Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Hitman]

Hi,

Just lost around 25Mb (usually had 65/18) off my infinity speed so I've unlocked my modem to get my stats, factory defaulted the modem and found high values for interleaving (using DSL Stats 5.3) 1075/DL 489/UL, see stats below.

Just wondered what would cause this, I have a BT engineer coming Monday to check my line although a quite line test I can only hear a slight hiss any advice would be appreciated and any advice on settings I need for BT infinity or is the default fine?

I can supply more info if required.

Thanks.

xdslcmd info --stats
xdslcmd: ADSL driver and PHY status
Status: Showtime
Retrain Reason:   1
Last initialization procedure status:   0
Max:   Upstream rate = 18537 Kbps, Downstream rate = 45632 Kbps
Bearer:   0, Upstream rate = 13338 Kbps, Downstream rate = 38946 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):    6.6       6.5
Attn(dB):    14.7       0.0
Pwr(dBm):    13.5       6.6
         VDSL2 framing
         Bearer 0
MSGc:      18      26
B:      63      21
M:      1      1
T:      64      64
R:      14      10
S:      0.0523      0.0524
L:      11936      4883
D:      1075      489
I:      78      32
N:      78      32
         Counters
         Bearer 0
OHF:      22924862      1106544
OHFErr:      79      2
RS:      106531253      4002391
RSCorr:      66161      42174
RSUnCorr:   1977      0

         Bearer 0
HEC:      490      0
OCD:      22      0
LCD:      22      0
Total Cells:   26865846      0
Data Cells:   669565235      0
Drop Cells:   0
Bit Errors:   0      0

ES:      49      143
SES:      22      0
UAS:      486      464
AS:      57753

         Bearer 0
INP:      5.00      4.00
INPRein:   0.00      0.00
delay:      14      6
PER:      2.51      6.73
OR:      76.21      38.00
AgR:      39022.13   13376.00

Bitswap:   27124/27125      6275/6278

Total time = 1 days 15 hours 58 min 48 sec
FEC:      239464      367851
CRC:      7576      252
ES:      49      143
SES:      22      0
UAS:      486      464
LOS:      2      0
LOF:      19      0
LOM:      0      0
Latest 15 minutes time = 13 min 48 sec
FEC:      212      504
CRC:      0      0
ES:      0      0
SES:      0      0
UAS:      0      0
LOS:      0      0
LOF:      0      0
LOM:      0      0
Previous 15 minutes time = 15 min 0 sec
FEC:      8712      547
CRC:      18      0
ES:      3      0
SES:      0      0
UAS:      0      0
LOS:      0      0
LOF:      0      0
LOM:      0      0
Latest 1 day time = 15 hours 58 min 48 sec
FEC:      66161      41680
CRC:      79      2
ES:      11      2
SES:      0      0
UAS:      0      0
LOS:      0      0
LOF:      0      0
LOM:      0      0
Previous 1 day time = 24 hours 0 sec
FEC:      62310      253639
CRC:      3998      109
ES:      11      8
SES:      11      0
UAS:      428      417
LOS:      1      0
LOF:      9      0
LOM:      0      0
Since Link time = 16 hours 2 min 33 sec
FEC:      66161      42174
CRC:      79      2
ES:      11      2
SES:      0      0
UAS:      0      0
LOS:      0      0
LOF:      0      0
LOM:      0      0
#

Discovery Phase (Initial) Band Plan
US: (7,32) (871,1205) (1972,2782)
DS: (33,859) (1216,1961) (2793,3970)
Medley Phase (Final) Band Plan
US: (7,32) (871,1205) (1972,2782)
DS: (33,859) (1216,1961) (2793,3970)
     VDSL Port Details        Upstream        Downstream
Attainable Net Data Rate:       18615 kbps          45860 kbps
Actual Aggregate Tx Power:          6.7 dBm           13.5 dBm
====================================================================================
  VDSL Band Status   U0   U1   U2   U3   U4   D1   D2   D3
  Line Attenuation(dB):   3.0   18.7   30.2   N/A   N/A   9.6   24.7   39.9
Signal Attenuation(dB):   3.0   18.0   29.2   N/A   N/A   12.7   24.5   39.9
        SNR Margin(dB):   6.9   6.5   6.5   N/A   N/A   6.7   6.6   6.7
         TX Power(dBm):   -7.0   -25.1   6.4   N/A   N/A   10.4   8.0   6.8
#

[burakkucat]

Welcome to the Kitz forum. 

I won't attempt to analyse your statistics . . . there are others here who can do it far better than I.

With regards to the HG612 configuration, it is advisable to leave everything "as is" once the device has been unlocked to allow the harvesting of statistics.

[kitz]

Quick analysis from me

>> INP:      5.00      4.00
>> delay:      14      6

Thats pretty high, I cant recall seeing INP 5 before. Delay of 14 is high interleaving. This suggests that the DLM is seeing a lot of noise.  Both of these will reducing the speed at which you can sync at.

Despite high INP and Interleaving you are still seeing a fair few errors.  That line has only been up about 16hrs, yet theres some UAS suggesting probable multiple resyncs.

Judging from your errors (CRC/SES/OCD)... Im going out on a total limb here and gut feeling from reading line stats for many years that your line is experiencing sudden noise bursts, possibly even voice related (ie when phone rings/voice in use).  I stress that the later is just a guess, however I do suspect SHINE whether that is from voice or some other source I dont know. If its voice then its a possible HR Fault. 

We'd need to see some DSLstat graphs to say for sure.

PS - Im moving this thread to FTTC, where its more likely to be seen by those who read fttc stats.

[WWWombat]

I concur with Kitz - this seems to be a very heavy level of intervention by DLM, on both upstream and downstream, suggesting it has been trying a few settings to get rid of errors or resyncs.

The current setting has left 11 ES over 16 hours, which seems like a small level, and alone would suggest that things are OK like this. However, I too am worried about the size of the UAS counter (the number of seconds the service has been unavailable).

Definitely worth having one of the stats harvesting programs running - preferably before the engineer arrives, so we can gauge the effectiveness of the visit. See Kitz's link, and this part of the forum

[Hitman]

Thanks a lot everyone, that's a big help!

I thought as much to leave the settings at default, this does seem the general rule from the web.

This line has been quite problematic over the years almost an engineer visit yearly, when I first got infinity my stats were 72/18 which has steadily dropped and settled on 65/18 for the last 2 years, which I was told this is normal due to the extra infinity customers that had been added to the cabinet, I think I was one of the few early adopters.

I've been put on a new pair nearly each time an engineer has been out and the last visit the engineer said I was now on the last copper pair and if it went again in the future BT would have to install a new multipair in the street, there's a few inspection holes between me and the cabinet (i'm about 80m away from the fibre cabinet) and most problems have stemmed at the first inspection hole which is 10m away.

I'm getting mixed info from BT which makes me a bit nervous as my first call to them with this current problem the operator did some checks and said there was a fault within the local exchange and the engineer probably won't need access to my property although she took my mobile number just in case, then a day later I got a voice mail/missed call saying the problem was not in the exchange but at my property, so I called them back, this operator had me do some checks (unplugging phone/modem etc..) and said the engineer will need access Monday which was fine, however after a few confirmation txt's I've just got a txt saying the fault is within my property with the equipment and if it's my own cause I will be charged £129  , although this is a similar response I've had to earlier faults I've reported in the past, so I think i'll flash back my modem beforehand just in case.

I'll get some more stats up today as requested, thanks again for the responses appreciated!

Edit: the 16hrs uptime relates to when I unlocked my modem.

[Hitman]

Ok I remembered I took snapshots yesterday so have some details now, I've started dslstats again in the background so will leave this running for today to get a longer run on stats and can post these up tonight if these aren't a long enough run.

[kitz]

Thanks for the graphs.

The one Im most interested in is SNR Margin (dB).  I'd like to see that run for a longer period. From that time frame it looks perfectly normal but I suspect that at some point we may see spiking.

I'd also like to see if anything happens to the SNR margin when (a) the phone is in use (b) the phone rings.   

With regard to (b) you may need a careful bit of timing.  DSLstats usually samples every minute, so what I do is watch when it samples, so if it samples at 40 seconds past the minute, then using my mobile I will dial my landline at about 30/35 seconds past and let it ring.  BT's ringback service wasnt working last time I checked.

Re the QLN graphs etc, you can adjust the scale of those so that the whole spectrum fits into the one graph (see screen cap below)

[kitz]

when I first got infinity my stats were 72/18 which has steadily dropped and settled on 65/18 for the last 2 years, which I was told this is normal due to the extra infinity customers that had been added to the cabinet, I think I was one of the few early adopters.

Yes thats true.  Most people will lose speed the more users there are on the cab.  I dont think its unfair to say that the average figure for the forum regs is now in the region of at least 15Mb, probably more like 20Mb just through crosstalk.

so I think i'll flash back my modem beforehand just in case.

Its highly unlikely that the engineer would notice, they just look at the box and test with their own equipment. Unless you tell them, they wont know.
In fact some of us on here have been quite open with the engineers and say we are getting stats from them..  I had one engineer who looked at my graphs and said he was going home that evening and unlocking his own. However I wouldn't mention it unnecessarily unless we can find a good example of SNRm spiking.

[Hitman]

Thanks again Kitz, I've done an electrical switch on with various appliances/lights etc.. first thing, this did not effect the stats at all then did a phone test and this spikes a bit but nothing I would say as substantial, tried 3 phone tests - off hook test, phone ring and phone answer - see the 3 spikes midway on the graph...

I'll change the settings as you advise and leave it running.



This is the adjusted QLN stats...

[burakkucat]

If possible, may we please see the Hlog graph expanded to the same X-axis scale as the previously shown QLN graph, above?

If you are any way concerned about having an unlocked HG612 when the engineer visits, just stop the statistics harvesting program and remove the Ethernet patch cable from the LAN2 port of the device before she/he arrives. 

[Hitman]

If possible, may we please see the Hlog graph expanded to the same X-axis scale as the previously shown QLN graph, above?

If you are any way concerned about having an unlocked HG612 when the engineer visits, just stop the statistics harvesting program and remove the Ethernet patch cable from the LAN2 port of the device before she/he arrives. 

Hi,

Hlog attached and put the latest snr margin up (5hrs logging), the monitoring has been on all day but nothing showing as serious that I can see....

[NewtronStar]

If you are any way concerned about having an unlocked HG612 when the engineer visits, just stop the statistics harvesting program and remove the Ethernet patch cable from the LAN2 port of the device before she/he arrives. 

which i forgot to do and was quizzed by the engineer it felt like the spanish inquisition   

[Hitman]

I'll make sure mine looks perfectly normal 

Thanks.

[burakkucat]

Hlog attached and put the latest snr margin up (5hrs logging), the monitoring has been on all day but nothing showing as serious that I can see....

Thank you. The Hlog graph looks good and the SNRM graph is, to all intents and purposes, a pair of straight lines. So nothing odd there.

[Hitman]

Update on progress,

SNR went up late last night as it had started raining earlier, see graphs below...

BT Engineer has just left and he said he's getting a battery fault, which is voltage leakage from another line and suspects an inspection hole is possibly water logged and shorting the pairs, he said he's calling in for a ground engineer to attend and inspect all the covers and isolate the water damaged pair cable and/or inspection hole, hopefully he will get here today.