Trinitarian broadband!!!

[#1]

I think it wouldn't be a bad idea if agents, on putting (in as simple a way as possible) relevant technical detail on a ticket (as well as the formal stuff for internal), then encouraged customers to the forum for a more detail explanation of what it all meant, if they so wished. A lot of customers won't bother because as long as their fault is fixed, that's all they will worry or care about.

[#2]

Advodat,
Back on thread...! 
Anotherone and I have been through all of the date acquired to day - I am grateful for the information sent via PM and that provided by Adam it has helped to explain what has been happening on your line, specifically the last 5 disconnects...
25 Aug 08:54 xDSL
26 Aug 10:52 xDSL
26 Aug 11:14 xDSL
27 Aug 01:12 xDSL
27 Aug 03:45 PPP
These can all be explained away by configuration changes and or testing being performed by PlusNet either directly or as a consequence of BTOR / PN systems implementing requested changes.  These are not the issues we are seeking to profile for you.  Thank you though for diligently gathering the diagnostics, it helps a great deal.
Attempted logins to the router from .66 - given your comment about the RS configuration having been lost, I would suspect that those ATTEMPTED (and failed logins) coming from "admin" and "Admin" were RS trying to connect repeatedly with the wrong credentials.  There is therefore nothing to be concerned about.
HTH?
Can you please let Adam know of suitable dates for each of his requests - via PM might be best.
Please when you are able, keep the logs coming when there is a disconnect.  Please keep an eye on the RS graphs, if there is anything other than the evening dip, please post the graphs.
If you encounter your usual issue, it would be good to remember the browser back button and try again when the link comes up.
When you need further help, please shout, we will be listening!
Cheers,
Kevin

[#3]

Good morning Avodat,
Over night reflections...!
There is one thing which has not been tried yet to see if it makes any material difference to the broadband operation, which is to move the router to the master socket.  I know there is no power there and it is a LJU (not NTE5) but a test (even short just to gather stats) could be illuminating.
The line's attenuation of 22.5dB suggests that the synch speed should be 17Mbps - yours is only 13Mbps.  Obtaining the same in the master socket might suggest that the internal wiring has no material affect on the line's performance and thus the degradation arises from an obscure issue on the copper circuit (phone line) or in the exchange.  This comparison could be done before or after the firmware upgrade.

@Adam / CRT,
The customer advises two important points...
1. There was never any issues when service was provided by O2 - one assumes this was LLU?
2. The issue has only become manifest since May this year
In other words the line has previously performed well.  Could the above facts point to exchange side issues or changes?  Are there maintenance records for the exchange which could be inspected?  If copper tests all report clean, what is the possibility of a marginal line card?  Can this be progressed as an intermittent BROADBAND issue please?  Check for best d/e-side, lift n shift, TPM?
I know that the recent changes are the normal "first line approach" to addressing stability, however if that is the right "circumvention" then usually the DLM steps in and raises SNRM, applies interleaving and caps US.  The DLM appeared to be happy with a 3dB SNRM, no interleaving and an uncapped US therefore does the "cause" lay somewhere else?
On the basic measures, this line is performing 3+Mbps below expectations - one presumes that the line is not banded?  The absence of a nte5 socket (BTOR's demarcation point) means that the BT installed extensions / wiring cannot be eliminated from the equation by the subscriber.  In such circumstances if BTOR attend and find that the fault is in THEIR installed internal wiring, is the subscriber still vulnerable to being given a no fault found in BT line / equipment charge?
Kevin

[#4]

I now have the router inside the front door plugged into the master phone socket - with a very long power cable running all the way from the kitchen to feed it. My desk-top is not very happy though - it is only showing 52mbps, as opposed to its usual 130mbps!
Rx noise has dropped to 6.4 and is beginning to rise, or not, as the case now is!

[#5]

Avodat,
Thank you for that. The SNRM is about where one would expect it to be, though the .4 bit implies a change in back ground noise. What we are more interested in is the DS synch. A full set of modem stats when time permits would be useful please.
Kevin

[#6]

Afternoon all,
Thanks for getting touch with me via private message today Avodat, I've got everything queued and ready to process the firmware update for you between 10am and 12:30pm tomorrow so we should hopefully start to see some improvements. I'll be sure to drop a another reply to your private message to confirm we have seen the update complete.
I'm glad to see that we're plugged into the master socket and I hope we see some marked improvements to the connection over the testing period!

@Townman - I think there could well be some differences being seen in the broadband service on the BT equipment as opposed to the O2 LLU equipment and this wouldn't be the first time we have noticed it. There is not doubt that better speeds have been seen on LLU in the past and this could well be a factor in play here. We certainly haven't see a gap as large as 4mbps with this type of technology shift so I'd be inclined to say it could be a factor but shouldn't be the sole cause.
Unfortunately it would be impossible for us to look at these side by side as we can't see any of the line information (inclusive of faults and diagnostic testing) prior to the service moving over to us. With regards to the issue starting in May, it would be tough to narrow this down to any one cause. It could have been a firmware update to the router around that time, it could have been maintenance carried out by the supplier or it could have been another internal/external factor.
Ideally I think we need to see where the firmware update leave us and then see what more we can do. Personally I think there are benefits to be had from changing the socket as well as replacing some of the copper 9where needed) but we do reach a bit if a wall here given that the line isn't with us. Contacting the phone provider should give the option to have the immediate copper checked/replaced as well as the master socket however I would hazard a guess at this being at cost to Avodat which I find unacceptable given the age of the equipment in place now.
Ideally this should be upgraded as part of having a broadband service running on it however I'm afraid that this isn't an option for us to request the work free of charge so I would presume that this is the same with the phone provider.
We could go down the route of pushing for an engineer to attend on a broadband fault side, I'm unsure on their stance regarding the test socket however they will still look to levy a charge if this issue is caused by the extension wiring in the home along with the router so the next few days of testing will be crucial in testing both of these areas in order to eliminate them if they are indeed not the cause.
I'm very happy to push down this route pending a few days connected at the master socket and after we have the firmware up to date if we then see no improvements although I'm afraid this could be a tip into the unknown.

[#7]

Quote from: Adam

I think there could well be some differences being seen in the broadband service on the BT equipment as opposed to the O2 LLU equipment and this wouldn't be the first time we have noticed it. There is not doubt that better speeds have been seen on LLU in the past and this could well be a factor in play here. We certainly haven't see a gap as large as 4mbps with this type of technology shift so I'd be inclined to say it could be a factor but shouldn't be the sole cause.

Adam,
For clarity, Avodat's comparison has never been about synch speed but about connection stability with PlusNET.  He has observed that his brother in London had connection stability issues with PlusNET, which raised the question in respect of TG582n firmware versions.
The question of synch speed is one I have raised against expectation implied by the line's attenuation.
Kevin

[#8]

Afternoon all,
Apologies for not making it to the thread today, with the outages it has been a tough day to be working Twitter however the good news is that we should have seen the firmware update around 11:30 today. Our friends in the product team pushed this out to the hardware and I would now be interested in seeing if this has made any kind of difference with the stability of the connection.
@Townman - I understand the issue we have with Advodat's connection isn't related to the speeds but more the stability of the service. I think I may not have explained the reference I made very well so apologies on that one. Speeds have been the the main difference seen on connections shifting from LLU to BT equipment and I fear I may have gone off on a tangent about this without making the point that we haven't ever really seen stability differences in a straight switch between the two sets of equipment.
We have seen stability corrections with a very prominent fault if a connection is switched between the two but our situation is a little different. Granted, there could well be something in that sentence that we could base further fault diagnostics on but given the results we have in previous tests, I think it could be difficult to pull anything from the results to support this to a level where we can have an engineer identify and fix easily.
Once we see if we have any positive outcome from todays fix I would hope that we can start looking into the possibility of getting the copper looked at!

[#9]

Evening Adam,
Thank you for the after hours update and clarification.  Good to hear that the FW update proceeded to plan, given all that has been happening today.  Strange how things go really bad on Friday afternoons... seen all to many of those over the years!

Avodat,
How are things with you?  Is the router still in the LJU?  Could you please post your line stats from the router when you have a moment.

Quote from: avodat

My desk-top is not very happy though - it is only showing 52mbps, as opposed to its usual 130mbps!

That is not anything to be concerned about UNLESS you do a lot of data transfer between the PC and the laptop.  52Mbps is 4x faster than the DS line speed.  In other words, your WiFi connection is still much faster than th ebraodband link.
Kevin

[#10]

It seems a bit slow, overall.
It is chaotic! The noise graph has been all over the place - a multitude of horizontal, vertical and even a diagonal line at one point, and the dB figure at the top has varied between a fraction (less than one) and 43dB  (it currently says: ags dB ) and sync speed went up into the millions as well (it currently says: end Kbps). The data in the spaces at the bottom and top of the RouterStats page has varied from the sane to the insane!  All this since 12.20 when I connected my laptop to the router again.
Uptime: 0 days, 8:13:12    
DSL Type: ITU-T G.992.5    
Bandwidth (Up/Down) [kbps/kbps]: 440 / 12.925    
Data Transferred (Sent/Received) [MB/MB]: 56,88 / 955,35    
Output Power (Up/Down) [dBm]: 12,6 / 0,0    
Line Attenuation (Up/Down) [dB]: 10,8 / 22,5    
SN Margin (Up/Down) [dB]: 31,0 / 6,6    
System Vendor ID (Local/Remote): TMMB / ----    
Chipset Vendor ID (Local/Remote): BDCM / IFTN    
Loss of Framing (Local/Remote): 0 / 0    
Loss of Signal (Local/Remote): 0 / 0    
Loss of Power (Local/Remote): 0 / 0    
Loss of Link (Remote): -    
Error Seconds (Local/Remote): 15 / 0    
FEC Errors (Up/Down): 0 / 4.369    
CRC Errors (Up/Down): 0 / 25    
HEC Errors (Up/Down): 0 / 320 generateTasks()  
The Telnet log just says   "(Admin)"   and nothing else. It has been showing several odd pages this afternoon!  Some have been almost blank, others have had very different data than usual and this is the first blank one.

[#11]

Did you make a backup of the Routerstats.ini file as I suggested the other day?
Edit: if you touch anything on the Router Login/logout configuration page in Routerstats, it promptly starts with a default .ini file or corrupts the original. If you have a backup you can copy it back to the .ini should it get fouled up.
I'd Close Routerstats and start it again just in case that's all it needs.

[#12]

Anotherone - there was no .ini file(s).
I have shut this down, this afternoon, but it continues its crazy journey. I'll close it overnight and start it again in the morning. I assumed this was PlusNet making any necessary changes. It crashed about 20 mintes ago, which has ended its crazy run, it seems, and it is now looking fairly normal at 6.6mB and time up has been restored to its correct state and the graphs llok nerer being normal.

[#13]

Hi Avodat,
Do you recall my CONFESSION in respect of RouterStat's ability to looes the "plot" - do forgive the pun please!
From what you describe, I would suspect that RouterStats is getting its internal messages twisted.  What is displayed on the SUMMARY tab please when everything is looking "wrong"?  Remember that stop / start plotting will fix the issues.
The modem stats are interesting in part - they are little different than that seen at the end of the BT installed extensions - though inconclusive, another "you should do" test has been performed and shown to deliver no discernible difference.
If you are able, could you please take a look at the BITS PER TONE graph - this can be found on the BITLOADING tab.

@Adam,
Just thinking, does anything need doing with the open fault ticket to stop it "dying" ?

Thanks,
Kevin

[#14]

Probably a good idea to make sure Routerstats is running correctly before doing anything else Kevin.
avodat, in the folder where you have installed RouterStats you should have the following files at least -
RouterStats.ini    RouterStats.chm    RouterStats.exe    libeay32.dll     ssleay32.dll    routerstats_history.txt   install.txt    routerstats.htm
Have a look tomorrow. When Routerstats is running ok and the settings are correct make a backup copy of that Routerstats.ini file

[#15]

Townman - that also changes from having data to no data.
We'll see what tomorrow brings.
Anotherone - there is no .ini file. I'll see after it has had a chance to settle overnight.
I've attached sample graph from earlier this evening.