Print

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

[burakkucat]

You should show the visiting engineer those SNRM graphs when she/he arrives. If she/he is of Black Sheep's flock, the graphs will be accepted as evidence that "something is up".

[Roxy]

You should show the visiting engineer those SNRM graphs when she/he arrives. If she/he is of Black Sheep's flock, the graphs will be accepted as evidence that "something is up".

Definitely burakku - that's the plan. Going to spend some time tomorrow trying to piece it all together.
And I hope you're right, I do look forward to them confirming it requires some investigation.

[Roxy]

Before any one gets excited, he didn't turn up. I had to call in again only to be told 'due to lack of technical assistance, he hasn't arrived and we could reschedule for Monday. When I pressed them about what 'technical assistance' means, the gentleman said that they didn't have enough engineers.

He actually got assy with me, when I told him I don't have any faith in that because it keeps happening.

I only waited this two weeks or so on the appointment, so I could get a Saturday booking!

[Black Sheep]

Oh dear. I'm sorry to hear that, Roxy. 

It's a sad fact that for one reason or another, appointments can be missed by under-staffing or other in-life tasks running beyond their expected commitment time. I know it's no consolation to you, but I have seen first-hand the problems encountered when trying to organise a 20,000 strong engineering work-force with a workstack that changes by the second.

What I can't abide is lies though !! 'Technical assistance' should have been made to sound more like 'Lack of resource' in the first instance. If folk are straight up from the get-go, then most considered folk will accept the predicament ?.

[Roxy]

EDIT: Note that in regards to ES/hour, the figures below have increased significantly since posting. It's at
======================================================================

Engineer gone now, was with me for about two hours.

Firstly, he showed up asking if he was here to boost speeds or deal with low speeds - clearly they did not read their notes again. I explained that we have fluctuating SNR and the line is interleaving every five days after a line reset by engineers. He immediately clarified that if it was REIN that he couldn't help. As a gesture he turned on his little K208 radio and waved it in the air a minute then said 'I can't hear anything, but I'll drive around the neighbourhood to be sure.' Sadly at this time, line stats were indicating no fluctuation in SNR ...typical.

I could tell right away he was another one that was quite chatty, however he seemed more interested in a quick OR 'resolution' as opposed to finding an actual solution(will have an example below).
I tried to convince him if nothing else, that his report concerning REIN, would potentially make or break any progress in the matter - if he didn't support us on this, we may be out of luck.

List of things that did /didn't happen :

A) Couldn't do pair test at the house because it's a Huawei cabinet and suddenly their equipment does not work with those any more.
B) Line/error test with interleaving on showed little CRC activity yet a fair amount of FEC activity.
C) He visited the cabinet to do a pair test (passed with similar results as last time). Then reset the DLM/line. He claims there were only a few CRC errors in the first four minutes and that's acceptable to him.
D) He called his office to speak to someone concerning REIN. After about five minutes, they both agreed there was no REIN, no indication of REIN and nothing out of the ordinary. He was going to pass me over on the phone, though the other gentlemen declined to speak with me.
E) Thirty minute waffle conversation in response to me questioning and discussing the situation.

Only after a bit of back and forth did he ask to see the graphs I mentioned at the start of his visit. I showed him the SNR over eight days, polled at two times a day. He asked if we could see them more per hour and thus I presented several days by the hour, showing clear spikes and dips at random times (except in the middle of the night). He wasn't phased by the numbers. I wanted him to look at the DS CRC vs DS FEC where the former was in the fives and tens while the latter was in the fifty thousand+ a day. I also asked him about bitloading and tones/frequency - he said he hadn't a clue about that, they're only trained to look at ES/hour and CRC errors.

These were my points and his answers in bold:

1. Why the fluctuation in SNR at random times throughout the living hours of the day? Because interference is every where and you cannot control it. 3db or 4db up and down daily is no big deal.
2. Why the interleaving on my line?Because at anything around 900-1000 meters, a line can't handle fast path speeds, so DLM trains the line using interleaving to ensure that it can sustain at least some of the high speed and keep your line from resynching. If you were closer to the cabinet, it would have more flexibility in it's fast path requirements.
2a. Fine, though why at almost exactly six days after each line reset does the interleave get applied? Because there must be a significant amount of CRC errors flagging up and it is correcting them.
3. If the above is true, then why does it not affect Upstream, especially as DS and US are tied together? Don't know.
4. Now that interleaving is off, I can see on the stats that my ES/hour is 55 and rising. There are no FEC and CRC count is 2341. Well when I reset the line and checked CRC, only a handful came up, and there were no FEC errors either. I don't know what software you're using, but that's not what I see.
5. If the above is true, then why not do another test and see where the CRC are at now? We're looking for incrementing errors. I've done my final tests, there's nothing more I can do.
6. What's the bottom line here? If the connection is not actually resynching, and you're only losing five or six mbps then there's nothing you can do. OR aren't going to commit to a REIN engineer and if you had true REIN (he kept making REIN out to be an absolute all or nothing) than my van speakers would blow out from the interference.

He even ended it by giving the me the speech, 'hey, I only get 8mbps, you're lucky!'

Let it also be known that when he spoke to his guy at the centre, he didn't plead my case whatsoever. He said "I need you to do a REIN test/check for me on such and such line such and such etc. I can't see any problems any where, I can't hear any significant REIN and the speeds are in acceptable limits." He never said 'Interleaving is on all the time, with repeat symptoms.' or 'the SNR is all over the place' or 'Look I've already gotten 50es/hour' or 'the EU has weeks of data showing there's an issue' etcetc.


Sure, he was nice enough but his knowledge was suspect, he didn't know how to do a basic REIN walk around (nor was interested) and when questioned on things, the responses were more excuses than hard facts.


BT claims is calling tomorrow, and not sure what to say really other than I am not satisfied? I have evidence of something amiss, my line is on constant interleave and the neighbours are still suffering as well.

Additionally, and I'll run this by Paul some more - now that interleaving is off, and I have about a four to five day window, I will keep logging. Is this a good time to cap my downstream in hopes it reduces the ES/hour and keeps me on fastpath? I am to understand that anything over 30 or 40/hr is too many?
My max attentuation is around 47 and I synced at 46, with an IP profile of 45.5 and a throughput of 44mbps. I am conscientious that DLM will most likely cap me at what is a throughput of just under 40mbps, so the question becomes, whether I can find the capped sweet spot in between those 5mbps.

Here are stats since the uptime today:






Thanks again everyone that's following this

[les-70]

  Looking at other connections it seems wise to keep below about 800ES/day to stay on fast path.  The limit may be 24*60 ES/day but if your close to the limit a few bad days may hit you.  If your over 800ES I would initially cap to 40 Mb/s and see what the error rate becomes.

[Roxy]

  Looking at other connections it seems wise to keep below about 800ES/day to stay on fast path.  The limit may be 24*60 ES/day but if your close to the limit a few bad days may hit you.  If your over 800ES I would initially cap to 40 Mb/s and see what the error rate becomes.

Hey Les. Thing is, interleaving will put it on about 40mbps any ways.
It's resynched twice since the engineer left yesterday and in the past four hours I'm at 2075ES/hr.

However tomorrow I will cap it and see what happens - out of curiosity if anything.

EDIT: Never got a chance to manually cap it, interleaving beat me to it - was quick this time, only took two days.
EDIT 2 : Yes REIN engineer on the way, fingers crossed SNR spikes.

[Roxy]

Engineer today was the same one as second to last(so NOT a REIN engineer.... ..), that did a pair replacement for us on the D side. I explained the issue, brought him up to speed, talked about concern over REIN or some type of interference. I said I had graphs, which at no point was he interested in looking at.

He wouldn't initially do a PQT, because they are now failing when used on a Huawei cabinet(with the new upgraded cards - apparently these are being installed in regards to Vectoring trials? and thus are not compatible with JDSU equipment at this time).

Instead, he got out his K208 radio and we found some distinct noise in the room with the computer and the modem/router, that was very strong at the floor. We followed it outwards into the hallway and then it ended.
I went down stairs and turned off the strip of halogen lights in the downstairs hall, and the noise stopped.

He also checked the downstairs (original) socket and noticed one of the extension cables was still connected improperly, and thus rewired that. Furthermore, he installed a RF3 filter. He said he's never had any success with those, but it's worth a try.

When doing the error test after the RF3 install, he said the FEC were almost non existent on the interleaved line, where as before they were in the thousands. Then he recalced the line and left.

Sadly, as of one hour and twenty-five minutes up time, I'm already at 14000 CRC and an ES/hour of 600. And my SNR is dipping to under 5b.



He said he wasn't convinced that the lights were the problem and that it is probably something else. Additionally REIN engineers cannot and will not be assigned until the normal engineers confirm it's REIN related. And I'm thinking 'so figure it out and confirm or deny.' It's like these guys always want to leave it for someone else to do...

I guess then I'm supposed to once again 'observe and report.' I feel like I am going in circles.

EDIT: He also made comment that in some cases they can't/won't triangulate a REIN source and if they do, they cannot tell you whom is the culprit.

[Bald_Eagle1]

EDIT: He also made comment that in some cases they can't/won't triangulate a REIN source and if they do, they cannot tell you whom is the culprit.

I suppose that's to avoid the potential risk of irate users going round & posting a brick through the lounge window of the culprit if he/she decides to do nothing to reduce the interference 

[NewtronStar]

Sadly, as of one hour and twenty-five minutes up time, I'm already at 14000 CRC and an ES/hour of 600. And my SNR is dipping to under 5db.

Are the FEC counts still non existent ?

[Dwight]

Roxy,
How old is your property? Do you site your BB equipment near to the CH or other stuff like that? I know you say others suffer in your area. Just wondering if these could be your issue. Externally what is near to you mobile phone mast wise and police mast wise. Could they be cause of it.
Yes total agree BT sound like chocolate fire guard when it comes to REIN, and might have an agenda for fault logging types to discourage pad press. But phone lines are low voltage low amp-age, so to generate an electric filed to disrupt it your talking more like an iron! So a bad earth circuit in your house could be just building up charge which then dissipates at a certain voltage.
You could also get an independent in like this guy to look at it! Burytec.co.uk
Best of luck
Dwight

[Roxy]

Sadly, as of one hour and twenty-five minutes up time, I'm already at 14000 CRC and an ES/hour of 600. And my SNR is dipping to under 5db.

Are the FEC counts still non existent ?

Correct. Currently at 42258 CRC, 5683 HEC and 0 FEC. Whenever we're back on fastpath or interleaving removed, there are never any FEC.

Roxy,
How old is your property? Do you site your BB equipment near to the CH or other stuff like that? I know you say others suffer in your area. Just wondering if these could be your issue. Externally what is near to you mobile phone mast wise and police mast wise. Could they be cause of it.
Yes total agree BT sound like chocolate fire guard when it comes to REIN, and might have an agenda for fault logging types to discourage pad press. But phone lines are low voltage low amp-age, so to generate an electric filed to disrupt it your talking more like an iron! So a bad earth circuit in your house could be just building up charge which then dissipates at a certain voltage.
You could also get an independent in like this guy to look at it! Burytec.co.uk
Best of luck
Dwight

The home is newer than 1990 I believe. My modem and router are near one another and about five feet from the main computer. There's nothing else plugged in around it. I've actually though about moving it, but my understanding is fibre only works at the master socket, so...?

I don't know of any thing externally other than an old CCTV footage pole up the road about a hundred feet. That DOES give off some interference, but that's the only thing I can think of.

[Black Sheep]

Roxy,
How old is your property? Do you site your BB equipment near to the CH or other stuff like that? I know you say others suffer in your area. Just wondering if these could be your issue. Externally what is near to you mobile phone mast wise and police mast wise. Could they be cause of it.
Yes total agree BT sound like chocolate fire guard when it comes to REIN, and might have an agenda for fault logging types to discourage pad press. But phone lines are low voltage low amp-age, so to generate an electric filed to disrupt it your talking more like an iron! So a bad earth circuit in your house could be just building up charge which then dissipates at a certain voltage.
You could also get an independent in like this guy to look at it! Burytec.co.uk
Best of luck
Dwight

Dwight ........... we could be a 'Chocolate fireguard' or an 'Ashtray on a motorbike' ...... it matters not IF the issue is REIN ??. It is only through goodwill that we even offer to look at the possibility of it being REIN affecting a circuits performance.

If the EU has dodgy central-heating stats, or has decided to buy a sat-receiver from Pakistan ....... this is NOT an Openreach problem.

[Roxy]

So the REIN engineer booked for today, turned out to be ..you guessed it..the same guy as last night.
I found him outside and asked what was going on. He told me that a colleague was on his way over and they were going to look at it together, I assume he meant my issue(s). I also let him know that right at that moment, there were dips in the SNR - the nasty drops I had been experiencing. He just sat there and nodded his head, pointing to his lunch and chewed along...  I don't know if he ever 'heard' any interference or attempted to find any. I briefly saw this colleague of his and then the guy was off.

Nobody asked for any graphs or information.

A bit later he comes by and says he's going to disconnect us at 'the joint' so a.k.a. your internet is going off.
He went over to the manhole/thing by one of the neighbours and started working in there, while also making trips to the cabinet. After a few hours he calls and says to give it half hour and the service will resume.
I questioned what had he done/changed - he tells me that he's put us on another port and also redone our line from the cabinet to the box. I informed him that's interesting because two other guys already changed ports to which he said 'oh, that's odd.....' *awkward silence*  I also asked what the need to rerun the line and his answer was 'because it was suggested to me.' I'm like suggested for the purpose of what? What happened to the walk around with colleague concerning REIN?

The internet is back up and I'm at a speed of 36.8mbps, and a max attainable of 36.9.
Already at 400 CRC errors on 50 minutes of uptime and SNR is falling under 5db now.

Simply mind boggling awful.

[Black Sheep]

I can't really comment conclusively on your specific job, as I don't know what the engineers found with their tests ?? I do have to tentatively remind folk looking in though, that OR have a specific mandate of tests to follow and if they pass then that is our remit fully satisfied.

OR have (out of their own purse, as it is a non-CP chargeable task), sent two men to try and resolve what is a very intermittent fault at best. As the KITZ REIN page states ....... 'tracking down REIN faults can prove quite tricky' ........ and as a REIN engineer myself I can confirm that. The last briefing we had it stated that it took on average nationwide approximately 3.8 visits to determine the source of the REIN.

There's no mystery as to why it would be the same engineer as what turned up last night. It often happens with myself. If I'm visiting on a bog-standard BB fault and subsequently find the issue is likely to be REIN, then there has to be an official REIN case raised to satisfy various protocols and red-tape.

What I always do, unless it's absolute proof-positive that it is REIN utilising our WHOOSH systems, is collate information from BB circuits on the same serving DP as the affected EU, and also others in the same locality. I can more often than not make an informed decision as to whether the interference is local (only the EU affected) or remote (other circuits affected). I can only do this with BT Classic circuits, or GEA products .... as I don't have access to LLU's systems.

All REIN cases are monitored by both OR and the ISP. So you are at liberty under the FOI act to ask what the engineer has done thus far ? If REIN has been found emanating from another premises and the offending item sourced, you will NOT be told where it was found, but you will be informed that REIN is/was the cause of you issues.
The problem is, OR can not demand somebody cease using their dodgy set-top box or whatever we find is causing the problem ?, unless (I believe) it is a radio-ham who is knowingly broadcasting. So, it is possible that REIN can be found but nothing can be done about it.

I am trying to generalise things here, as every case is different.