Print

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

[N0STIE]

Ok, I capped the sync to 35000 10000 100000, DS SnR is now significantly higher but for some reason the interleaving depth went up to 1123 from 849. And for reason MDWS shows a resync at 14:03?? I followed the steps your provider. Is that supposed to happen?

I have a question. If I eventually get fast path back with sync capped to 35000 and then I get rid of HG612 and will only be using W9980, will sync remain capped?

[burakkucat]

I have a question. If I eventually get fast path back with sync capped to 35000 and then I get rid of HG612 and will only be using W9980, will sync remain capped?

No. It is a transient state and is exited by a resynchronisation event . . . either using the HG612 or changing to another device.

[les-70]

 One resync should be reported with a switch off and a later resync.   When fast path returns see what error rates you get, you should be able to raise the speed cap in careful stages until the ES are about 0.5 or 0.3 of the 2880 per day on most days. More errors than may be a big risk that a bad day brings back interleaving. 

  See how the ES per day go with the cap.  If you sustain less than ~20ES/day it should go back to fast path eventually.

  If you use the W9980 nothing will be capped but you won't know the stats and can't cap the speed. You will back where you started.

[N0STIE]

No. It is a transient state and is exited by a resynchronisation event . . . either using the HG612 or changing to another device.

Thanks for explaining

One resync should be reported with a switch off and a later resync.   When fast path returns see what error rates you get, you should be able to raise the speed cap in careful stages until the ES are about 0.5 or 0.3 of the 2880 per day on most days. More errors than may be a big risk that a bad day brings back interleaving. 

  See how the ES per day go with the cap.  If you sustain less than ~20ES/day it should go back to fast path eventually.

  If you use the W9980 nothing will be capped but you won't know the stats and can't cap the speed. You will back where you started.

20ES/day combined DS and US or just DS? I'd rather have sync capped than interleaving applied. Since resync at 2pm, only 2 ES appeared.

[les-70]

  Just DS that is the one interleaved.  US just needs to always be less than 2880.

[N0STIE]

My US isn't noisy at all, just few ES a day, SnR is always above 6dB. Will see how it goes through peak times where errors tended to be quite high (5-10/h). But why did interleaving depth increase to 1123?

[ejs]

If the INP and delay didn't change, the increase of the interleaving depth is probably not significant. Depth isn't the only interleaving parameter, there's also the block size, and the other framing parameters such as the amount of FEC data. It'll be the modem adjusting all the parameters to achieve the desired result (INP level within the delay constraint).

[N0STIE]

If the INP and delay didn't change, the increase of the interleaving depth is probably not significant. Depth isn't the only interleaving parameter, there's also the block size, and the other framing parameters such as the amount of FEC data. It'll be the modem adjusting all the parameters to achieve the desired result (INP level within the delay constraint).

What I've noticed is INP increased from 3.00 to 3.50

[N0STIE]

Only 9 ES on DS in 24 hours, I guess not that bad. Is it usually long time before DLM switches back to fast path with that low amount of errors?

[ejs]

What I've noticed is INP increased from 3.00 to 3.50

I'm not sure if that increase is significant or not. It is probably not the DLM's doing, if it changed at the instant that you capped the rate.

Perhaps when you cap the rate, what you're actually capping is the net data rate, which does not include bandwidth used to carry FEC data. So with the rate capped, there's more bandwidth available to carry FEC data, which the modem might opt to use otherwise it might just be wasted. A line profiles specifies the minimum INP level and maximum delay, so the modem could work things out so that it has greater than the minimum INP level specified.

[j0hn]

I'm not sure if that increase is significant or not. It is probably not the DLM's doing, if it changed at the instant that you capped the rate.

Perhaps when you cap the rate, what you're actually capping is the net data rate, which does not include bandwidth used to carry FEC data. So with the rate capped, there's more bandwidth available to carry FEC data, which the modem might opt to use otherwise it might just be wasted. A line profiles specifies the minimum INP level and maximum delay, so the modem could work things out so that it has greater than the minimum INP level specified.

That's exactly correct. Capped at 50mb my INP is 3, at 25mb INP increases to 3.5. Increasing the cap again decreases the INP.

Only 9 ES on DS in 24 hours, I guess not that bad. Is it usually long time before DLM switches back to fast path with that low amount of errors?

I would expect it to take around 10 days from interleaving being applied going by recent observations.

[les-70]

  Looking at your stats from yesterday you may need to cap at 30000 to be sure to recover fast path.  How much of cap you need depends on the nature of the noise on your line and it looks like you need a bit more to be sure that you not just disappointed that nothing happens.

[N0STIE]

  Looking at your stats from yesterday you may need to cap at 30000 to be sure to recover fast path.  How much of cap you need depends on the nature of the noise on your line and it looks like you need a bit more to be sure that you not just disappointed that nothing happens.

I'll give it few more days. Since 2am there was no single DS ES.

[N0STIE]

Can anyone look at my stats and confirm that everything is fine? Uptime just over 4 days and only 83 DS ES

xdslcmd info --stats
xdslcmd: ADSL driver and PHY status
Status: Showtime
Retrain Reason:   0
Last initialization procedure status:   0
Max:   Upstream rate = 8091 Kbps, Downstream rate = 48548 Kbps
Bearer:   0, Upstream rate = 7934 Kbps, Downstream rate = 34998 Kbps

Link Power State:   L0
Mode:         VDSL2 Annex B
VDSL2 Profile:      Profile 17a
TPS-TC:         PTM Mode(0x0)
Trellis:      U:ON /D:ON
Line Status:      No Defect
Training Status:   Showtime
      Down      Up
SNR (dB):    8.4       6.1
Attn(dB):    21.8       0.0
Pwr(dBm):    5.7       5.6
         VDSL2 framing
         Bearer 0
MSGc:      36      26
B:      31      238
M:      1      1
T:      64      53
R:      10      16
S:      0.0290      0.9577
L:      11568      2130
D:      1123      1
I:      42      255
N:      42      255
         Counters
         Bearer 0
OHF:      131744877      1299211
OHFErr:      183      55
RS:      3345256660      2532578
RSCorr:      1080374      422
RSUnCorr:   4870      0

         Bearer 0
HEC:      709      0
OCD:      36      0
LCD:      36      0
Total Cells:   3323036857      0
Data Cells:   1103322147      0
Drop Cells:   0
Bit Errors:   0      0

ES:      83      49
SES:      0      0
UAS:      60      60
AS:      368805

         Bearer 0
INP:      3.50      0.00
INPRein:   0.00      0.00
delay:      8      0
PER:      2.79      12.73
OR:      120.03      20.09
AgR:      35117.67   7954.34

Bitswap:   226807/226839      39/42

Total time = 1 days 6 hours 29 min 8 sec
FEC:      1080506      422
CRC:      183      55
ES:      83      49
SES:      0      0
UAS:      60      60
LOS:      0      0
LOF:      0      0
LOM:      0      0
Latest 15 minutes time = 14 min 8 sec
FEC:      4989      0
CRC:      2      0
ES:      1      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:      1610      0
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 = 6 hours 29 min 8 sec
FEC:      99054      24
CRC:      11      3
ES:      5      3
SES:      0      0
UAS:      0      0
LOS:      0      0
LOF:      0      0
LOM:      0      0
Previous 1 day time = 24 hours 0 sec
FEC:      226312      128
CRC:      38      10
ES:      17      9
SES:      0      0
UAS:      0      0
LOS:      0      0
LOF:      0      0
LOM:      0      0
Since Link time = 4 days 6 hours 26 min 45 sec
FEC:      1080374      422
CRC:      183      55
ES:      83      49
SES:      0      0
UAS:      0      0
LOS:      0      0
LOF:      0      0
LOM:      0      0
#

[les-70]

  Unfortunately I don't think the exact return to fast path ES limit is known.  You had 27ES one day, that may be bit high or OK?  The overall average ES rate is not key just the max number you get in one day during the return to fast path interval.    You may be alright at the moment but a slight extra speed cap would make it more certain especially if you get more errors on an odd day.