New "Old" topic - Speed or lack thereof

[#1]

The line itself is looking ok now, I've checked and the profile on our side is correct but I'm hoping the test I've performed has kicked things along a bit for you.
Loop Loss:  24.6  42.0 
SNR Margin:  5.3  6.0 
Errored Seconds:  0  6 
HEC Errors:  0 
Cell Count:  27496  236892 
Speed:  1239  8211 

[#2]

Quote from: Chris

The line itself is looking ok now, I've checked and the profile on our side is correct but I'm hoping the test I've performed has kicked things along a bit for you. Loop Loss:  24.6  42.0  SNR Margin:  5.3  6.0  Errored Seconds:  0  6  HEC Errors:  0    Cell Count:  27496  236892  Speed:  1239  8211

Chris I dont know how intrusive your test was supposed to be but it didnt show anything on Routerstats,  Shows as connected for 23 hours plus  which is when I did a powerdown.
and that also shows CRC errors in excess of 2million and HEC errors in excess of 24million!! 

I put any improvement NOW showing as being down to the fact that the REIN stopped at 17.07pm 

[#3]

It won't drop the sync of the line, it'll disconnect the PPP (authentication).  I can see the disconnect in our RADIUS logs below:

I guess we'll find out tomorrow whether this improves your speeds.

[#4]

Quote from: cedlor

I put any improvement NOW showing as being down to the fact that the REIN stopped at 17.07pm

That will certainly be the case. If you synced when there was no REIN, the minute it starts, it drops the SNRM so low (but not quite enough to lose sync) that the errors increase enormously and so throughput drops dramatically as data has to be continuously retransmitted to try and get some error free data, when it stops, things return to near normal.
Anyway some positive results from the REIN engineer. With that wires replaced you should see a cleaner after dark SNRM graph. A current Bits/Tone might be interesting to compare, but there'd still be some tones knocked out because of the earlier REIN.
Maybe in a day or two, a graceful resync and a Bits/Tone grabbed before any REIN starts would hopefully show the potential, but don't rush to do it and aggravate DLM, remember line ILQ is Amber.

[#5]

Quote from: Chris

I guess we'll find out tomorrow whether this improves your speeds.

Ok well we can forget Thursday ( Tomorrow in the quote) - as often happens and as recorded there was no rein so it worked as expected - no better - no worse.
FRIDAY Rein on at  08-34 until  13.14pm  Probably short day as half term/holiday week.

[#6]

Wasn't there some talk about enabling interleaving at the start of this thread? The stats look like interleaving has been off (they all have zero FEC) for the entire duration of this thread.

[#7]

Hi EJS
Yes there was but there has been some discrepancy in BT reports and PN reports as to whether it is on or off.   As you say mainly off  and I will leave it like that for now until the rein reporting is finished.
Anotherone mentions it a few posts up - and says it is off. - Thanks for spotting it though.

[#8]

Quote from: Anotherone

r. With that wires replaced you should see a cleaner after dark SNRM graph.

I attach night time graphs - can t say I can see it any smoother - will get to the Bits - tomorrow.

[#9]

OH NO!!!   
Looks like its hitting weekends now 

Download  Speed
0.11 Mbps

0 Mbps 7.15 Mbps
Max Achievable Speed
Download speedachieved during the test was - 0.11 Mbps
For your connection, the acceptable range of speeds is 2 Mbps-7.15 Mbps.
IP Profile for your line is - 7.24 Mbps
2. Upstream Test: -provides background information.
Upload Speed
0.68 Mbps

0 Mbps 0.83 Mbps
Max Achievable Speed
Upload speed achieved during the test was - 0.68Mbps
Additional Information:
Upstream Rate IP profile on your line is - 0.83 Mbps

This test was not conclusive and further testing is required.This might be useful for your Broadband Service Provider to investigate the fault.
If you wish to carry out further tests,please click on 'Continue' button.If not, please close the window using 'Exit' button and contact your ISP for further assistance with these results.

[#10]

Based on your current stats I'm not surprised.
Your SNRm is very low and there's errors picking up again.
Our faults team are checking to see if there's any updates on the REIN case.

[#11]

Not sure what happened here -perhaps the noise got so bad it dropped the signal altogether or perhaps Matthew did a reset?
Anyway a bit later looks like the REIN has stopped for today.

[#12]

It looks to me that an intrusive CLT was carried out judging by the duration of the drop. When the test has completed and sync is re-estabished, the effect is the same as doing a resync, so as previously explained the connection is made at the current Traget SNRM. As noise is present (the REIN) making some tones unusable and a number with reduced bitloading, then the sync speed achieved will be lower. Later on the REIN stopped so the SNRM then increased.
When you have a very low SNRM then the CRC rate will increase significantly affecting throughput speed. As previously mentioned, if at such time throughput speed is important, then the only option is to gracefully resync when the REIN is present.
As Interleaving has been mentioned again, to save re-reading the thread, just refresh us, IIRC, it was your choice to want Interleaving Off to achieve lower pings?

[#13]

Hi - yes , I do understand that I can get a little more speed by doing a resync  but most of the time I am happy to leave it as is. After all I shouldnt HAVE to de a resynch twice a day to maintain a workable speed.
Interleaving - yes I said to leave it off   - again I realise that if it were on some of the errors would disappear  but  I wanted to just leave as is so as not to "mask" the rain situation. In fact IIRC  I thought  the Rein engineer
said he switched it off and that if I wanted it on I could just ask PN.  ( Surprised me as I thought it was off already, but maybe I midheard/ misunderstood)
I have Rein today as usual though not quite so bad.and the internet is usable.
Just had a senior moment thought RS had given up at the end of May  but is has inserted todays files in the middle of the run of graphs as J comes before M  .
Keep meaning to ask then forget- There is another  dsl cabinet that is only 30 meters from my pole but is therefore a bit further away from the exchange
Does anyone know if it would be a simple/difficult/impossible task for BT/OR to test if the REIN is affecting that cabinet and could my connection be made through that cabinet as against the other one.
Would it make any difference as possibly that cabinet is fed from the one I am connected to..
I have found another radio ( abit newer - at least the dial is in khz) and will do some more sleuthing when the REIN returns to being bad.


1. Best Effort Test: -provides background information.
Download  Speed
6.2 Mbps

0 Mbps 7.15 Mbps
Max Achievable Speed
Download speedachieved during the test was - 6.2 Mbps
For your connection, the acceptable range of speeds is 2 Mbps-7.15 Mbps.
IP Profile for your line is - 5.85 Mbps
2. Upstream Test: -provides background information.
Upload Speed
0.93 Mbps

0 Mbps 0.83 Mbps
Max Achievable Speed
Upload speed achieved during the test was - 0.93Mbps
Additional Information:
Upstream Rate IP profile on your line is - 0.83 Mbps

[#14]

Still not fixed - drove me mad yesterday BTWtest reports upload 0.02 - took me 3 goes to get the test to load before it timed out!!.   Router stats before restart 105million+ errors
The problem has now extended to Thursdays so today as I really need to use email I have restarted the router. Did not want to as I feel it must distort the problem.
Anyway the result was remarkable - I know you have said before that it will improve the throughput but didnt realise it would be this much. It would be too much of a coincidence for the rein to have stopped at exactly the same time.
So now it has me wondering if when the noise kicks in the router somehow remembers that
and holds it until something jogs it - either the rein stops or I restart the router.
I only ask if that is possible as the noise margin is exactly back to where it was before
the rein kicked in. Maybe just a daft idea - is it what is expected?
Well rein had not ended but has now!!   so do I do another restart. Probably just leave it
as at least it is usable and see what happens next time rein kicks in.

[#15]

Hi,
Remind me (save me scanning the whole thread) which router are you using?
Some routers will - having found tone bins unusable - black block them out from use and not consider reusing them until there is a resynch.  You can look for this characteristic by capturing the bits per tone plot before and after the resynch.
If the REIN / SHINE is still persisting, expect the link to drop again and a return to poor performance.  One would have hoped that by now the BTOR REIN team had located the issue, even if they cannot remedy it.  It might be worth while talking with friends and neighbours nearby and get them to join you in monitoring their service performance.  several people doing this together might identify some more effected than others which might help locate the source.
Kevin
Typo corrected