Print

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

[plexy]

Thanks BC, Ronski and Kitz,

I managed to persuade the wife it was freezer defrosting time last time we were away, so it was all off.

I haven't looked back on the plusnet forums or tickets since I last posted, im sick of trying to put forward the concept that the line worked at full whack for weeks, then deteriorated and became error prone, latent, slower etc. Ive just given up, its not worth my time arguing with plusnet and I can pay to exit the contract and switch to a better ISP once FTTPoD is available in this area

They did email me today and said an engineer is coming tomorrow, with this cryptic note

BOT - FTTC Logged Faults - Post SFI until this time.

So ill be in tomorrow, armed with graphs, stats etc to see whatever this bod finds.

Kitz, thank you for posting on their forums re the speed DLM profile. I hope they get back to you if they have not already. BTW - it may be a bug/reporting issue but DSL stats showed my US SNRM go *up* from its customary 6 to 20 today. It soon recovered back down. Not sure if thats just a reporting thing?

[roseway]

That peak in the US SNRM is almost certainly a false report. The router may have reported a false value just for one sample, and DSLstats may have mishandled it. Whatever the reason, it's very unlikely that it represents reality.

[Ronski]

Ive just given up, its not worth my time arguing with plusnet and I can pay to exit the contract and switch to a better ISP once FTTPoD is available in this area

One of Plus Nets biggest support problems, and I expect this applies to a lot of the big companies is that you always get different support staff, they never read the full ticket history and you often end up going around in circles.

The next big problem, is the ISP has to convince Open Reach that there is a problem. Last year I went around in circles trying to get PN to sort out my brothers broadband (complete lack of), in the end after three months I emailed the CEO of BT at 17:00 on a Sunday, I received a reply within minutes saying it would be sorted ASAP. The next morning I had a phone call from high up support member at PN, because it had been elevated to a CEO level complaint this gave them higher access into Open Reach. We had an engineer on site 3 days in a row, and it was fixed by the Thursday. During in those four days that one member of staff at PN kept gave me constant updates, and gave me his direct number, it was how support should be.

Obviously going to a more expensive ISP will get you better support, as in one person will handle the problem, but then they still have to deal with BT OR.

If you're going to go with FTTPoD then you may as well stay with PN, because the chances of a fault will be so greatly reduced.

[Black Sheep]

I know it's going OT slightly (sorry) ..... but I deal with 'Tech support' as an engineer daily, and I have to say SKY and Zen's 'Tier 2' support are very switched on !! Obviously, I deal with SKY a heck of a lot more, but Saeed and Will on their Tier 2 are very good.
I don't get any golden numbers, just the normal 0845 3661038 for SKY customers.

Thought it good to give credit where credits due. 

[plexy]

Hi Ronski, very valid points there. I think they must have found something as they have escalated it to BT.

A engineer was due out yesterday between 1pm and 6pm, but was a no-show. Shame as me and 4 kids had to wait in all day which was a nightmare in itself. Another poor show by plusnet tbh (Though I accept that the engineer is BTOR my contractual relationship is with Plusnet and its their duty of care to ensure that appointments are kept IMHO).

Thanks for the insight there Blacksheep, I probably should have spent more time shopping around for ISP I guess. I did consider AA or Bogons for their technical know/techie friendliness how but were too much ££ for the wife to let me get

BTW all, line stats are getting worse. Sync speed has gone down slightly, INP is still 3, delay still 8 and interleaving is at 1041. Used to get no CRC's on the DS but now they are stacking up, while the CRC's on the US remain about the same. Never had ES on the DS before till now.

SNRM also seems to be weird now. I now get dips to 0 in the DS SNRM as well as the US, though I am yet to discover if the DS dips correlate to phone being off hook.

In terms of experience, max download speed throughput seems to be 57mbps (down from 60) now (sync showing 65). Upload throughput seems to have lost ~1mbps and is now at 15 (numbers based on the mean result of collective tests ran to different servers). Ping remains low to mid 20's, with a small amount of occasional packet loss (nothing id worry about at the moment)

      Per second   Per minute   Per hour     Per day

CRC   Up   0.01      0.58      35.0      841   
   Down   0.39      23.7      1421      34102   

FEC   Up   0.17      10.0      603      14462   
   Down   10.6      634      38053      913275   

HEC   Up   0      0      0      0   
   Down   0      0      0      0   

ES   Up   0.01      0.42      25.0      600   
   Down   0      0.04      2.47      59.2   

*edit* just noticed - the SNRM changes seem to correlate. When the DS SNRM dropped, the US SNRM went up at that exact moment, and then US SNRM seems to have a dip at the exact moment the DS SNRM flattens out/recovers. Any thoughts on whats causing that?

[burakkucat]

Are you able to see the computer monitor displaying the real-time SNRM graph, as produced by Eric's DSLstats utility, when using the telephone?

If so, use the phone and call 17070, option 2. Listen to what should be a quiet line whilst watching the monitor. When you get bored with that, clear down the call and re-initiate it, this time taking option 1, ring-back. When prompted, return the handset to the rest and watch the monitor as the ringing voltage is applied to the line.

What, if anything, is shown by the SNRM graph? 

[plexy]

Are you able to see the computer monitor displaying the real-time SNRM graph, as produced by Eric's DSLstats utility, when using the telephone?

If so, use the phone and call 17070, option 2. Listen to what should be a quiet line whilst watching the monitor. When you get bored with that, clear down the call and re-initiate it, this time taking option 1, ring-back. When prompted, return the handset to the rest and watch the monitor as the ringing voltage is applied to the line.

What, if anything, is shown by the SNRM graph? 

Yes I am and its usually a big dip in the US SNRM only. That graph attached (*edit* in last post), around 17:30 was an incoming call, ring for maybe 20 s then a few mins chat. The US SNRM went to 0. DS SNRM was unaffected. Ill do another test Tuesday when im home using 17070 (kids asleep now) to see if we get any DS change now - by using your exact instructions !

[burakkucat]

That graph attached (*edit* in last post), around 17:30 was an incoming call, ring for maybe 20 s then a few mins chat. The US SNRM went to 0. DS SNRM was unaffected.

I see a drop in the US SNRM to 1.9 dB at approximately 1735 hours and not to zero.    At 1850 hours the US SNRM performs a 'step rise' to 7.2 dB as the DS SNRM hits zero. (Or does it? Could it be a graphing artefact?)

Ill do another test Tuesday when im home using 17070 (kids asleep now) to see if we get any DS change now - by using your exact instructions !

Sure, no problem. I'll keep an eye open for the new graph.

[plexy]

Hi BC - sorry for not yet performing the tests, worked called me away so am just back last night. I *will* do them soon (btw -agreed the graph I posted does not show SNRM go to 0. Thats in earlier graphs. Sometimes it would dip a little, other times all the way to 0. But yes in that graph I was wrong to state it hit 0.

Nothing much to report before today other than a drop in throughput down to 0.5mbps down the other day for no reason. Fixed itself a few hours later, but thankfully managed to catch the problem via the BTW speed test and line diagnostics.Was told by plusnet its been escalated and they would get back to me.

Then, an interesting thing happened this morning. Came back from dropping off the kids on the school run to find, unannounced, a BTOR van parked on my drive. the guy said he was booked two days ago so had no idea why PN hadn't notified me. Think he was SFI, not sure. Said he'd been doing it 17 years and he arrived in one of the huge vans with all kinds of gubbins in the back.

His line tester didn't show anything was wrong other than the sync being much lower than it should be. I showed him my graphs and stats, which he was very interested in and said 'Yep, looks like a HR fault at least, probably some REIN or crosstalk too'. He did his line tests with this thing plugged into my electric socket, which he said showed nothing in the house was doing anything noisy (guess that rules our REIN on my side at least. Glad I wasn't going mad).

He also looked at the topography of the local wiring, which explained why my attenuation is that of a 5-600m line instead of the 250 it is to the cabinet. Unfortunately for me,if I lived in the house across the street, id be just shy of 250 wire length.  In my place its more like 550. Ive attached a little map - black is my house, green is the cab, blue is the DP down the manhole. At least im not as unlucky as my neighbour diagonally opposite who appears to have another 200m run at least, even though he is ironically closer to the cab than me. Sorry for the pic quality, etc.

The DP down the manhole was drenched, but it looked like only small amount of water had ingressed into the casing - the silica was only a little soggy He also replaced my D side pair from the cab up to the DP, and then reset DLM on the cab. He said if this doesn't resolve it, a new pair may need to come from the DP to the property. And if that doesn't resolve it, then its a problem inside the cab with the actual terminating equipment.

Current sync is 78.5mbps and I have the graphing on the go. Upstream SNRM still dips when off hook, but seems to only be by 1db currently. will keep an eye on that one. DS SNRM doesnt change off hook.

Current errors so far are;

      Per second   Per minute   Per hour     Per day

CRC   Up   0      0.28      16.7      402   
   Down   0.01      0.76      45.8      1100   

FEC   Up   0.09      5.24      314      7542   
   Down   0      0      0      0   

HEC   Up   0      0      0      0   
   Down   0.01      0.81      48.8      1170   

ES   Up   0      0.14      8.22      197   
   Down   0.01      0.44      26.1      627   

What do you guys think of these stats? More ES and HEC on the DS, but is it enough for DLM to be concerned?

[burakkucat]

That's definite progress.

Think he was SFI, not sure. Said he'd been doing it 17 years and he arrived in one of the huge vans with all kinds of gubbins in the back.

If not an SFI engineer, he probably could be one (once the paperwork re: promotion has been 'shuffled'). The fact that he mentioned 17 years definitely rules out Black Sheep but it could have been a cousin from the same flock, i.e. Grey Sheep. 

A huge van with all kinds of gubbins in the back (with which I could 'play') would have made my day. 

As for your current statistics, I really can't say. Perhaps the owner of a VDSL2-experienced eye will be able to comment?

I'll try and 'whistle up' an Eagle for you . . . 

P.s. Thank you for the map. The D-side length is now perfectly understandable.

[Black Sheep]

Ha ha ..... cheers B*Cat.

Bit of boring info, but around our way, you could generally tell the skill-set of an engineer by the van he drove. This will obviously differ geographically (especially the outlying areas such as parts of Scotland and Wales), but usually the 'Huge van' which we call 'Tonners', will be a jointers van.
The Transit, or Vivaro, will be CAL/OMI ..... or CSE ..... or CSE with broadband.
The really small vans with ladders on top can also be CAL/OMI, or Private Wire engineers, or Electric, Lighting & Power engineers, Survey Officers, Precision Test Officers.
The smallest of all, like the Ford Ka, will have the MDF 'Jumper engineer' in them.

As I say, this is NOT written in the laws of the land, just a rough idea .......... if you're that interested. 

[Bald_Eagle1]

As for your current statistics, I really can't say. Perhaps the owner of a VDSL2-experienced eye will be able to comment?

I'll try and 'whistle up' an Eagle for you . . . 

As those stats show averages over periods of time, we can't see if there are any spikes etc. that MIGHT indicate a 'problem'

My preference would be to see a montage of ongoing stats over a period of a few days, hopefully registering 'before & 'after' any changes in performance, particularly when making/receiving phone calls.

[hake]

I have heard that FTTC involves traffic sharing the fibre between the cabinet and the exchange. and that the greater the number of users of the fibre, the more the speed tends to diminish.  I infer that FTTC is more susceptible to contention issues.  I understand that people who connect to the cabinet first get the full bandwidth of the fibre.  As more use FTTC through that cabinet, the pioneer FTTC users may believe that their connection is getting slower.

Someone please disabuse me if this understanding if I am mistaken.

I find PlusNet extremely helpful and they have been consistently so over the eleven years I have been a customer.  They are beholden to BT Wholesale and Openreach.

[Chrysalis]

I have heard that FTTC involves traffic sharing the fibre between the cabinet and the exchange. and that the greater the number of users of the fibre, the more the speed tends to diminish.  I infer that FTTC is more susceptible to contention issues.  I understand that people who connect to the cabinet first get the full bandwidth of the fibre.  As more use FTTC through that cabinet, the pioneer FTTC users may believe that their connection is getting slower.

Someone please disabuse me if this understanding if I am mistaken.

I find PlusNet extremely helpful and they have been consistently so over the eleven years I have been a customer.  They are beholden to BT Wholesale and Openreach.

who did you hear this from?

The contention between cabinet and exchange is very low, lower than isp backhauls, and its so low that the upstream bandwidth is generally uncontended as is enough capacity for all to use at once.  I have never heard of congestion between cabinet and exchange.

However there is signal contention on the D side aka crosstalk.

[Black Sheep]

I tend to side with Chrysalis on this (about the contention being with the ISP's backhaul), although I'm sure our very own Walter has commented otherwise on this somewhere ??

He used B4RN's symmetrical fibre provision as a direct comparison to BTOR's, and I'm sure there was something in there along the lines of what Hake is asking over ?? Apologies if I've got that wrong.