I’m still having problems with line #3 after the engineer’s visit this week, but lesser ones. Big noise spikes mean that connections are being broken and sync rates need to be reduced to allow for this.
I have captured an SNRM-vs-time graph - see
https://ibb.co/zPkZ4vZ.
But in addition, the
upstream of line 3 continues to be a big problem, with more than 9dB [!] of noise variation through the day, in the same pattern that is now becoming familiar: flat noise-vs-time graphs, with vertical cliffs between levels. The upstream high SNRM level over the last 24 hrs was 10.5dB and the low level was 1.6dB, with a target SNRM set at 6dB.
There is one difference from the earlier picture though: the appearance of spikes overlaid on the ‘plateaux’ graph.
On the SNRM-vs-time graph there are huge
upward-going spikes - ie. noise disappearing for very short periods. These are the opposite of a problem of course. A new feature. Just an oddity. God only knows why such ultra-quiet brief moments are happening.
More importantly, there are also downward spikes in the upstream, ~2.5 dB high, and these are wrecking upstream reliability, since during part of the day the upstream SNRM is only 1.6dB, for many hours, until the noise level drops and the graph goes up to the next plateau. The modem chooses to hang on in there and does not drop the link, despite the downward-going spikes bring high enough to make the upstream SNRM go negative for a very short period.
Regarding the graph, unfortunately the
y-axis scale is very unhelpful, as the
y-axis max is far too high making it difficult to read
y-values off accurately. The big upward-going spikes are creating crazy all-time high values and so stretching the y-axis range.
My main requests though:
I got the following stats from the modem and I would be very grateful if (i) someone would read them for me and comment. I would like to know what you think in general and (ii) specifically just how bad you think the error counts are. Also, I’m ashamed to say that, due to the once more increased pain drugs, I have forgotten the meaning of some of the fields in these reports, and (iii) I would be very grateful if some kind folks would go through the exact definitions of the various fields relating to
errors. (I can look up the framing-related variables in G.992.3.) Here’s line #3:
VMG1312-B10A
Login: admin
Password:
> xdslctl info --show
xdslctl: ADSL driver and PHY status
Status: Showtime
Last Retrain Reason: 8000
Last initialization procedure status: 0
Max: Upstream rate = 280 Kbps, Downstream rate = 2940 Kbps
Bearer: 0, Upstream rate = 344 Kbps, Downstream rate = 2606 Kbps
Link Power State: L0
Mode: ADSL2 Annex A
TPS-TC: ATM Mode(0x0)
Trellis: U:ON /D:ON
Line Status: No Defect
Training Status: Showtime
Down Up
SNR (dB): 6.0 6.1
Attn(dB): 64.5 40.7
Pwr(dBm): 17.8 12.4
ADSL2 framing
Bearer 0
MSGc: 63 12
B: 76 4
M: 2 16
T: 1 8
R: 16 16
S: 1.8554 7.2453
L: 733 106
D: 1 4
Counters
Bearer 0
SF: 36390 35897
SFErr: 0 0
RS: 1255461 323009
RSCorr: 2 0
RSUnCorr: 0 0
ReXmt: 0 0
ReXmtCorr: 0 0
ReXmtUnCorr: 0 0
Bearer 0
HEC: 0 0
OCD: 0 0
LCD: 0 0
Total Cells: 3600553 473763
Data Cells: 18185 5381
Drop Cells: 0
Bit Errors: 0 0
ES: 4181 5031
SES: 2543 709
UAS: 24374 22609
AS: 587
Bearer 0
INP: 29.00 2.00
INPRein: 0.00 0.00
delay: 8 7
PER: 16.00 16.30
OR: 34.49 8.83
AgR: 2628.53 351.95
Bitswap: 99/106 12/12