Print

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

[burakkucat]

Like WWWombat, I would have appreciated sight of the results of a TDR scan on the circuit. 

Would it be possible that you could borrow a suitable device and then perform the measurement yourself?

[samwise78]

Yeah ..... the job notes would be a great help, cheers.

Here are what I was given by BT chat later on, his notes don't align exactly with what he actually said to me but read into that what you will:

"On arrival all tests  ok.
Pqt passed with good results.
Vdsl test passed with slightly below speeds at property.
Drove to PCP where speeds were still low, carried out a DLM reset and speeds did not increase.
Dcoe stated that the interleaves was preventing the speeds so has been removed.
Speeds now higher however I have advised Mr <samwise> that if he wants higher speeds then he
will need to discuss with his service provider about a more thorough diagnosis as the d side connection
is in very good condition.
Ltok on completion using eclipse."

Like WWWombat, I would have appreciated sight of the results of a TDR scan on the circuit. 

Would it be possible that you could borrow a suitable device and then perform the measurement yourself?

None of the ex-Openreach guys I know have a JDSU, and they're a bit steep to buy for this sole purpose - are there any affordable* devices you could point me at (make/model) that could fulfill the same function adequately?

*I realise affordable for one person isn't necessarily affordable for another..

Cheers
Sam

[Weaver]

Doesn't user CrazyTeeka have a JDSU ? He/she will know how much they are/have tips on where to get one from. I suspect there's no substitute on the cheap, as these devices have a lot of costly hardware in them which is essential.

[burakkucat]

None of the ex-Openreach guys I know have a JDSU, and they're a bit steep to buy for this sole purpose - are there any affordable* devices you could point me at (make/model) that could fulfill the same function adequately?

Occasionally a Tester 301C will appear on eBay. (A simple TDR device, subsequently replaced by the Hawk, subsequently replaced by the JDSU HST-3000c or the Exfo AXS-200/365 -- to place it in the chronological sequence of the devices that have been issued to Openreach technicians.)

Another option would be a specialist equipment hire businesses . . . though I don't know what a day's hire would typically cost.

Of course there is always Walter's Wheelbarrow, if you are located somewhere in Surrey. 

[Weaver]

Yes, hiring might be a great option. I hired a logic analyser once, that would have cost ~£15k to buy. (We kept hiring it for so long that they asked us if we wanted to just buy it eventually.)

[WWWombat]

Drove to PCP where speeds were still low, carried out a DLM reset and speeds did not increase.
Dcoe stated that the interleaves was preventing the speeds so has been removed.
Speeds now higher

I guess this shows the standard DLM reset is to put interweaving profile in place, but that DCoE was able to shift it to an open profile.

However, it still doesn't ring true...

Yes, interweaving + FEC does steal some of the line's speed and latency. However, a JDSU plugged in at the cabinet should easily cope with an 80/20 sync even with such a profile. The zero-distance open speed is probably 160/40... so plenty of headroom.

If they couldn't achieve 80/20 at the cab, something is wrong.

[samwise78]

Hopefully the boost visit on Weds will be more productive - assuming it is actually a boost appointment of course..
Perhaps they'll be willing to view the hlog we can see to investigate further too.
If the guy/gal is receptive and willing I'll try and get a snap/snaps of the TDR test results to share

G.INP enabled overnight here and have now moved DSLstats to an always on PC so should be logging around the clock, I need to change a BIOS setting on the machine so it comes back after a power outage but apart from that should be good to go

Cheers all
Sam

[samwise78]

aand just seen two resyncs within 10 minutes of each other, one preceded by US noise margin nosedive, second one by DS margin nosedive.
 
At work all day so can't really investigate
Grrrr

<edit>
On closer inspection checking the upstream SNRM per band shows U0 spiking upwards to 16dB just before the first resync and then to around 11dB just before the second, what would cause this?

Cheers
Sam

[Black Sheep]

I guess this shows the standard DLM reset is to put interweaving profile in place, but that DCoE was able to shift it to an open profile.

However, it still doesn't ring true...

Yes, interweaving + FEC does steal some of the line's speed and latency. However, a JDSU plugged in at the cabinet should easily cope with an 80/20 sync even with such a profile. The zero-distance open speed is probably 160/40... so plenty of headroom.

If they couldn't achieve 80/20 at the cab, something is wrong.

Absolutely spot-on.

[samwise78]

Absolutely spot-on.

Will see what tomorrow's visit reveals, fingers crossed they get to the bottom of it.

Just had another weird resync, US margin on U0 skyrocketed while DS went in the other direction..

Why would this happen and what would cause it?

Cheers
Sam

Sent from my SM-G930F using Tapatalk

[digitalnemesis]

Will see what tomorrow's visit reveals, fingers crossed they get to the bottom of it.

Just had another weird resync, US margin on U0 skyrocketed while DS went in the other direction..

Why would this happen and what would cause it?

Cheers
Sam

Sent from my SM-G930F using Tapatalk

Noise/harmonics on line.

Can you try to short your line for a few seconds?

[Black Sheep]

It really could be down to anything I like to stick to what we know, rather than what we don't ..... and that is the engineers aren't getting full synch at the Cab, which they should be able to do each and every time once the profile is reset to 'Open'.

This smacks of a) A faulty DSLAM port ....... b) A high-resistance (HR) fault on the Quante termination strips in the DSLAM Cab ..... or c) A split pair between the DSLAM Cab and the Copper Cab.
The resultant fix will require a 'Lift & Shift' should it be either of these three scenario's, as we don't have access rights to the DSLAM Cab.

This advice is obviously given, assuming the previous visit and the subsequent notes are correct. Fingers crossed for a result. 

[samwise78]

Thanks both

Agreed BS and thanks for the breakdown of possible causes (and the fix)
Will report back tomorrow afternoon following the appointment.

Cheers
Sam

[samwise78]

So, my boost appointment was cancelled.
I only found out when the engineer didn't show up, checked the fault on the BT website and it had been closed.
After speaking with a supervisor via Live Chat found out that apparently there's a MSO for this area (although it isn't reported on any of the checkers) with a fix date of 19/08?
Somehow I think I'll still have the same line conditions/symptoms after that date and will need to make another appointment, but will happily be proved wrong..

[samwise78]

So my line has returned to flappyville - had 8 resyncs in the last 24hrs all for LOS or RDI (last 5 today have all been RDI)
Seeing loads more errors and not really sure what's happening to the line, HLOG looks terrible.

Anyone got any ideas/advice?

Attached yesterday's HLOG graph and today's for a comparison..

Thanks
Sam

<edit> better make that 8 resyncs today - one at 14:09 Reason 8000 LOS???
Dread to think what action DLM is going to take, something is amiss though