Poor line revisited

[#1]

For a while I thought my line was becoming more stable but we seem to get more frequent dropouts and resyncs than before.
Each resync seems to cause various changes:
Att goes to 47 and we connect at G.992.5 with a sync of around 6600.
Att goes to 43 (or 43.5) and we connect at G.992.3 with a sync between 5700 and 5900.
[quote=Current stats]Line state Connected
Connection time 0 days, 0:21:21
Downstream 6,606 Kbps
Upstream 1,036 Kbps
ADSL settings
VPI/VCI 0/38
Type PPPoA
Modulation ITU-T G.992.5
Latency type Interleaved
Noise margin (Down/Up) 3.4 dB / 5.6 dB
Line attenuation (Down/Up) 47.0 dB / 24.9 dB
Output power (Down/Up) 0.0 dBm / 13.0 dBm
Loss of Framing (Local) 20
Loss of Signal (Local) 13
Loss of Power (Local) 0
FEC Errors (Down/Up) 1639005 / 4294967235
CRC Errors (Down/Up) 32 / 2147480000
HEC Errors (Down/Up) nil / 0
Error Seconds (Local) 2955
Even with the lowest sync we have recently had which was around 5500 we still get buffering at various times in the day when watching iPlayer, YouTube, etc.
I have done a quiet line test with no perceivable noise.
Any ideas what else can be done or looked into?

[#2]

Firstly, IIRC the attn will vary between ADSL2 and ADSL2+ as the range of tones varies and the higher frequencies affect (increase) the aggregate attn.
You should be able to "fix" the modem/router to use a particular "mode" instead of "auto" somewhere in the GUI, though you seem to have a problem at the (common) lower frequencies, judging by the CRC errors - definitely a problem somewhere.
Standard Qns. any extensions with ring wire connected? All phones filtered? Is it the same connected to the test socket via a filter?
Edit: Strangely, you seem to have 138.47 error seconds per minute  

[#3]

Not seen anywhere where the mode can be configured from within the BT Homehub 2.
No extensions, no ring wire connected and have a NT2000 negating the need for any separate filters.
Difficult to get at the NT2000 at present but will try the test socket later this evening or tomorrow.

[#4]

Hi Mav,
Just wondering how things went with connecting to the test socket and if you need any help with this.
Adam

[#5]

Hi Adam.
Was only able to connect to the test socket this morning.
[quote=Current Stats:]Line state Connected
Connection time 0 days, 0:03:27
Downstream 6,624 Kbps
Upstream 1,011 Kbps
ADSL settings
VPI/VCI 0/38
Type PPPoA
Modulation ITU-T G.992.5
Latency type Interleaved
Noise margin (Down/Up) 3.3 dB / 5.9 dB
Line attenuation (Down/Up) 47.0 dB / 25.0 dB
Output power (Down/Up) 0.0 dBm / 12.9 dBm
Loss of Framing (Local) 44
Loss of Signal (Local) 15
Loss of Power (Local) 0
FEC Errors (Down/Up) 278217 / 3
CRC Errors (Down/Up) 10 / 2147480000
HEC Errors (Down/Up) nil / 20
Error Seconds (Local) 1
Will leave it a few days and see how things go.
Edit: Have you seen many recent disconnects from your end?

[#6]

Hi Mav, no problem.
Only the one drop in the past 3 days:
<img src="http://ccgi.psmith12.plus.com/visradius/generated/image13627504042767.png"/>

[#7]

Thanks Adam.
That drop would have been when I plugged into the Master socket this morning.
As I alluded previously, it's not so much the disconnections (although I have been getting more than I would like) but the quality of the line.
I'll report back Monday as to how things seem to be.

[#8]

Line just resynced back to ADSL2.
[quote=Stats now:]Line state Connected
Connection time 0 days, 0:02:45
Downstream 6,002 Kbps
Upstream 1,011 Kbps
ADSL settings
VPI/VCI 0/38
Type PPPoA
Modulation ITU-T G.992.3
Latency type Interleaved
Noise margin (Down/Up) 3.4 dB / 6.0 dB
Line attenuation (Down/Up) 44.0 dB / 24.9 dB
Output power (Down/Up) 0.0 dBm / 12.9 dBm
Loss of Framing (Local) 44
Loss of Signal (Local) 15
Loss of Power (Local) 0
FEC Errors (Down/Up) 65970 / 0
CRC Errors (Down/Up) 11 / 2147480000
HEC Errors (Down/Up) nil / 12
Error Seconds (Local) 7608

Still connected to test socket.
NM is always low - would this have an impact on stability?

[#9]

Hi Mav,
Look's like your line has been fairly stable and one of the drops was caused by a speed profile update.
<img src="http://ccgi.psmith12.plus.com/visradius/generated/image13629942577667.png"/>
Stability can be impacted if your NM is too low, it really depends on whether your line can handle it. Your line does appear to be holding up OK at the moment.

[#10]

Quote from: chrispurvey

Your line does appear to be holding up OK at the moment.

It seems to stabilise for a while then goes through further periods of instability.
Also concerned that it appears to alternate between ADSL2 and ADSL2+.
Also, there are many times when I may be downloading and my wife complains that she has problems streaming where the streaming, surely, should be prioritised.
Why should the NM be so low and is there anything that can be done about it?
Thanks for your input, Chris.

[#11]

Quote from: Mav

Also concerned that it appears to alternate between ADSL2 and ADSL2+.

Your router determines whether you connect via ADSL2/+ looking at your line this should not be causing you any issues.

Quote

Also, there are many times when I may be downloading and my wife complains that she has problems streaming where the streaming, surely, should be prioritised.

Streaming is prioritsed, is this happening on a regular basis?

Quote

Why should the NM be so low and is there anything that can be done about it?

We could set your line profile to a higher dB if needed.

[#12]

Just had another drop.
[Quote= Stats:]
Line state Connected
Connection time 0 days, 0:00:43
Downstream 6,531 Kbps
Upstream 996 Kbps
ADSL settings
VPI/VCI 0/38
Type PPPoA
Modulation ITU-T G.992.5
Latency type Interleaved
Noise margin (Down/Up) 3.2 dB / 6.1 dB
Line attenuation (Down/Up) 47.0 dB / 24.9 dB
Output power (Down/Up) 0.0 dBm / 12.9 dBm
Loss of Framing (Local) 0
Loss of Signal (Local) 0
Loss of Power (Local) 0
FEC Errors (Down/Up) 28129 / 0
CRC Errors (Down/Up) 0 / 2147480000
HEC Errors (Down/Up) nil / 0
Error Seconds (Local) 0

Event Log doesn't seem particularly helpful:
[quote=Event Log:]15:29:17  11 Mar SNTP Synchronised to server: 213.123.26.170
15:29:17  11 Mar SNTP Systemtime update: time setting 00:04:01 > new time setting: 15:29:17
00:03:10   1 Jan CONFIGURATION saved by TR69
00:02:36   1 Jan CONFIGURATION saved by TR69
00:02:25   1 Jan [CWMP] Connection Request NOT ALLOWED.
00:01:39   1 Jan LOGIN User admin logged in on [HTTP] (from 192.168.1.10)
00:01:37   1 Jan PPP link up (Internet) [212.159.***.***]
00:01:37   1 Jan PPP CHAP Chap receive success : authentication ok
00:01:36   1 Jan PPP CHAP Receive challenge (rhost = ptn-ag01)
00:01:35   1 Jan PPP CHAP Receive challenge (rhost = bras-red12.l-ste)
00:01:29   1 Jan xDSL linestate up (ITU-T G.992.5; downstream: 6531 kbit/s, upstream: 996 kbit/s; output Power Down: 19.8 dBm, Up: 12.9 dBm; line Attenuation Down: 47.0 dB, Up: 24.9 dB; snr Margin Down: 6.2 dB, Up: 3.0 dB)
00:01:27   1 Jan usbmgr: USB device isn't matched the configuration
00:01:26   1 Jan usbmgr: try /etc/usbmgr/class/module
00:01:26   1 Jan usbmgr: try /etc/usbmgr/class/09/module
00:01:26   1 Jan usbmgr: try /etc/usbmgr/class/09/00/module
00:01:26   1 Jan usbmgr: try /etc/usbmgr/class/09/00/00/module
00:01:26   1 Jan usbmgr: class:0x9 subclass:0x0 protocol:0x0
00:01:26   1 Jan usbmgr: buffer line: remove T
00:01:26   1 Jan usbmgr: USB device isn't matched the configuration
00:01:26   1 Jan usbmgr: try /etc/usbmgr/class/module
00:01:26   1 Jan usbmgr: try /etc/usbmgr/class/09/module
00:01:26   1 Jan usbmgr: try /etc/usbmgr/class/09/00/module
00:01:26   1 Jan usbmgr: try /etc/usbmgr/class/09/00/00/module
00:01:26   1 Jan usbmgr: class:0x9 subclass:0x0 protocol:0x0
00:01:26   1 Jan usbmgr: buffer line: remove T
00:01:23   1 Jan LOGIN User Basic logged in on [HTTP] (from 192.168.1.10)
00:01:05   1 Jan xDSL linestate down
00:00:55   1 Jan DHCS server up
00:00:55   1 Jan FIREWALL event (1 of 1): deleted rules
00:00:48   1 Jan FIREWALL level changed to Standard.
00:00:41   1 Jan FIREWALL event (1 of 1): modified rules
00:00:41   1 Jan FIREWALL event (1 of 1): created rules
00:00:30   1 Jan WIRELESS interface turned on.
00:00:29   1 Jan usbmgr: start 0.4.8
00:00:19   1 Jan KERNEL Warm restart
00:00:19   1 Jan KERNEL Warm restart
15:23:36  11 Mar IDS proto parser : tcp null port (1 of 1) : 152.2.31.233 212.159.***.*** 0060 TCP 40882->0 [S.....] seq 3738647187 win 14600
Not sure how useful the above info is.
Maybe we should try a NM reset and see what happens?

Quote from: chrispurvey

Quote Also, there are many times when I may be downloading and my wife complains that she has problems streaming where the streaming, surely, should be prioritised. Streaming is prioritsed, is this happening on a regular basis?

Usually, yes, but strangely not over the weekend. But SWMBO came into to me and ask me to stop downloading about 10 minutes before the connection dropped.

[#13]

Yet another drop and lower resync. Are you seeing these, too.

Quote

Line state Connected Connection time 0 days, 0:02:04 Downstream 5,814 Kbps Upstream 1,004 Kbps ADSL settings VPI/VCI 0/38 Type PPPoA Modulation ITU-T G.992.3 Latency type Interleaved Noise margin (Down/Up) 2.8 dB / 5.8 dB Line attenuation (Down/Up) 43.5 dB / 24.9 dB Output power (Down/Up) 0.0 dBm / 12.9 dBm Loss of Framing (Local) 0 Loss of Signal (Local) 2 Loss of Power (Local) 0 FEC Errors (Down/Up) 102091 / 0 CRC Errors (Down/Up) 0 / 2147480000 HEC Errors (Down/Up) nil / 21 Error Seconds (Local) 1897

[#14]

Hi Mav,
Yeah we can see these drops,
<img src="http://ccgi.psmith12.plus.com/visradius/generated/image1363080251716.png"/>
I'm going to place a SNR reset and set your downstream to 6dB and see if this helps. This will take around 4 hours to take effect.

[#15]

@Mav you need to run routerstats and plot the SNR of your line for a few days, if this is truly a new problem hopefully you will see some indication in the SNR plots to show when the problem happens ... then you have to find it!