet (but probably a different module,) didn't reset.

09/02/2015 14:37 67660 20000 5.9 10.0 13.5 1.3 66236 25872 PTM 11038 91 0 1949 13516 0 2078134 0 0 0 0 577.259 0 1949 0 1164246533 2373135 33615 11038 91 0 0 984833 0 0 0 0 1 1 2058 78 7 0 0 0 0 636155 9791 0 0.9 11.1 19.0 0.0 7.9 17.3 26.6 0.9 11.0 19.0 0.0 9.6 17.1 26.6 9.0 9.8 10.1 0.0 5.9 5.9 5.8 -4.9 -44.8 0.2 0.0 10.4 7.4 7.4 Showtime 0 0 L0 17a ON ON No Showtime 12.8 0.0 239 236 1 1 21 5 0 16 0.1129 0.3771 17008 5410 259 0 259 0 3978312443 0 275254454 0 0 0 0 26 26 0.00 0.00 1.78 6.15 0.00 0.00 107.56 202.87 1100799 81 VDSL2 42 0 0 0 0.00 0.00 0.00 0.00 0 0 0 0 0 0 0 0 0 0 0 0 0 0
09/02/2015 14:38 67660 20000 5.9 10.0 13.5 1.3 66236 25845 PTM 11038 91 0 1949 13516 0 2078194 0 0 0 0 577.276 0 1949 0 1164280148 2382936 33615 11038 91 0 0 1621571 0 0 0 0 1 1 2058 78 7 0 0 0 0 636738 9801 0 0.9 11.1 19.0 0.0 7.9 17.3 26.6 0.9 11.0 19.0 0.0 9.6 17.1 26.6 9.6 9.8 10.1 0.0 5.9 5.9 5.8 -4.9 -44.8 0.2 0.0 10.4 7.4 7.4 Showtime 0 0 L0 17a ON ON No Showtime 12.8 0.0 239 236 1 1 21 5 0 16 0.1129 0.3771 17008 5410 259 0 259 0 3986115669 0 275270921 0 0 0 0 26 26 0.00 0.00 1.78 6.15 0.00 0.00 107.56 202.87 1100844 81 VDSL2 45 0 0 0 0.00 0.00 0.00 0.00 0 0 0 0 0 0 0 0 0 0 0 0 0 0
09/02/2015 14:39 79987 20000 10.1 8.5 13.4 0.4 95418 25595 PTM 0 0 0 0 0 0 10 0 0 0 0 0.003 0 0 0 6046 1484 0 0 0 0 0 96417 0 0 0 0 1 1 0 0 0 0 0 0 0 0 0 0 0.9 11.1 18.8 0.0 7.8 17.1 26.5 0.9 11.0 18.8 0.0 9.6 17.0 26.5 14.8 15.4 6.0 0.0 0.0 0.0 0.0 -4.7 -36.2 -1.2 0.0 10.4 7.3 7.4 Showtime 0 0 L0 17a ON ON No Showtime 12.7 0.0 239 236 1 1 23 5 0 16 0.0955 0.3771 20104 5410 0 0 0 0 1498795 0 161 0 0 0 0 26 26 0.00 0.00 1.65 6.15 0.00 0.00 116.09 202.87 0 6 VDSL2 0 0 0 0 0.00 0.00 0.00 0.00 0 0 0 0 0 0 0 0 0 0 0 0 0 0
09/02/2015 14:40 79987 20000 3.0 8.2 13.4 0.4 93992 25404 PTM 0 0 0 0 0 0 70 0 0 0 0 0.019 0 0 0 42324 11281 36278 0 0 0 0 732943 0 0 0 0 1 1 0 0 0 0 0 0 0 636526 9797 0 0.9 11.1 18.8 0.0 7.8 17.1 26.5 0.9 11.0 18.8 0.0 9.6 17.0 26.5 11.7 13.2 6.2 0.0 3.1 3.0 2.9 -4.7 -36.2 -1.2 0.0 10.4 7.3 7.4 Showtime 0 0 L0 17a ON ON No Showtime 12.7 0.0 239 236 1 1 23 5 0 16 0.0955 0.3771 20104 5410 0 0 0 0 10723915 0 17638 0 0 0 0 26 26 0.00 0.00 1.65 6.15 0.00 0.00 116.09 202.87 14 25 VDSL2 14 19 0 0 0.00 0.00 0.00 0.00 0 0 0 0 0 0 0 0 0 0 0 0 0 0