Kitz ADSL Broadband Information
adsl spacer  
Support this site
Home Broadband ISPs Tech Routers Wiki Forum
 
     
   Compare ISP   Rate your ISP
   Glossary   Glossary
 
Please login or register.

Login with username, password and session length
Advanced search  

News:

Pages: [1] 2

Author Topic: Recap and review, and Statfest  (Read 3923 times)

Weaver

  • Senior Kitizen
  • ******
  • Posts: 11459
  • Retd s/w dev; A&A; 4x7km ADSL2 lines; Firebrick
Recap and review, and Statfest
« on: May 08, 2019, 02:16:26 AM »

A few specific, problems and a general question about what you think of the ES and the CRC counts? Especially the upstream.

[Stats for all four lines follow.]

(i) What is a good guideline for the acceptable number of CRCs and ES per unit time? [Apologies if I have asked this before but I would appreciate a recap if need be.]

(ii) Lines 1 and 2 are 300k lower downstream than lines 3 and 4.  Why is this?

They are also 200k lower than they have been historically. Why?

(iii) The upstream on some of the lines seems to be showing a few CRCs and ES. Should I be concerned?

(iv) This is so even on the lines where the upstream SNRM is keeping up at a reasonable level, near 6dB. Opinions? Should I consider raising the upstream target SNRM on those lines?


Stats for line 1

Code: [Select]
xdslctl: ADSL driver and PHY status
Status: Showtime
Last Retrain Reason: 8000
Last initialization procedure status: 0
Max: Upstream rate = 449 Kbps, Downstream rate = 3148 Kbps
Bearer: 0, Upstream rate = 532 Kbps, Downstream rate = 2883 Kbps

Link Power State: L0
Mode: ADSL2 Annex A
TPS-TC: ATM Mode(0x0)
Trellis: U:ON /D:ON
Line Status: No Defect
Training Status: Showtime
Down Up
SNR (dB): 3.1 5.9
Attn(dB): 65.0 40.6
Pwr(dBm): 18.4 12.4

ADSL2 framing
Bearer 0
MSGc: 52 12
B: 33 62
M: 4 1
T: 3 1
R: 10 12
S: 1.4841 3.7267
L: 787 161
D: 2 8

Counters
Bearer 0
SF: 18556726 977626
SFErr: 4 316
RS: 807217590 1501767
RSCorr: 69371 33957
RSUnCorr: 18 0

ReXmt: 2232 0
ReXmtCorr: 2059 0
ReXmtUnCorr: 18 0

Bearer 0
HEC: 20 583
OCD: 2 0
LCD: 2 0
Total Cells: 2051043372 378623893
Data Cells: 19431926 9596608
Drop Cells: 0
Bit Errors: 598 59200

ES: 14 219
SES: 10 0
UAS: 174 164
AS: 301608

Bearer 0
INP: 26.00 2.00
INPRein: 0.00 0.00
delay: 8 7
PER: 16.13 16.77
OR: 28.74 8.58
AgR: 2899.49 538.85

Bitswap: 91436/91517 201/201

Total time = 3 days 19 hours 58 min 36 sec
FEC: 71867 37564
CRC: 450 316
ES: 14 219
SES: 10 0
UAS: 174 164
LOS: 1 0
LOF: 9 0
LOM: 0 0
Latest 15 minutes time = 13 min 36 sec
FEC: 74 72
CRC: 0 0
ES: 0 0
SES: 0 0
UAS: 0 0
LOS: 0 0
LOF: 0 0
LOM: 0 0
Previous 15 minutes time = 15 min 0 sec
FEC: 91 87
CRC: 0 0
ES: 0 0
SES: 0 0
UAS: 0 0
LOS: 0 0
LOF: 0 0
LOM: 0 0
Latest 1 day time = 19 hours 58 min 36 sec
FEC: 14986 9917
CRC: 0 103
ES: 0 72
SES: 0 0
UAS: 0 0
LOS: 0 0
LOF: 0 0
LOM: 0 0
Previous 1 day time = 24 hours 0 sec
FEC: 25742 8728
CRC: 1 65
ES: 1 45
SES: 0 0
UAS: 0 0
LOS: 0 0
LOF: 0 0
LOM: 0 0
Since Link time = 3 days 11 hours 46 min 47 sec
FEC: 69371 33957
CRC: 4 316
ES: 4 219
SES: 0 0
UAS: 0 0
LOS: 0 0
LOF: 0 0
LOM: 0 0
NTR: mipsCntAtNtr=0 ncoCntAtNtr=0

Stats for line 2 (which was struggling a bit possibly?):

Code: [Select]
xdslctl: ADSL driver and PHY status
Status: Showtime
Last Retrain Reason: 8000
Last initialization procedure status: 0
Max: Upstream rate = 421 Kbps, Downstream rate = 3256 Kbps
Bearer: 0, Upstream rate = 515 Kbps, Downstream rate = 2776 Kbps

Link Power State: L0
Mode: ADSL2 Annex A
TPS-TC: ATM Mode(0x0)
Trellis: U:ON /D:ON
Line Status: No Defect
Training Status: Showtime
Down Up
SNR (dB): 3.3 5.4
Attn(dB): 65.0 41.4
Pwr(dBm): 18.2 12.4

ADSL2 framing
Bearer 0
MSGc: 54 11
B: 18 62
M: 8 1
T: 5 1
R: 16 12
S: 1.7231 3.8462
L: 780 156
D: 1 8

Counters
Bearer 0
SF: 16531658 15788
SFErr: 733 70
RS: 619937191 268369
RSCorr: 3085163hhi 21344ggc y> &      C !.’
RSUnCorr: 4013 0

ReXmt: 13353 0
ReXmtCorr: 12801 0
ReXmtUnCorr: 4013 0

Bearer 0
HEC: 3842 130
OCD: 111 0
LCD: 111 0
Total Cells: 1759209296 326831307
Data Cells: 81901041 10719297
Drop Cells: 0
Bit Errors: 98356 10907

ES: 1388 449
SES: 13 0
UAS: 172 172
AS: 268695

Bearer 0
INP: 29.00 2.00
INPRein: 0.00 0.00
delay: 8 8
PER: 16.15 16.34
OR: 29.71 8.32
AgR: 2793.37 522.12

Bitswap: 61805/63872 2759/2759

Total time = 26 days 21 hours 31 min 19 sec
FEC: 3085163 21344
CRC: 733 70
ES: 1388 449
SES: 13 0
UAS: 172 172
LOS: 1 0
LOF: 0 0
LOM: 0 0
Latest 15 minutes time = 1 min 19 sec
FEC: 2 4
CRC: 0 0
ES: 0 0
SES: 0 0
UAS: 0 0
LOS: 0 0
LOF: 0 0
LOM: 0 0
Previous 15 minutes time = 15 min 0 sec
FEC: 48 4
CRC: 0 0
ES: 0 0
SES: 0 0
UAS: 0 0
LOS: 0 0
LOF: 0 0
LOM: 0 0
Latest 1 day time = 21 hours 31 min 19 sec
FEC: 364891 4311
CRC: 354 31
ES: 79 22
SES: 0 0
UAS: 0 0
LOS: 0 0
LOF: 0 0
LOM: 0 0
Previous 1 day time = 24 hours 0 sec
FEC: 1222315 8708
CRC: 0 23
ES: 0 17
SES: 0 0
UAS: 0 0
LOS: 0 0
LOF: 0 0
LOM: 0 0
Since Link time = 3 days 2 hours 38 min 13 sec
FEC: 3085163 21344
CRC: 733 70
ES: 141 53
SES: 3 0
UAS: 0 0
LOS: 1 0
LOF: 0 0
LOM: 0 0
NTR: mipsCntAtNtr=0 ncoCntAtNtr=0

Stats for line 3

Code: [Select]
xdslctl: ADSL driver and PHY status
Status: Showtime
Last Retrain Reason: 8000
Last initialization procedure status: 0
Max: Upstream rate = 302 Kbps, Downstream rate = 3292 Kbps
Bearer: 0, Upstream rate = 376 Kbps, Downstream rate = 3132 Kbps

Link Power State: L0
Mode: ADSL2 Annex A
TPS-TC: ATM Mode(0x0)
Trellis: U:ON /D:ON
Line Status: No Defect
Training Status: Showtime
Down Up
SNR (dB): 2.4 5.7
Attn(dB): 64.5 40.4
Pwr(dBm): 18.4 12.4

ADSL2 framing
Bearer 0
MSGc: 51 11
B: 36 5
M: 4 16
T: 3 8
R: 10 14
S: 1.4888 8.0000
L: 849 110
D: 2 4

Counters
Bearer 0
SF: 121326080 807600
SFErr: 2781 27
RS: 891722660 891916
RSCorr: 2185741 41447
RSUnCorr: 10481 0

ReXmt: 174724 0
ReXmtCorr: 129062 0
ReXmtUnCorr: 10819 0

Bearer 0
HEC: 9484 27
OCD: 113 0
LCD: 113 0
Total Cells: 1468181162 1722842719
Data Cells: 429602891 55990364
Drop Cells: 0
Bit Errors: 422693 3298

ES: 771 69
SES: 100 0
UAS: 1067 986
AS: 1943110

Bearer 0
INP: 26.00 2.00
INPRein: 0.00 0.00
delay: 8 8
PER: 15.91 17.00
OR: 28.65 8.00
AgR: 3147.27 382.50

Bitswap: 389253/390103 25962/25962

Total time = 46 days 19 hours 51 min 24 sec
FEC: 2266530 3559301
CRC: 7111 80
ES: 771 69
SES: 100 0
UAS: 1067 986
LOS: 9 0
LOF: 70 0
LOM: 0 0
Latest 15 minutes time = 6 min 24 sec
FEC: 92 0
CRC: 0 0
ES: 0 0
SES: 0 0
UAS: 0 0
LOS: 0 0
LOF: 0 0
LOM: 0 0
Previous 15 minutes time = 15 min 0 sec
FEC: 94 9
CRC: 0 0
ES: 0 0
SES: 0 0
UAS: 0 0
LOS: 0 0
LOF: 0 0
LOM: 0 0
Latest 1 day time = 19 hours 51 min 24 sec
FEC: 10864 1270
CRC: 0 2
ES: 0 1
SES: 0 0
UAS: 0 0
LOS: 0 0
LOF: 0 0
LOM: 0 0
Previous 1 day time = 24 hours 0 sec
FEC: 24956 1386
CRC: 4 2
ES: 3 2
SES: 0 0
UAS: 0 0
LOS: 0 0
LOF: 0 0
LOM: 0 0
Since Link time = 22 days 11 hours 45 min 9 sec
FEC: 2185741 41447
CRC: 2781 27
ES: 451 23
SES: 19 0
UAS: 0 0
LOS: 1 0
LOF: 0 0
LOM: 0 0
NTR: mipsCntAtNtr=0 ncoCntAtNtr=0

Stats for line 4

Code: [Select]
xdslctl: ADSL driver and PHY status
Status: Showtime
Last Retrain Reason: 8000
Last initialization procedure status: 0
Max: Upstream rate = 439 Kbps, Downstream rate = 3272 Kbps
Bearer: 0, Upstream rate = 499 Kbps, Downstream rate = 3118 Kbps

Link Power State: L0
Mode: ADSL2 Annex A
TPS-TC: ATM Mode(0x0)
Trellis: U:ON /D:ON
Line Status: No Defect
Training Status: Showtime
Down Up
SNR (dB): 2.3 6.2
Attn(dB): 64.5 40.5
Pwr(dBm): 18.4 12.4

ADSL2 framing
Bearer 0
MSGc: 53 11
B: 35 62
M: 4 1
T: 3 1
R: 10 12
S: 1.4545 3.9735
L: 847 151
D: 2 8

Counters
Bearer 0
SF: 231525391 534848
SFErr: 2500 1231
RS: 1655064000 1871824
RSCorr: 872227 558936
RSUnCorr: 10446 0

ReXmt: 117454 0
ReXmtCorr: 102106 0
ReXmtUnCorr: 10646 0

Bearer 0
HEC: 10018 2575
OCD: 93 0
LCD: 93 0
Total Cells: 1807900101 120696083
Data Cells: 779350822 134729470
Drop Cells: 0
Bit Errors: 461531 242852

ES: 490 835
SES: 15 0
UAS: 281 281
AS: 3750352

Bearer 0
INP: 26.00 2.50
INPRein: 0.00 0.00
delay: 8 8
PER: 16.09 16.88
OR: 29.33 8.05
AgR: 3133.75 505.38

Bitswap: 653436/663706 6285/6285

Total time = 46 days 20 hours 39 min 45 sec
FEC: 874599 596969
CRC: 2501 1313
ES: 490 835
SES: 15 0
UAS: 281 281
LOS: 0 0
LOF: 0 0
LOM: 0 0
Latest 15 minutes time = 9 min 45 sec
FEC: 109 37
CRC: 0 0
ES: 0 0
SES: 0 0
UAS: 0 0
LOS: 0 0
LOF: 0 0
LOM: 0 0
Previous 15 minutes time = 15 min 0 sec
FEC: 238 81
CRC: 0 0
ES: 0 0
SES: 0 0
UAS: 0 0
LOS: 0 0
LOF: 0 0
LOM: 0 0
Latest 1 day time = 20 hours 39 min 45 sec
FEC: 12691 10144
CRC: 1 40
ES: 1 35
SES: 0 0
UAS: 0 0
LOS: 0 0
LOF: 0 0
LOM: 0 0
Previous 1 day time = 24 hours 0 sec
FEC: 15100 8426
CRC: 3 5
ES: 3 4
SES: 0 0
UAS: 0 0
LOS: 0 0
LOF: 0 0
LOM: 0 0
Since Link time = 43 days 9 hours 45 min 51 sec
FEC: 872227 558936
CRC: 2500 1231
ES: 489 789
SES: 15 0
UAS: 0 0
LOS: 0 0
LOF: 0 0
LOM: 0 0
NTR: mipsCntAtNtr=0 ncoCntAtNtr=0

Logged

Weaver

  • Senior Kitizen
  • ******
  • Posts: 11459
  • Retd s/w dev; A&A; 4x7km ADSL2 lines; Firebrick
Re: Recap and review, and Statfest
« Reply #1 on: May 08, 2019, 02:25:48 AM »

Line 2 upstream is going up and down a bit, all over the place, but the waveform height is not great so it is only dropping from 6dB to 5dB, not a big deal, like line 3. After the earlier quoted stats were captured, I initiated an ADSL reconnect, and some time after that, here is the latest day of upstream SNRM history: line #2 upstream SNRM:

Logged

hacktrix2006

  • Reg Member
  • ***
  • Posts: 305
    • xDSL Stats:
Re: Recap and review, and Statfest
« Reply #2 on: May 08, 2019, 03:01:27 AM »

Have you tried ADSL1 mode? It could help slightly, is there a pattern with the SNR drops?

SES CRC ES seem decent for your line length, not sure about the LOS (Loss Of Signal) and LOF (Loss of Frame) though.
Logged

Weaver

  • Senior Kitizen
  • ******
  • Posts: 11459
  • Retd s/w dev; A&A; 4x7km ADSL2 lines; Firebrick
Re: Recap and review, and Statfest
« Reply #3 on: May 08, 2019, 05:13:10 AM »

I used to use ADSL1, for many years, the arrival of ADSL2 gave me a performance boost. I haven’t used it since the end of 2015 and all the DSLAM hardware was upgraded at the time so it isn’t a fair comparison. So to be a fair comparison I would have to look at it again, in isolation.

I would lose all the single bit bins and any other advantages of improvements to the core protocol design. Ought to need to re-read the intro to G.992.3
Logged

DiggerOfHoles

  • Member
  • **
  • Posts: 64
Re: Recap and review, and Statfest
« Reply #4 on: May 08, 2019, 09:27:20 AM »

This may help? or not in process.
Logged

j0hn

  • Kitizen
  • ****
  • Posts: 4093
Re: Recap and review, and Statfest
« Reply #5 on: May 08, 2019, 01:08:26 PM »

This may help? or not in process.


Most likely or not.
In what way will posting snippets of SIN472 help?
Why not just post a link to the whole thing like you have multiple times in the last few days.

I'm not convinced randomly adding snippets of/links to SIN documents adds anything.
Logged
Talktalk FTTP 550/75 - Speedtest - BQM

DiggerOfHoles

  • Member
  • **
  • Posts: 64
Logged

j0hn

  • Kitizen
  • ****
  • Posts: 4093
Re: Recap and review, and Statfest
« Reply #7 on: May 08, 2019, 03:03:38 PM »

https://www.btplc.com/sinet/SINs/pdf/472v2p9.pdf

I'm guessing my sarcasm didn't come across well.
You may as well put that link in your signature.

Still not sure what it adds, specifically for the OP.
Logged
Talktalk FTTP 550/75 - Speedtest - BQM

DiggerOfHoles

  • Member
  • **
  • Posts: 64
Re: Recap and review, and Statfest
« Reply #8 on: May 08, 2019, 05:56:59 PM »

Ought to need to re-read the intro to G.992.3
Logged

kitz

  • Administrator
  • Senior Kitizen
  • *
  • Posts: 33879
  • Trinity: Most guys do.
    • http://www.kitz.co.uk
Re: Recap and review, and Statfest
« Reply #9 on: May 08, 2019, 07:36:55 PM »

I'm confused.  Isn't that what Weaver said he may do?

Weaver will have read in it the past, but you may not be aware and I'm sure he won't mind me mentioning this, is he is in constant pain and sometimes things get a bit hazy when he is on strong pain killers, so may ask our help to refresh his memory or just double check something/wants a 2nd opinion.

I can fully sympathise as I too am on similar medication and some days if someone shows me something from SINET, you may as well just be showing me a blurred document. :/
Logged
Please do not PM me with queries for broadband help as I may not be able to respond.
-----
How to get your router line stats :: ADSL Exchange Checker

ejs

  • Kitizen
  • ****
  • Posts: 2078
Re: Recap and review, and Statfest
« Reply #10 on: May 08, 2019, 08:38:10 PM »

I always think that if you have to ask if some numbers are bad, and you don't otherwise notice anything wrong, the answer is generally going to be no, the numbers are OK.

If you want more upstream bandwidth, I'd suggest trying having upstream interleaving off. You won't necessarily get vastly more errors, the current FEC counts shouldn't really become the CRC counts without interleaving because it doesn't really work like that. The modem and DSLAM will be aiming for the same bit error rate with or without interleaving.
Logged

DiggerOfHoles

  • Member
  • **
  • Posts: 64
Re: Recap and review, and Statfest
« Reply #11 on: May 08, 2019, 08:41:19 PM »

Ought to need to re-read the intro to G.992.3 in conjunction with sin.

I am lazy , old and new to this posting malarkey.

I'm guessing my sarcasm didn't come across well.

No it come across as being a lazy without a /s.

But what do I know about PPP? Nothing as it would appear to not exist?

What's that all about?

So tramadol all round and away with the fairies!

'he is in constant pain and sometimes things get a bit hazy when he is on strong pain killers'

As I resemble that comment Weaver has my sympathy it's not as good as people would have you believe.

 
Logged

Weaver

  • Senior Kitizen
  • ******
  • Posts: 11459
  • Retd s/w dev; A&A; 4x7km ADSL2 lines; Firebrick
Re: Recap and review, and Statfest
« Reply #12 on: May 08, 2019, 09:16:30 PM »

>I always think that if you have to ask if some numbers are bad, and you don't otherwise notice anything wrong, the answer is generally going to be no, the numbers are OK.

Very good point. Wisdom.

Recently there has been a small loss of upstream throughput, after very careful measurements. But yesterday the upstream was superb at 1.53 Mbps on speedtest2.aa.net.uk. (Even though I don’t know much about it and I don’t have a definition of exactly what the results’ units are, it does seem to be a useful reference and I compare it to its own self not to other testers.)

I suppose I could also have asked about a comparison with some other Kitzens’ upstream CRC counts. And also a stupid, ignorant question: at 6dB upstream, should the CRC count be approximately zero?

The thing that fixes me most though was why the line 1 + 2 downstream sync rates are so low, 400k down in one case. Can you see it? Am I missing something ? Just plain _more noise_ these days? Is the only reason that I haven’t seen it with line 4 because that has not resynched in an age and is running at very low d/s SNRM now?

I think looking back performance figures might be bad at this time of year generally, because of the long daylight (and the weather - drought even?).

Even if I lose 4 * 400k downstream I have plenty in reserve as the downstream in winter was heading towards almost 10.7 Mbps measured. So I can’t really complain.

> If you want more upstream bandwidth, I'd suggest trying having upstream interleaving off.
I had never thought of that. An excellent idea. I will report back.

> The modem and DSLAM will be aiming for the same bit error rate with or without interleaving.

Well, that’s the spec isn’t it, to aim for a certain but error rate however you get there. I really have not been thinking straight at all.

Logged

Weaver

  • Senior Kitizen
  • ******
  • Posts: 11459
  • Retd s/w dev; A&A; 4x7km ADSL2 lines; Firebrick
Re: Recap and review, and Statfest
« Reply #13 on: May 08, 2019, 09:56:12 PM »

I forgot something. I raised the upstream target SNRM of line 2 temporarily from 6dB to 9dB and made many careful speed measurements before during and after the change. I lost about 90 kbps sync rate by going up to 9dB. The effect on measured speeds was exactly that, I lost what I expected to lose and did not derive any compensatory benefit from reduced L3 retransmissions, so as I was hardly surprised to learn, the error rate was already too low to begin with.

It’s not like before when, I was getting approx one corrupted packet for every hundred, fixed the reliability and so got a big performance gain.

So that was just a sanity check. Demonstrates that extremely low error rates mean practically nothing in terms of the effect of L3 ReTx on performance.
Logged

kitz

  • Administrator
  • Senior Kitizen
  • *
  • Posts: 33879
  • Trinity: Most guys do.
    • http://www.kitz.co.uk
Re: Recap and review, and Statfest
« Reply #14 on: May 08, 2019, 11:37:06 PM »

Whilst J0hn's comment may have appeared to have been harsh, he was completely correct that it had nothing to do with SINET or BT/Openreach.

Weavers remark about reading G.992.3 was about basic core DSL technology - in particular the advantage G.992.3/G.992.5 has over G.992.1 when it comes to bit load and RS coding gain.
 
Even though a line may have exactly the same amount of tones and the same SNR, adsl2 can quite often eke more out of a line (without opening up the full range of adsl2+ tones if they aren't needed), but it was a fair enough comment by hacktrix just to check & see.   It's all about the underlying technology and core standards of xDSL as opposed to anything to do with the carrier or their network implementation.

I thought I was trying to be nice and explain why Weaver had asked certain questions, especially when some of your comments didn't make sense to me or why you'd repeatedly pick up on a line or words that Weaver had said..  and it's not the first time you've done this.

Please play nice or not at all...  and it's not recommended to start the sarcasm with me thank you. 
Your comments about 'tramadol' were about as far removed as what G.992.3 has to do with SINET in this particular case. 
Logged
Please do not PM me with queries for broadband help as I may not be able to respond.
-----
How to get your router line stats :: ADSL Exchange Checker
Pages: [1] 2