Print

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

[Bald_Eagle1]

I have been remotely monitoring a VDSL2 connection & attach a montage of 8 days of ongoing stats.

Do any of you have any suggestions as to what may be causing such high error counts, only subsiding following a connection resync, sometimes 'on the fly' (Retrain Reason 2) & sometimes by what appears to be a Loss of Signal (Retrain Reason 0)?

FWIW, the stats have been obtained via a Huawei HG612 modem connected to an ECI DSLAM.

The connection appears to be currently capped at 32.4 Mbps DS sync speed, with relatively low US sync speed.

Although not shown in the montage, DS Interleaving depth is currently at 2379.

Also:-

INP:      9.50      0.00
PER:      2.63      16.25
delay:   16.00      0.00
OR:      72.76      19.20

[ryant704]

Looks like external interference and rather bad considering you have a INP of 9.5 :/

Retrain Reason 0 is used when the user resets, 2 is when the DLM intervened with your line.

Any major works happening near you? Just look for REIN, EI, RFI in and out of the home.

QLN and Hlog could be of some help.

[Bald_Eagle1]

Looks like external interference and rather bad considering you have a INP of 9.5 :/

Retrain Reason 0 is used when the user resets, 2 is when the DLM intervened with your line.

Well, yes. But the user is adamant they haven't reset the connection or reboted the modem.
The monitoring equipment is connected directly to the modem's LAN2 port to which the user doesn't really have access.

These are apparently the retrain reasons for VDSL2 connections:-

Retrain Reason Los Detector              0
Retrain Reason Rdi Detector              1
Retrain Reason Negative Margin           2
Retrain Reason Too Many Us FEC           3
Retrain Reason CReverb1 Misdetection     4


Reasons 0 & 2 are quite common, 1, 3 & 4 quite rarely seen.

Any major works happening near you? Just look for REIN, EI, RFI in and out of the home.

QLN and Hlog could be of some help.

I am starting to suspect an intermittent HR or capacitance issue, somewhere in the D-side cabling, only clearing temporarily following a reason 2 or reason 0 resync.

I haven't seen much of the way of changes in Hlog or QLN graphs over the monitoring period.

The connection's DS sync speed was originally capped at a ridiculously low 15 Mbps when monitoring commenced.
A DLM reset was arranged, but until the 'inbterference' issues are resolved, there is every chance that DLM will cap sync speeds at lower & lower levels again.

[NewtronStar]

The SNR is kind of low on the DS 3.9 and 6.0 should be the norm and Interleaving is high at 2300 it could be consistent noise on the line, have they done the Quiet Line test and when did the user first notice these issues as the DLM should adjust in 2 weeks if the noise has gone !

[Bald_Eagle1]

Quiet Line Test seems fine.

The user noticed low speeds from day one.

The connection was capped/banded at 15 Mb when 24/7 monitoring commenced 4th June.

DLM was reset 6th June at 40 Mbps, but US sync speed has remained unusually low throughout.
That didn't last for long though.

The D-side connection is approximately 820m in length.


04/06/2013 14:43 - RESYNC detected (DS 14999 Kbps, US 4415 Kbps)
05/06/2013 06:53 - RESYNC detected (DS 14999 Kbps, US 4252 Kbps)
06/06/2013 10:28 - RESYNC detected (DS 39997 Kbps, US 4677 Kbps)
07/06/2013 05:58 - RESYNC detected (DS 39997 Kbps, US 4812 Kbps)
07/06/2013 06:22 - RESYNC detected (DS 39997 Kbps, US 4745 Kbps)
08/06/2013 04:40 - RESYNC detected (DS 37880 Kbps, US 4723 Kbps)
08/06/2013 20:00 - RESYNC detected (DS 36530 Kbps, US 4210 Kbps)
09/06/2013 05:36 - RESYNC detected (DS 37803 Kbps, US 4305 Kbps)
09/06/2013 08:25 - RESYNC detected (DS 37991 Kbps, US 4181 Kbps)
10/06/2013 05:55 - RESYNC detected (DS 34999 Kbps, US 4210 Kbps)
10/06/2013 07:47 - RESYNC detected (DS 34999 Kbps, US 4291 Kbps)
17/06/2013 16:50 - RESYNC detected (DS 34999 Kbps, US 4298 Kbps)
18/06/2013 16:51 - RESYNC detected (DS 34999 Kbps, US 4497 Kbps)
21/06/2013 05:51 - RESYNC detected (DS 32399 Kbps, US 4490 Kbps)
21/06/2013 08:50 - RESYNC detected (DS 32399 Kbps, US 4348 Kbps)
21/06/2013 12:26 - RESYNC detected (DS 32399 Kbps, US 4341 Kbps)
22/06/2013 04:34 - RESYNC detected (DS 32399 Kbps, US 4490 Kbps)
23/06/2013 21:51 - RESYNC detected (DS 31585 Kbps, US 4475 Kbps)
25/06/2013 04:15 - RESYNC detected (DS 32982 Kbps, US 4511 Kbps)
25/06/2013 10:41 - RESYNC detected (DS 34486 Kbps, US 4291 Kbps)
26/06/2013 05:26 - RESYNC detected (DS 34999 Kbps, US 4426 Kbps)
26/06/2013 05:34 - RESYNC detected (DS 34999 Kbps, US 4341 Kbps)
30/06/2013 05:00 - RESYNC detected (DS 32399 Kbps, US 4348 Kbps)
30/06/2013 05:31 - RESYNC detected (DS 32399 Kbps, US 4210 Kbps)
04/07/2013 04:18 - RESYNC detected (DS 32399 Kbps, US 4557 Kbps)
04/07/2013 12:53 - RESYNC detected (DS 32398 Kbps, US 4380 Kbps)
05/07/2013 22:23 - RESYNC detected (DS 28730 Kbps, US 4365 Kbps)
05/07/2013 22:50 - RESYNC detected (DS 28875 Kbps, US 4497 Kbps)
06/07/2013 06:15 - RESYNC detected (DS 29998 Kbps, US 4295 Kbps)


Some of those resyncs reported Retrain Reason 0 & some reported Retrain Reason 2.
Only one of them was user initiated.


The 2Wire router that has been used for a while was replaced by the original HH3 at 17:37 yesterday.
The connection/DLM didn't seem too happy about it though.

It now seems to be banded at 30 Mbps.

[NewtronStar]

it looks does look strange when they have 3 bands active on the medley Final Band Plan on the US
I have only 2 Bands active on the US but getting higher sync's than the user but still you said they were 820 meters from Cab thats quite good, as for me I would say I am 725 meters from Cab and you can look at my current stats as the DLM is happy again with my line.

[ryant704]

You can see the interference is affecting the higher VDSL2 tones, what ISP is he with?

You can do what I did and knock on peoples house further down the road who are with the same ISP and send him an email. Though can be rather difficult if people don't agree, you could try and request for a REIN engineer though you will probably have to wait 1 - 2 weeks.

How is DS throughput, can he still get close to his IPP?

I'm assuming the Interference should be rather obvious, trains tracks, electric fence, Sodium chipsets etc but that would make it too easy!

It may be worth getting a ECI Modem, personally I've never had DLM intervention since I've plugged the HG612 in. When using an ECI it use to intervene every 7 days.

On the final note has a 'lift and shift' be done yet? I doubt it will resolve anything but might be worth putting a request in.

[Bald_Eagle1]

An engineer's visit is earmarked for this afternoon where the deteriorated stats etc. will be discussed, hopefully resulting in a lift & shift or at least checking/remaking external joints as the connection is definitely experiencing increasingly random disconnections (coinciding with the warmer weather?).


As a result of these resyncs, sync speeds are now down to 17460 Kbps DS & 3688 Kbps US.

I won't be in attendance for the engineer's visit as the site is a very long way from my home, but someone local with a good understanding of xDSL matters will be attending on behalf of the user.

[NewtronStar]

external joints as the connection is definitely experiencing increasingly random disconnections (coinciding with the warmer weather?).

The Dreaded Expansion & Contraction of metal during High and Low temperatures it still amazes me the stuff in the cabs can survive one years worth of our up & down climate 

Looking forward to see the stats after L&S 

[Bald_Eagle1]

The engineer visited yesterday & swapped the pairs from the PCP toward the end user.

As one of the poles was classed as being in a dangerous condition he was unable yeterday to swap the last 2 spans of overhead pairs.

DLM was reset (on fastpath), but we THINK it may still be banded at 35Mb for the moment.

There has therefore been some improvement in sync /attainable speeds.
The last 24 hours of ongoing stats do still indicate an issue though - possibly within the last 2 spans of as yet not swapped overhead cable?

There were also 2 overnight resyncs:-

12/07/2013 01:56 - RESYNC detected (DS 33908 Kbps, US 6144 Kbps) - Retrain Reason 2
12/07/2013 02:59 - RESYNC detected (DS 34187 Kbps, US 5892 Kbps) - Retrain Reason 0

Hopefully the stats are sufficiently convincing to encourage an engineer to obtain a cherry picker & swap the final 2 spans & maybe if the issues still continue, arrange a DSLAM port lift & shift.

If all that was done & the issues still continue a detailed REIN investigation would be needed.

This morning's snapshot montage & the latest 24 hours of ongoing stats are attached for reference (showing the results of yesterday's partial line swapping).

 

[Chrysalis]

be happy he got the pair swap, I cant see how lift and shift can be more effective than a pair swap and now days to get a pair swap is extremely diffilcult.