Print

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

[hacktrix2006]

Right hi all, remember i said about my SNR on down stream being 3DB etc, well yesterday my line has taken a right good old dump.

I have removed all extensions (Data/Phone) and even tested in the test socket with ADSL Micro filters tried 8 all new and its doing the same thing.
The SNR on the Upstream will either go from 6.0db to 0.4db when the phone is picked up and disconnect the internet (Which is why i decided to check filters) or go the other way and have an SNR of 6554.xx when the phone gets picked up (I was shocked as well thought it was a mis-read on my router so even swapped the routers over and the same thing still happens). The error count(s) also go through the roof.

So now the brain is spinning what can this be, there was noise on the line but its intermittent and i have reported it as a Phone fault.

Now all 8 Micro Filters show the same issue when plugged into the test socket, the MK3 and ADSL Nation pre-filered plate shows the same issue when they are connected.

I have changed routers/modems same thing am i right in thinking the High resistance fault is back??
I have been running a logging session whilst i have been doing the test i will upload them if the graphs show the dips.

[burakkucat]

. . . am i right in thinking the High resistance fault is back??

Yes, it certainly reads as if you have one (or more) joints showing HR (or semi-conductive) tendencies.

[hacktrix2006]

Yes, it certainly reads as if you have one (or more) joints showing HR (or semi-conductive) tendencies.

Thought so as calling the number now increases only the upstream errors and decreases or increases snr however picking the phone up now results in broadband going completely.

Does this also in the test socket now as well.

Sent from my MotoG3 using Tapatalk

[burakkucat]

That is fairly conclusive. The problem comes in identifying the defective joint.

Performing a TDR trace, whilst looking towards the exchange battery, and making a call to that circuit could give a good indication of the distance to suspect joint.

[hacktrix2006]

That is fairly conclusive. The problem comes in identifying the defective joint.

Performing a TDR trace, whilst looking towards the exchange battery, and making a call to that circuit could give a good indication of the distance to suspect joint.

Thanks for that info, Engineer should be coming on 9th so unsure weather to make the engineer a brew first before saying what i suspect the issue is or just let him/she be.

My guess the defective joint will either be in the exchange or cab (Engineers cable joint busting when messing with birds nest, it happens quite often), failing that it might be on a road that was recently worked on however the fault didn't show its head till Wednesday (2nd) just gone at around 6pm so that might just rule out roadworks as that was end of last week.

What i find really funny is according to the router i have seen max attenable shoot from 8mb to 9.5mb its as if the connection is trolling me big time.

Just had the connection disco whilst typing this

Line rate (kbit/s)   7611    555    
CRC errors                   79191    0    
FEC errors                   1305    0    
HEC errors            6252    0

Thats in a few seconds and that is going to require a profile reset after its fixed as its now disconnecting ever 10 seconds.

[NewtronStar]

Thanks for that info, Engineer should be coming on 9th so unsure weather to make the engineer a brew first before saying what i suspect the issue is or just let him/she be.

Being nice or offering them coffee/Tea will not change the result they are there to do a job and not socialize with the human race it can be seen as bribery to offer a bacon sandwich to an OR Engineer which goes against OFCOM rules.

[hacktrix2006]

I am always nice to BTOR staff, although i might offer a strong tea or coffee they might need it before finding that dodgy joint(s).
In fact i am nice to anyone that comes to my door until they do something rude or abusive then its nice hat off.

[NewtronStar]

You get three types of OR Engineers the open chatty type and then the one that clinches his/her chin with thumb and makes odd noises while doing the diagnostic tests, the third one is the quiet type which ignores anything you say and keeps their findings to themselves

I find the one that clinches there chin and makes odd noises is the better type of engineer they can see an issue and are trying to resolve it, the Chatty type may be young and inexperienced with a hopeful career ahead, the quiet type just doesn't like the job.

[Weaver]

I always try to be extremely nice to BTOR engineers who are normally excellent locals, always very quiet - the opposite of garrulous. Unfortunately many of them don't seem to understand a lot of what I say to them and communications with the ISP don't seem to be brilliant which isn't very helpful to the engineer. I haven't yet had a female BT engineer. Occasionally I have put AA staff on the phone to try and help the BT engineer a bit if the engineer seems short of details.

Offers of tea and coffee always get turned down, probably because they have already had a dozen such already.

[burakkucat]

. . . and communications with the ISP don't seem to be brilliant which isn't very helpful to the engineer.

b*cat opens a sleepy eye, wide, and muses "Could that be a criticism of A&A?" 

[Weaver]

@Burakkucat - could be, not sure - the system doesn't seem very joined up in that the engineer often doesn't seem to have too much information. But that doesn't tell me where things could be improved.

AA are definitely not completely perfect. That would be weird. They have quite a lot of work on their plates. But there are actual humans about, that’s what makes it good.

[hacktrix2006]

Update on the line.

BTOR come out (Sent Over ground engineer instead of a UG one), whom did his tests PQ test past (Like always) but other tests failed so he went of to CAB17 then after a while rung back to say he swapped legs as one leg was burnt (Which makes me wonder what caused it).

The line has now been reset to 6DB Fast uncapped upstream as of today so will be running my monitoring Raspi once i have remade the SD-Card. No more noise on the line plus as a bonus new faceplate NTE5C and test socket.

At some stage i will upload my setup script in the forum for Headless Wireless/Wired Raspi with RDP (so you can use Windows Remote Desktop) access on here to update what was made on GitHub 3 years ago.

As it does everything for you in one clean .sh file.

Will update the thread after i have tested for a few days lets just hope DLM will be happy and put me back to 3db SNR on downstream.

Edit: My DSLStats Raspi is up link in signature. It might be slow as the QOS setting for that device is set to 64kbps up and 128kbps down.

[hacktrix2006]

@burakkucat @NewtronStar @Weaver

Seems the Internet gods hate me again, Just rebooted my router after rebooting my Raspi Monitor (Like stats to tie correctly) and the line was stable at 6db on the downstream 1 hour later it looks like my SNR has taken a major dump its in the 1.5-1.9db range from nowhere (Currently showing on my Web Monitor which is linked in my Signature). Before the reboot the SNR on the Downstream was 7.2db.

I have also noticed that the Dip's i have been seeing on the Upstream is tied to BTOR working in the cab.
Now it seems that the SNR is not recovering at all right now the line is erroring out like no tomorrow, i have checked the Line using the number ending 70 and performed a QLN little clicks on the line again with some slight noise.

So would the last fault be returning? can't do anything right now as i am getting ready to go under on the operating slab and to be slightly honest i am getting abit of a CBA attitude now with the line due to the stress its causing having to bash my head against the brick wall.

Will keep my Raspi Webserver going for a few days see how it goes but if someone can have a look would be grateful.

I know i can reboot the router and it will more then likely go back to 6db however i know with what its doing now its going to kill my speed and more likely band the profile.

[burakkucat]

Hmm . . .

I don't like all those CRCs without any FECs, it seems as if your circuit is on fast-path. Ah, yes, the Stats Summary confirms it --

Interleave depth:          1      1

Your synchronisation speed is reasonable --

Connection speed (kbps):   7995      1088

I have also noticed that the Dip's i have been seeing on the Upstream is tied to BTOR working in the cab.
Now it seems that the SNR is not recovering at all right now the line is erroring out like no tomorrow, i have checked the Line using the number ending 70 and performed a QLN little clicks on the line again with some slight noise.

So would the last fault be returning?

That does appear to be the case. If it is not the original fault reappearing then it is another one developing. I would not expect normal day-to-day activity in the PCP, by Openreach, to cause a disturbance to your circuit . . . unless there is a problem with the jumpering (between the E- & D-sides) for your circuit.

[hacktrix2006]

It's down even lower now 0.8db SNR on downstream. I guess i will have to put up with it till after my operation as i need to keep my stress levels down as much as possible.

Beside that DLM will kick in soon no doubt and force a resync to a much lower profile.
Although during gaming session tonight and with my father on Youtube watching stuff not had a hickup or lag spike, or anything like lag from packetloss which is strange.

Yes i am on fastpath will always be on that as i am a gamer and every ms counts.
As for the PcP bit, its a bird nest in Cab17 where i am connected to. I have seen it and a few engineers have said a few words when i tell them i am connected via cab17 which is not for this forum.

Anyhows thanks for your reply @burakkucat i am grateful for them.