Kitz ADSL Broadband Information
adsl spacer  
Support this site
Home Broadband ISPs Tech Routers Wiki Forum
 
     
   Compare ISP   Rate your ISP
   Glossary   Glossary
 
Please login or register.

Login with username, password and session length
Advanced search  

News:

Author Topic: Help with possible line banding / issue  (Read 1753 times)

stuf83

  • Just arrived
  • *
  • Posts: 3
Help with possible line banding / issue
« on: August 15, 2019, 10:09:00 PM »

Hi

This is my first post on here as i was directed to post here from someone on the Talk Talk forum so hoping someone can please help me and give some advice. We had issues with our broadband disconnecting every night between 8pm & 10pm and after numerous attempts to fix, including an engineer visit, we bought our own router and it has stopped disconnecting and been stable for 68 days at 63meg.

Before trying the new router, the speed was reducing every night after the drop so when the new router remained stable we were pleased and hoped the DLM would increase the speed. Since we got to 68 days and the DLM hadn't increased the speed, we switched the router off for 30mins and thought this may resync higher but it didn't. The router reconnected at the same speed. Given that we used to connect much higher and that the max rate on the line is being reported at 75meg I wondered if our line is banded? Can anyone tell from the stats below?

If it's not banded can someone tell me why there is almost a 12 meg difference between the actual and max rate please?

Stats when the new router was first installed and remained stable for 68days:

Line Status:    Connected   
DSL Up Time:    68 days, 5 hours and 30 minutes   
DSL Modulation Type:    VDSL2   
Annex Type:    Annex A/L   
   Upstream   Downstream
Current Rate (kbps)   19902   63700
Max Rate (kbps)   19749   75997
SNR Margin (dB)   6           6.1
Line Attenuation (dB)   23.6          12.1
Errors (pkts)           97          26

Router stats now (rebooted 5 days ago after the router was stable for 68 days)

Line Status:    Connected   
DSL Up Time:    5 days, 8 hours and 50 minutes   
DSL Modulation Type:    VDSL2   
Annex Type:    Annex A/L   
                           Upstream    Downstream
Current Rate (kbps)   19566   63200
Max Rate (kbps)      19579      75647
SNR Margin (dB)      6              6.3
Line Attenuation (dB)     23.6          12.1
Errors (pkts)               7              127
Logged

burakkucat

  • Respected
  • Senior Kitizen
  • *
  • Posts: 38300
  • Over the Rainbow Bridge
    • The ELRepo Project
Re: Help with possible line banding / issue
« Reply #1 on: August 15, 2019, 10:25:28 PM »

Welcome to the Kitz forum.  :)

I would be helpful, in terms of understanding the back-story, if you will please disclose the make/model of the TalkTalk supplied modem/router and the one you purchased, to replace it. Also, do you know the make of the cabinet DSLAM? Is it Huawei or ECI?

I'm not too sure about your circuit (line) being banded. The DS synchronisation speed at 63700 kbps and the 63200 kbps are nicely rounded . . . However if we look at the SNRM, at 6.0 & 6.3 dB US/DS respectively, there is no "spare overhead" which would be a good indicator that a circuit is banded. (Perhaps j0hn or A.N.Other will be able to comment, please.)
Logged
:cat:  100% Linux and, previously, Unix. Co-founder of the ELRepo Project.

Please consider making a donation to support the running of this site.

stuf83

  • Just arrived
  • *
  • Posts: 3
Re: Help with possible line banding / issue
« Reply #2 on: August 15, 2019, 10:52:22 PM »

Thanks for the warm welcome and quick reply  :)

The Talk Talk router was the Fast 5364-3 and the router we bought is the Archer VR600 - V2 which has been very stable. As for the cabinet DSLAM, it looks to be an ECI.

Can I ask, when you say there is no spare overhead where would you see that? I thought the spare overhead would be the difference between the actual and the max rate or am I on the wrong train of thought?

Thanks :)
Logged

burakkucat

  • Respected
  • Senior Kitizen
  • *
  • Posts: 38300
  • Over the Rainbow Bridge
    • The ELRepo Project
Re: Help with possible line banding / issue
« Reply #3 on: August 16, 2019, 12:03:55 AM »

If the circuit was banded or capped, I would expect to see the SNRM to be significantly greater than the default of 6 dB. That I am seeing 6.0 & 6.3 dB for the US/DS, respectively, is why I mentioned no spare overhead and, thus, why I think that your circuit is not banded.

As you believe that your circuit is connected to an ECI DSLAM, then there will be no G.Inp applied and the reduction of the target SNRM, in one dB steps, (whilst the circuit remains stable) will also not be implemented.  :(

An xdslctl info --vendor command line (or the equivalent for your modem/router) will return details relating to the line-card chipset of the DSLAM. Sight of that will positively identify the manufacturer.

The Fast 5364-3 is a device of which I am unfamiliar. (b*cat plods off to perform a search . . .)
Logged
:cat:  100% Linux and, previously, Unix. Co-founder of the ELRepo Project.

Please consider making a donation to support the running of this site.

j0hn

  • Kitizen
  • ****
  • Posts: 4093
Re: Help with possible line banding / issue
« Reply #4 on: August 16, 2019, 05:30:20 AM »

What modem is it you're using?

Are the stats from modem/routers webpage or via telnet commands?


It's missing a few bits of info like if Interleaving is enabled.
The max attainable is exaggerated when interleaving is active.

As burakkucat says the SNRM suggests the line is syncing at/near its highest possible rate in its current state.

The line dropped from 63700 to 63200 when you rebooted the modem.
Horribly rounded figures that suggest possible banding but DLM doesn't usually reduce a lines rate for a single resync with 30 minutes grace given.

I can't believe DLM would reduce the banding on a line that resynced once in 68 days.
Banding is the exact same sync every time, with the capped sync going up if a line is stable (few or no resyncs over an extended time period) and down if there's multiple resyncs in a short time.

I'm not familiar with the Archer VR600 - V2.
Hopefully someone can advise on how to get additional stats.
« Last Edit: August 16, 2019, 05:32:59 AM by j0hn »
Logged
Talktalk FTTP 550/75 - Speedtest - BQM

broadstairs

  • Kitizen
  • ****
  • Posts: 3697
Re: Help with possible line banding / issue
« Reply #5 on: August 16, 2019, 10:27:53 AM »

I'd be surprised if it is an ECI cabinet. My experience of these is that even with a cap the sync is around 6db. ECI cabinets are in my view very problematic and do not behave well. I also believe caps are usually at or close to a very rounded figure, mine is 60000kbps giving me an actual of 59995 kbps. I have also found with the ECI cabinet that even BTOR cannot explain why I am capped, I have also had interleaving set now for several years with again no reason BTOR can find for DLM doing what it does. I can get interleaving off if I force my sync as low as around 35000kbps but DLM promptly sets it back as soon as I remove my arbitrary cap. As I said I suspect this is actually a Huawei cabinet. I am also on TalkTalk btw.

Stuart
Logged
ISP:Vodafone Router:Vodafone Wi-Fi hub FTTP

stuf83

  • Just arrived
  • *
  • Posts: 3
Re: Help with possible line banding / issue
« Reply #6 on: August 16, 2019, 11:15:06 AM »

Hi Thanks for the posts it looks like a ECI but I am no expert if you pop the coordinates in google maps street view its quite a good view of the cabinet once you move past the white van  :) 53°25'12.4"N 2°45'56.3"W

Many thanks 
Logged

kitz

  • Administrator
  • Senior Kitizen
  • *
  • Posts: 33881
  • Trinity: Most guys do.
    • http://www.kitz.co.uk
Re: Help with possible line banding / issue
« Reply #7 on: August 16, 2019, 11:48:07 AM »

Yup looks like an ECI.
Quote
Cabinet P20      FTTC Available from 15th November 2012      Phase 06b 2011      527      ECI      L35 4AA 

I think the VR600 has a BCM chipset but TP-link use their own f/w which disables access to the BCM shell.
You may be get some more line stat details using ejs's  StatPOSTer
Logged
Please do not PM me with queries for broadband help as I may not be able to respond.
-----
How to get your router line stats :: ADSL Exchange Checker

dorzb

  • Member
  • **
  • Posts: 98
Re: Help with possible line banding / issue
« Reply #8 on: August 16, 2019, 01:46:34 PM »

If the circuit was banded or capped, I would expect to see the SNRM to be significantly greater than the default of 6 dB. That I am seeing 6.0 & 6.3 dB for the US/DS, respectively, is why I mentioned no spare overhead and, thus, why I think that your circuit is not banded.

As you believe that your circuit is connected to an ECI DSLAM, then there will be no G.Inp applied and the reduction of the target SNRM, in one dB steps, (whilst the circuit remains stable) will also not be implemented.  :(

An xdslctl info --vendor command line (or the equivalent for your modem/router) will return details relating to the line-card chipset of the DSLAM. Sight of that will positively identify the manufacturer.

The Fast 5364-3 is a device of which I am unfamiliar. (b*cat plods off to perform a search . . .)

Aha, thank you burakkucat, as I've learnt something new from you.

My line's attainable net data rates are 102.526 Mbps (down) /31.690 Mbps (up), and it was capped at 40/10 for the last month due to a mistake made by BT. During which my SNRMs were 19.8 dB (up)/20.6 dB (down). My cap was then lifted back to normal (55/10) with the help from the kitzens here and my current SNRMs are 25.8 dB(up) / 15.6 dB(down). I'm planning to get an 80/20 package after my existing contract expires. Would I expect the SNRMs to be around 6.x dB by then, or would there still be some spare overhead?
« Last Edit: August 16, 2019, 01:50:55 PM by dorzb »
Logged

burakkucat

  • Respected
  • Senior Kitizen
  • *
  • Posts: 38300
  • Over the Rainbow Bridge
    • The ELRepo Project
Re: Help with possible line banding / issue
« Reply #9 on: August 16, 2019, 05:36:56 PM »

Yup looks like an ECI.

With a quick glance I can say, with confidence, that there is no doubt about it. We can see a cabinet equipped with ECI electronics.
Logged
:cat:  100% Linux and, previously, Unix. Co-founder of the ELRepo Project.

Please consider making a donation to support the running of this site.

burakkucat

  • Respected
  • Senior Kitizen
  • *
  • Posts: 38300
  • Over the Rainbow Bridge
    • The ELRepo Project
Re: Help with possible line banding / issue
« Reply #10 on: August 16, 2019, 05:44:08 PM »

Aha, thank you burakkucat, as I've learnt something new from you.

b*cat performs one of his best Japanese style bows.

Quote
Would I expect the SNRMs to be around 6.x dB by then, or would there still be some spare overhead?

It is a bit like kitz wondering how long is that ball of wool, the one I just about to chase around the floor and, in the process, unwind. Until it's done we really won't know.

Making a bit of a guess (but not wanting to prolong the hijack of stuf83's topic), I'll say "possibly".  ;)
Logged
:cat:  100% Linux and, previously, Unix. Co-founder of the ELRepo Project.

Please consider making a donation to support the running of this site.
 

anything