internet keeps dropping after 10pm

[#1]

I think there are enough graphs in this thread for us to see there are two different nights with different looking graphs at the same time of night.

[#2]

just to keep the forum updated as to whats happening

I phoned Plusnet spoke to faults and

he said he has turned up SNR TO 12DB MIGHT TURN UP TO 1`5DB  he said. also he said he set interleaving to MEDIUM and he was going to monitor it for 24hrs,ck let u know how i get on 

ta

will get ba

[#3]

Sight of the current DSL stats would be useful.

So they changed some settings which affect the DSL connection, what must have been the downstream target SNRM to 12 dB, and interleaving, medium is the standard level for interleaving being on. These are fairly typical actions to attempt to mitigate the problem without actually fixing the line or removing the source(s) of the interference, if there are actually any.

As usual, everyone else seems to completely ignore the issue with the upstream. We've seem the connection dropping, the downstream speed stays much the same each time, the upstream speed varies considerably - it should be obvious which one is the problem. Changing the downstream target SNRM isn't going to do anything to a problem with the upstream.

[#4]

Sight of the current DSL stats would be useful...............

he told me to leave the new modem on for 24hrs so he can monitor it , so i cant run routerstats and tgn582 modem at moment.

These are fairly typical actions to attempt to mitigate the problem without actually fixing the line or removing the source(s) of the interference, if there are actually any.

will i get a stable and decent speed ?

Changing the downstream target SNRM isn't going to do anything to a problem with the upstream.

so u reckon it will start dropping again

thanks

im getting fed up , dont feel im getting anywhere regarding a fix but i have learned alot if nothing else

[#5]

---

@glesgajohn wrote:

• http://192.168.1.254/expert_user.html
• get this
• Statistics -- xDSL
   

I meant more stats consisting of the table of numbers from the modem, like those in the very first post.

It's possible they also capped the upstream speed to 448k or could try that, but I don't think any amount of interleaving or increased target SNRM will be enough to hide a problem as severe as this, so yes, unfortunately I'd expect the problem to continue.

[#6]

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):	5.6	5.9
Attenuation (1 dB):	39.0	23.4
Output Power (1 dBm):	0.0	12.6
Attainable Rate (Kbps):	11844	1052
	Path 0		Path 1
	Downstream	Upstream	Downstream	Upstream
Rate (Kbps):	9726	1053	0	0
MSGc (# of bytes in overhead channel message):	92	67	0	0
B (# of bytes in Mux Data Frame):	24	32	0	0
M (# of Mux Data Frames in FEC Data Frame):	1	1	0	0
T (Mux Data Frames over sync bytes):	8	1	0	0
R (# of check bytes in FEC Data Frame):	4	6	0	0
S (ratio of FEC over PMD Data Frame length):	0.0818	0.9719	0.0	0.0
L (# of bits in PMD Data Frame):	2835	321	0	0
D (interleaver depth):	384	32	0	0
Delay (msec):	8	8	0.0	0.0
INP (DMT symbol):	2.00	2.00	0.0	0.0
Super Frames:	795015	713501	0	0
Super Frame Errors:	583	0	0	0
RS Words:	467949878	779573	0	0
RS Correctable Errors:	125806	31	0	0
RS Uncorrectable Errors:	1802	0	0	0
HEC Errors:	489	0	0	0
OCD Errors:	0	0	0	0
LCD Errors:	0	0	0	0
Total Cells:	219627649	23746836	0	0
Data Cells:	5129878	724184	0	0
Bit Errors:	0	0	0	0
Total ES:	61	0
Total SES:	10	0
Total UAS:	161	151

so yes, unfortunately I'd expect the problem to continue.=

i can only wait and see hope not

will u kindly post back your thoughts on the stats above plze

[#7]

Really not sure what has been done as opposed to what you believe you were told would be done.  As of these stats, the SNRM is at the normal setting of 6dB and the synch speeds are fairly hot.

Expect disconnections!

@MatthewWheeler can you please advise of the proposed approach here?

[#8]

The downstream may have been banded to 4864-9728 and has been given the greater than normal downstream INP of 2 (INP 1 is the usual on the downstream). The higher INP number indicates a higher amount of protection from noise. The downstream target SNRM is not really applicable with a banded speed, the only target SNRM with a banded profile is 6 dB, but the banding tends to limit the speed rather than the target SNRM.

The downstream interleaving depth is very high at 384, although this is not actually significant, because it's still the same 8ms delay, so the latency (ping times) will be the same as with ordinary interleaving levels.

The upstream is as normal, INP 2 is the standard level for the upstream.

So in summary, it's been given extra strong error protection on the downstream, and we'll have to wait and see if it's strong enough to fix / mitigate / workaround the problem.

[#9]

@ejs,

An aside - Thank you that's really a very helpful insight on something that I did not register - can you point me to any easy to digest reference material on this please?

Cheers.

[#10]

hi me again , yes you guesses, thats it started dropping again, started @ 3.20pm thats when i noticed it, faults guy was supposed to phone me today but not heard from him, yet .

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):	-4.7	4.5
Attenuation (1 dB):	39.0	23.5
Output Power (1 dBm):	0.0	12.6
Attainable Rate (Kbps):	6904	1004
	Path 0		Path 1
	Downstream	Upstream	Downstream	Upstream
Rate (Kbps):	9726	1024	0	0
MSGc (# of bytes in overhead channel message):	92	64	0	0
B (# of bytes in Mux Data Frame):	24	32	0	0
M (# of Mux Data Frames in FEC Data Frame):	1	1	0	0
T (Mux Data Frames over sync bytes):	8	1	0	0
R (# of check bytes in FEC Data Frame):	4	6	0	0
S (ratio of FEC over PMD Data Frame length):	0.0818	1.0000	0.0	0.0
L (# of bits in PMD Data Frame):	2835	312	0	0
D (interleaver depth):	384	32	0	0
Delay (msec):	8	8	0.0	0.0
INP (DMT symbol):	2.00	2.00	0.0	0.0
Super Frames:	5460081	4906711	0	0
Super Frame Errors:	61463	796	0	0
RS Words:	14828798	1212004	0	0
RS Correctable Errors:	8436	611	0	0
RS Uncorrectable Errors:	255943	0	0	0
HEC Errors:	300603	1874	0	0
OCD Errors:	7	0	0	0
LCD Errors:	7	0	0	0
Total Cells:	6839994	704857	0	0
Data Cells:	39996	12507	0	0
Bit Errors:	0	5097	0	0
Total ES:	1758	171
Total SES:	1063	18
Total UAS:	1677	594

[#11]

@MatthewWheeler,

Negative SNRM reported by the new router ... that does not inspire confidence!

[#12]

hi Townman, what damage does that do ? the faults guy never phone back, typical !

cheers

[#13]

---

@Townman wrote:

An aside - Thank you that's really a very helpful insight on something that I did not register - can you point me to any easy to digest reference material on this please?

---

Sorry, I am not aware of any such reference. There was the "ccSNR Best Practice Guide" which listed the rate bands with their available INP levels, but about a month ago BTWholesale changed their website so that almost everything requires a login for access.

[#14]

hi again(still dropping )

checked stats

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):	8.9	9.0
Attenuation (1 dB):	39.0	23.4
Output Power (1 dBm):	0.0	12.6
Attainable Rate (Kbps):	9900	972
	Path 0		Path 1
	Downstream	Upstream	Downstream	Upstream
Rate (Kbps):	9262	972	0	0
MSGc (# of bytes in overhead channel message):	59	65	0	0
B (# of bytes in Mux Data Frame):	144	30	0	0
M (# of Mux Data Frames in FEC Data Frame):	1	1	0	0
T (Mux Data Frames over sync bytes):	2	1	0	0
R (# of check bytes in FEC Data Frame):	10	6	0	0
S (ratio of FEC over PMD Data Frame length):	0.4991	0.9866	0.0	0.0
L (# of bits in PMD Data Frame):	2484	300	0	0
D (interleaver depth):	64	32	0	0
Delay (msec):	8	8	0.0	0.0
INP (DMT symbol):	1.00	2.50	0.0	0.0
Super Frames:	11035074	9983111	0	0
Super Frame Errors:	418567	3415	0	0
RS Words:	2688391	1358130	0	0
RS Correctable Errors:	443	0	0	0
RS Uncorrectable Errors:	12	0	0	0
HEC Errors:	7332107	8256	0	0
OCD Errors:	0	0	0	0
LCD Errors:	0	0	0	0
Total Cells:	7330365	760513	0	0
Data Cells:	249941	61384	0	0
Bit Errors:	0	0	0	0
Total ES:	13646	793
Total SES:	7146	61
Total UAS:	7563	1831

has anything changed . also this is what they sent me 

KBD

Service ID:	BBEU24002822	Circuit ID:	CBUK70076493	Current BRAS Profile(CVLAN rate):	9696
Product Type:	WBC End User ACCESS	Provisioned Bras Profile:	adsl10000	Profile Info:	WBC 4M - 9M Medium delay (INP 2) 6dB Downstream, UC Medium delay (INP 2) 6dB Upstream (ADSL2+)

xDSL Status Check

Circuit ID: CBUK70076493 Service ID: BBEU24002822 Telephone NO.: N.A. Test Executed On: 2017-04-16T17:03:32.321Z xDSL Status Test Summary Sync Status: Circuit In Sync General Information NTE Status:   NTE Power Status: Unknown Bypass Status:       Upstream DSL Link Information Downstream DSL Link Information Loop Loss: 24.1 39.5 SNR Margin: 4.5 4.8 Errored Seconds: 0 5 HEC Errors: 0   Cell Count: 10111 19359 Speed: 480 9726   Maximum Stable Rate (KBPS): 9088 Fault Threshold Rate (KBPS): 7270 Mean Time Between Retrains (Seconds): 1235 Mean Time Between Errors Upstream (Seconds): 43 Indicative Line Quality: R Mean Time Between Errors Downstream (Seconds): 20 Custom Thresholds MTBR_RED:   MTBE_RED:   MTBR_GREEN:   MTBE_GREEN:

Radius

[#15]

Considering the xDSL stats do not match the "Profile Info: 4M - 9M ..." details, I'm guessing someone has adjusted the parameters again. I'm assuming it's no longer on a banded profile, so has the standard downstream INP of 1, and they've set both the downstream and upstream target SNRM to 9 dB.