SNR reset request

[#1]

Hi,
For several months I suffered from lots of of disconnections that caused my link speed to drop. Following a re-organisation of cabling and a replacement router, I now have a stable connection.
Is it possible to try resetting the SNR for my line to try and improve the speed which is currently about 3300 kbps?
Thank you,
Simon

[#2]

The line's looking pretty good now, the SNR is at a target of 12db so I'll request a reset now. This should complete within a few minutes.

[#3]

Thank you, Chris.
I've now got a link speed of 5472 and achieving 4800 download.
Just have to see how it settles down over the coming days...
Best regards,
Simon

[#4]

Simon,
It would be worthwhile keeping a close watch on the stats - capture and save them.  During the next 10 days you can expect to see a number of disconnects as the DLM trains the line.  Seeing a rise in SNRM and a reduction in speed would point to something else requiring attention.
See the speed issues thread for more information.
Kevin

[#5]

Hi Townman,
I'm a bit new to this. Can you suggest a way for me to capture the data? I've got the new router supplied by Plusnet - the F@ST2704N, and the routerstats program doesn't seem to work with it.
So far the router is holding up well, and I've had no resets since the SNR margin was reset 11 hours ago.
Simon

[#6]

Hi Simon,
Until I get around to testing the 2704n I do not have any suggestions beyond copy and paste into a document (spreadsheet may be, one tab per sample) of some description.
When I next get a break in my xDSL session (currently 248 days) I will install the new router and explore all of these matters.
Cheers,
Kevin

[#7]

OK Kevin - I'll take copies of the 'Helpdesk' page.
248 days without a disconnect - I can only dream of a connection like that!
Simon

[#8]

There is also the Sagemcom interface that gives more info than the Plusnet one:-
http://192.168.1.254/expert_user.html
Go to Device Info > statistics > xDSL.

[#9]

The connection has been up for over 7 days now, and I've held the same IP address for a week.
Here's the xDSL information page from the Sagemcom interface...

Statistics -- xDSL
 
Mode:	ADSL_2plus	 	 
Traffic Type:	ATM	 	 
Status:	Up	 	 
Link Power State:	L0	 	 
 
 	Downstream	Upstream	 	 
Line Coding(Trellis):	On	On	 	 
SNR Margin (1 dB):	2.5	17.4	 	 
Attenuation (1 dB):	49.5	26.9	 	 
Output Power (1 dBm):	18.2	11.7	 	 
Attainable Rate (Kbps):	5404	1012	 	 
 
 	Path 0	 	Path 1	 
 	Downstream	Upstream	Downstream	Upstream
Rate (Kbps):	5472	448	0	0
 
MSGc (# of bytes in overhead channel message):	59	14	0	0
B (# of bytes in Mux Data Frame):	171	56	0	0
M (# of Mux Data Frames in FEC Data Frame):	1	1	0	0
T (Mux Data Frames over sync bytes):	1	1	0	0
R (# of check bytes in FEC Data Frame):	0	0	0	0
S (ratio of FEC over PMD Data Frame length):	1.0000	4.0000	0.0	0.0
L (# of bits in PMD Data Frame):	1376	114	0	0
D (interleaver depth):	1	1	0	0
Delay (msec):	0	1	0.0	0.0
INP (DMT symbol):	0.00	0.00	0.0	0.0
 
Super Frames:	0	0	0	0
Super Frame Errors:	60397	303	0	0
RS Words:	0	2391256	0	0
RS Correctable Errors:	0	0	0	0
RS Uncorrectable Errors:	0	0	0	0
 
HEC Errors:	114556	193	0	0
OCD Errors:	2688	0	0	0
LCD Errors:	2688	0	0	0
Total Cells:	244415707	723294702	0	0
Data Cells:	170305648	14025172	0	0
Bit Errors:	0	7169	0	0
 
Total ES:	36981	259	 	 
Total SES:	54	3	 	 
Total UAS:	331	52	 	 
  

Compared with what I had before this seems very good. Is there anything else I should do?
Thanks for your advice,
Simon

[#10]

Hi Simon,
Getting the U.S. uncapped would be good.

@CRT,
I know Chris has been looking at capped US configurations.  Is there yet any insight into how this is happening, when the provisioning system is reported to automatically configure uncapped US?  Is it still all down to manual provisions / alterations falling foul of BTw's (?) default settings on the ordering system?

Quote from: MattyC

I've gone ahead and uncapped that for you and slapped the wrist of the person that placed the order

If yes, again I ask, given the substantial volume of rework this generates, is there no way of having the default option on this system changed to uncapped?
Kevin

[#11]

Hi Kevin,
Thank you for your reply.
Do I need to raise a Support Ticket to get the US uncapped? What sort of difference do you think it would make to my connection.
I sort of expected to see a few disconnections after the SNRM was reset as the retraining happened, but it has been rock steady. Is this because the new configuration was good and didn't need any tweaking?
Thanks again,
Simon

[#12]

Hi Simon,
CRT might pick it up from here.  That is somewhat likely to be quicker than a ticket (aka question - see link below), but CRT are busy and if they do not pick this up today, you might wish to proceed with a ticket, at least that will be a formal request in the system.  If you do that, please post the ticket number here for the benefit of CRT.
Improvement - US will be a little swifter, which means responses will be faster and thus might lift data down load speed a little.
Your link is performing very well, so I doubt you will see training drops.  It is synching on the top side of expectation for your attenuation which is quite likely down to the apparent 3dB SNRM target.  Day time stats would be useful.  Interesting that the last current rate was better than the reported max rate.
Kevin

[#13]

I'm here!
I'll go uncap the upstream for you.
@Kevin
The majority of the cases I've seen are human error, obviously in an order process the less manual touches is generally better so fixing why the orders are dropping out is key.

[#14]

Cheers Chris,
The ordering system considerations are understood, but the BTw system defaults will not help human fallibility when performing manual SNRM (and the like) change orders.  I can think of nothing worse than to have to remember a number of things to check, when all that one is trying to do is alter one existing configuration parameter, it is all counter intuitive.  I bet that the change screen does not even show what the current settings are?
Kevin

[#15]

Quote from: Townman

It is synching on the top side of expectation for your attenuation which is quite likely down to the apparent 3dB SNRM target.

If you subtract 2dB from the attenuation before using the max speed calc, then for an attenuation of 48dB it's doing slightly worse than expected for a 6db target noise margin, despite the apparent 3dB SNRM target.
Perhaps this line is on some of the worse exchange equipment that doesn't even automatically change from ADSL2+ to ADSL2 when appropriate, but there's no way to tell without seeing the bitloading data.