General Discussion
  >> Fibre Broadband


Register (or login) on our website and you will not see this ad.


  Print Thread
Standard User mbdcs
(newbie) Tue 29-Mar-16 22:12:22
Print Post

Broadband fault, not sure if it's fixed.


[link to this post]
 
Would anyone like to comment on these line stats (from a talktalk HG633 on VDSL connection):


Upstream line rate (kbit/s):
474
Downstream line rate (kbit/s):
17045
Upstream noise safety coefficient (dB):
6.4
Downstream noise safety coefficient (dB):
6.7
Upstream interleave depth:
6
Downstream interleave depth:
1921
Line standard:
VDSL2
Upstream line attenuation (dB):
5.6
Downstream line attenuation (dB):
28
Upstream output power (dBm):
-4.9
Downstream output power (dBm):
10.3
Channel type:
Interleaved
DSL up-time:
0 days 7 hours 48 minutes 41 seconds

Here is the back story behind it:

I upgraded to FTTC (talktalk) 4 months ago. It was fine up until 3 days ago - I had a stable 22MB down and 1.8MB up, at 2km from the cabinet. Then I found it had lost connection Sunday morning. My router (Draytek 2760) mainly reported the state as 'training', and occasionally 'failed' Every now and again it appeared to sync, but wouldn't pass any data, and the wan interface was reported as down on the router's status page.

I phoned talktalk support, and after going through the basics, then then did a line test, which came back as 'faulty' apparently. Said there was nothing more their tech could do and that they were escalating it to Openreach to investigate, and advised they may need access to the property.

I left the draytek connected for the next 36 hours, and it just kept going through the training process, never with any success. Then I got back the standard talktalk router (HG633) that I'd leant to a friend (also on talktalk fibre, and had been having problems with loss of service, but on a different exchange). I plugged in the talktalk router expecting more of the same, but with less/different diagnostic info, and to my surprise it connected first time. Swapped back and forth a couple of times, and the draytek would never connect, but the talktalk one always did. So it seems the Draytek might have developed a fault, but I'm planning to try it on someone else's line to be sure.

The talktalk router has had constant internet for 24 hours now, although the speed is lower than I had before (17MB down, 0.4MB up). I had received a message saying that I had an openreach appointment for tomorrow, but this afternoon I phoned talktalk support for an update and asked if the fault had been resolved permanently. They looked at the staus and agreed it was working and cleared it down as resolved, so hopefully an engineer won't turn up tomorrow.

Do the line stats above suggest that it really is fine and healthy now? Why is my speed down, is that a DLM effect (wouldn't surprise me as the line was all over the place for over a day), or is it just the talktalk modem not as good as the one in the draytek? Does the talktalk router learn and remember the line performance - my friends line was also synced at 17MB down when he was using the talktalk router he borrowed off me (I'd never used it on my line before, always had the draytek on it). Talktalk support suggested reseting the router to make it relearn the line. I've held off that so far because I didn't want another loss of connection to possibily affect DLM badly, and also I've configured a lot of settings in the router, that I'd rather not have to go through again.
Standard User mbdcs
(newbie) Wed 30-Mar-16 02:40:36
Print Post

Re: Broadband fault, not sure if it's fixed.


[re: mbdcs] [link to this post]
 
I just swapped back to the Draytek router to see what it would do, and it too connected (as in the line synced and it got a WAN IP). It had higher sync rates than the HG633 - DL rate is a bit lower than it used to be, but UL is up to the previous 2MB (limit of the basic fibre package).

But it wouldn't pass any traffic/load any web pages. These are the connection stats:

General Tone Information | Refresh |
ATU-R Information
Type: VDSL2
Hardware: Annex A
Firmware: 05-07-04-03-00-07
Power Mngt Mode: DSL_G997_PMS_L0
Line State: SHOWTIME
Running Mode: 17A
Vendor ID: b5004946 544e0000
ATU-C Information
Vendor ID: b5004946 544eb204 [IFTN]
Line Statistics

Downstream Upstream
Actual Rate 19505 Kbps 1997 Kbps
Attainable Rate 24630 Kbps 2058 Kbps
Path Mode Interleave Interleave
Interleave Depth 817 26
Actual PSD 7. 6 dB 10. 1 dB
Near End Far End
Trellis ON ON
Bitswap OFF ON
SNR Margin 6 dB 6 dB
Attenuation 31 dB 44 dB
CRC 0 11670
FECS 9 s 44699 s
ES 0 s 6070 s
SES 0 s 0 s
LOSS 0 s 3101 s
UAS 52 s 315209 s
HEC Errors 0 0
RS Corrections 0 0
LOS Failure 0 0
LOF Failure 0 0
LPR Failure 0 0
NCD Failure 0 0
LCD Failure 0 0
NFEC 97 64
RFEC 16 16
LYSMB 613 6536

The UL stats look bad to me, for CRC, ES, UAS? I assumed they would reset to zero when a new connection is established, but the Draytek seems to be saving them?

What is making the HG663 UL sync 0.4MB, but the Draytek 2MB? I assumed it was DLM holding the line back, but now I'm confused! My line always used to have no problem supporting 2MB UL, so is there still a fault and it's just the low sync speed of the HG633 is getting around the issue? I suspect the line is ok, could it be a fault on the DSLAM?
Administrator MrSaffron
(staff) Wed 30-Mar-16 09:33:18
Print Post

Re: Broadband fault, not sure if it's fixed.


[re: mbdcs] [link to this post]
 
Probably seeing how the different chipsets cope when there are lots of errors, some go for higher sync and try to correct errors, and others opt for low sync but less errors.

Could be DSLAM issue, or more likely to be a joint/line/new noise source issue.

The author of the above post is a thinkbroadband staff member. It may not constitute an official statement on behalf of thinkbroadband.


Register (or login) on our website and you will not see this ad.

Standard User mbdcs
(newbie) Wed 30-Mar-16 09:58:02
Print Post

Re: Broadband fault, not sure if it's fixed.


[re: MrSaffron] [link to this post]
 
The Openreach engineer has just been (even though I tried to cancel the appointment). He tested the line, and his equipment synced at 20MB down , 1.9MB up, almost the same as the draytek has always done. He did physical line test and was amazed at how good the figures are, considering my line is 2km from the cabinet, probably 3km from the exchange. He also reset the line, so if there was any DLM on there, that should be cleared down now.

The talktalk router syncs at 17MB down, 0.4MB up, and the internet works, albeit slightly slower (probably due to the reduced upload speed and large interleaving, that's still present, despite apparently resetting DLM). The draytek syncs at 20mb down, and 1.9mb up, but the internet is barely usable - often no web pages will load at all, then occasionally it will load a couple of pages before stopping again.

The issue seems to be with the upload, but the openreach engineer said they only look at download sync speeds and error rate, his test equipment doesn't give him any upload error rates. He couldn't explain why the talktalk router would only sync at 0.4mb up.

So I'm no further on with identifying the problem, although I suspect it is the dslam rather than the line itself.
Standard User BatBoy
(sensei) Wed 30-Mar-16 10:04:06
Print Post

Re: Broadband fault, not sure if it's fixed.


[re: mbdcs] [link to this post]
 
The most recent change that Openreach has made is to roll-out G.INP to their ECI cabs. If you are on an ECI cab then it's likely this is what has happened to your line.

You can find what sort of cab you're on here http://www.kitz.co.uk/adsl/cabinet-lookup.htm - ECI or Huawei.
Standard User RobertoS
(elder) Wed 30-Mar-16 11:46:32
Print Post

Re: Broadband fault, not sure if it's fixed.


[re: mbdcs] [link to this post]
 
Are you sure you should be on Annex A? Or is that a quirk of the ECI cabinet? See my stats Huawei stats:-

# xdslcmd info --show
xdslcmd: ADSL driver and PHY status
Status: Showtime
Retrain Reason: 1
Last initialization procedure status: 0
Max: Upstream rate = 15555 Kbps, Downstream rate = 58076 Kbps
Bearer: 0, Upstream rate = 15641 Kbps, Downstream rate = 59504 Kbps
Bearer: 1, Upstream rate = 0 Kbps, Downstream rate = 0 Kbps
Link Power State: L0
Mode: VDSL2 Annex B
VDSL2 Profile: Profile 17a
TPS-TC: PTM Mode(0x0)


Kindness isn't going to cure the world of all its awfulness but it's a good place to begin. Daisy Ridley.
My broadband basic info/help site - www.robertos.me.uk. Domains, site and mail hosting - Tsohost.
Connection - AAISP Home::1 80/20. Sync 59504/15641kbps @ 600m. - BQM
Standard User Al1264
(member) Wed 30-Mar-16 12:02:43
Print Post

Re: Broadband fault, not sure if it's fixed.


[re: RobertoS] [link to this post]
 
Although the VDSL2 will be ITU G993.2 Annex B a lot of (V)DSL hardware still refers to the xDSL Annex A for xDSL over POTS or Annex B for xDSL over ISDN - the Annex A/B having a completely unrelated meaning in this context - I suspect this is what's being seen here.
Standard User RobertoS
(elder) Wed 30-Mar-16 20:32:57
Print Post

Re: Broadband fault, not sure if it's fixed.


[re: Al1264] [link to this post]
 
Ah. Thanks. That's clever of them! Not.

Kindness isn't going to cure the world of all its awfulness but it's a good place to begin. Daisy Ridley.
My broadband basic info/help site - www.robertos.me.uk. Domains, site and mail hosting - Tsohost.
Connection - AAISP Home::1 80/20. Sync 59504/15641kbps @ 600m. - BQM

Edited by RobertoS (Wed 30-Mar-16 20:33:15)

Standard User mbdcs
(newbie) Thu 31-Mar-16 01:33:16
Print Post

Re: Broadband fault, not sure if it's fixed.


[re: BatBoy] [link to this post]
 
In reply to a post by BatBoy:
The most recent change that Openreach has made is to roll-out G.INP to their ECI cabs. If you are on an ECI cab then it's likely this is what has happened to your line.

You can find what sort of cab you're on here http://www.kitz.co.uk/adsl/cabinet-lookup.htm - ECI or Huawei.


I've checked, and it is an ECI cabinet. No idea if G.INP has been enabled on it or not yet. My Draytek has the up to date firmware on it that is supposed to support G.INP, but it isn't reporting it as being enabled or active. Why would G.INP cause my upload speed to quarter, or be full of errors?

I tried the draytek on my friend's talktalk fibre connection (they are also having problems - ongoing, intermittent, infrequent loss of signal). It still wasn't right on there, they are very close to the cabinet, with very high max theoretical speeds, and the draytek router was possibly slightly better on their connection. Their talktalk router was only syncing 0.8mb up, despite being on the 2mb up service, and having a theoretical max upload speed of 25mb.

Back at home, and I've put the Draytek back on my line (after resetting to factory defaults and reconfiguring), and things seem to be looking a little better. It's online, syced at similar speeds to before all the issues. There still looks to be a lot of errors on the uplink, but ping times are looking reasonable, and the longer it's online, the more web sites actually seem to load up successfully!

The interleave depth has decreased to 308 down, and 1 up.

So it seems to be ok again tonight, but no idea what the problem was, if it was something me or the Openreach engineer did, or if it just got better by magic!
Standard User mbdcs
(newbie) Mon 04-Apr-16 00:39:57
Print Post

Re: Broadband fault, not sure if it's fixed.


[re: mbdcs] [link to this post]
 
Looks like I spoke too soon - the Draytek router stayed connected (and stable) for about 36 hours, and then kaput - lost sync and couldn't resync.

I swapped back to the talktalk router, and that synced up straight away. But still with an upload rate of 0.5MB, it also states the max attainable upload rate is 0.5MB as well. Which is clearly nonsense as until now, the line has always supported (and snyced at 1.8-2MB) with the Draytek. The openreach engineers test equipment synced at 1.9MB upload as well.

As above, my friend's line has an attainable upload rate of 25MB (according to the talktalk router), yet that too only syncs at 0.8MB. Are talktalk intentionally knobbling upload speeds on their 40/2 service?
  Print Thread

Jump to