Print

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

[kitz]

Is it possible that my line banding won't be removed?

It should do eventually... but by all accounts it would appear that you invoked ILQ of Scarlet which does appear to take longer to get rid of.   We more or less know what causes DLM to take action, but what we are unsure of is the recovery process.
It certainly seems to use some sort of doubler method for ILQ red -  Ive been ILQ red myself a couple of times after line faults and once when beta testing a new router...  on all occasions it recovered by itself over the course of a week or so.

You really went in for the kill though by doing rapid and many disconnects - which is what triggers scarlet.  Not only did you do it for one day, you did it for two in a row, so DLM will sense something really serious.   
Ive seen a few cases in the past whereby people swapping routers has invoked DLM after about only half a dozen disconnects.  adslmax is one of them, and iirc it took a couple of days for DLM to adjust back to normal and the MTBR has also been increased since then.

I dont think anyone has had the nerve before to go for the full 20 because we fully suspected that it could take a while for DLM to step back down.   You were warned.
Whilst we are fully aware of what can trigger DLM, the missing piece of the jigsaw is recovery.... particularly in the event of MTBR & ILQ Scarlet. 

 

[William Grimsley]

Is it possible that my line banding won't be removed?

It should do eventually... but by all accounts it would appear that you invoked ILQ of Scarlet which does appear to take longer to get rid of.   We more or less know what causes DLM to take action, but what we are unsure of is the recovery process.
It certainly seems to use some sort of doubler method for ILQ red -  Ive been ILQ red myself a couple of times after line faults and once when beta testing a new router...  on all occasions it recovered by itself over the course of a week or so.

You really went in for the kill though by doing rapid and many disconnects - which is what triggers scarlet.  Not only did you do it for one day, you did it for two in a row, so DLM will sense something really serious.   
Ive seen a few cases in the past whereby people swapping routers has invoked DLM after about only half a dozen disconnects.  adslmax is one of them, and iirc it took a couple of days for DLM to adjust back to normal and the MTBR has also been increased since then.

I dont think anyone has had the nerve before to go for the full 20 because we fully suspected that it could take a while for DLM to step back down.   You were warned.
Whilst we are fully aware of what can trigger DLM, the missing piece of the jigsaw is recovery.... particularly in the event of MTBR & ILQ Scarlet.

Thanks, that's really interesting.

I'm sorry for having to make you give up your time to write posts about something that I've caused, I feel bad. :/

[NewtronStar]

William could I ask a question here was your sync rate banded before your started the forced manual disconnects ?

[William Grimsley]

William could I ask a question here was your sync rate banded before your started the forced manual disconnects ?

No, it wasn't banded.

[NewtronStar]

William could I ask a question here was your sync rate banded before your started the forced manual disconnects ?

No, it wasn't banded.

Ok I see well it's up to you now wait until your sync changes and if your like me days can seem like months & months are like years when your waiting for DLM to do something.

Oh the DLM does not hang about to intervene when the users actions or hardware is at fault it has been said here in the past the DLM is just way to sensitive when things go skew whiff over 24 hours.

[jid]

One point I will make - DSL Stats are great, but to be honest when I had issues with my line, watching the stats all of the time was consistently getting to me.

Therefore, by just ignoring them and carrying on with things was better - the line eventually fixed itself. I don't know the full background into this thread, but it does seem like constantly watching and monitoring the line wont help it get fixed any quicker

[burakkucat]

. . . the last retrain was an RDI, what is that?

As mentioned, above, an RDI is a Remote Defect Indication.

I have found the relevant section of ITC-T Recommendation G.992.3 for you --

Section 8.12.1 xDSL line related primitives

. . .

Severely errored frame (SEF) defect: An SEF defect occurs when the content of two
consecutively received xDSL synchronization symbols does not correlate with the expected content
over a subset of the subcarriers. An SEF defect terminates when the content of two consecutively
received xDSL synchronization symbols correlates with the expected content over the same subset
of the subcarriers. The correlation method, the selected subset of subcarriers, and the threshold for
declaring these defect conditions are implementation discretionary.

. . .

Remote Defect Indication (RDI): An RDI defect is an SEF defect detected at the far-end and is
reported by the RDI indicator bit once per 15 to 20 ms (see Tables 7-8 and 7-15). The RDI indicator
bit shall be coded 1 to indicate that no SEF defect has occurred and shall be coded 0 to indicate that
an SEF defect has occurred since the last previous RDI indicator bit transmission. An RDI defect
occurs when a received RDI indicator bit is set to 0. An RDI defect terminates when a received
RDI indicator bit is set to 1.

. . .

A PDF copy of ITU-T Rec. G.992.3, which is entitled “Digital Line Systems – Access Networks (ADSL2)”, may be downloaded from this location.

(Although it was written for ADSL2, the basic principals are valid for VDSL2. One just needs to replace all mention of NSC=256 with NSC=4096. (NSC is the Number of Sub-Carriers.))

[William Grimsley]

. . . the last retrain was an RDI, what is that?

As mentioned, above, an RDI is a Remote Defect Indication.

I have found the relevant section of ITC-T Recommendation G.992.3 for you --

Section 8.12.1 xDSL line related primitives

. . .

Severely errored frame (SEF) defect: An SEF defect occurs when the content of two
consecutively received xDSL synchronization symbols does not correlate with the expected content
over a subset of the subcarriers. An SEF defect terminates when the content of two consecutively
received xDSL synchronization symbols correlates with the expected content over the same subset
of the subcarriers. The correlation method, the selected subset of subcarriers, and the threshold for
declaring these defect conditions are implementation discretionary.

. . .

Remote Defect Indication (RDI): An RDI defect is an SEF defect detected at the far-end and is
reported by the RDI indicator bit once per 15 to 20 ms (see Tables 7-8 and 7-15). The RDI indicator
bit shall be coded 1 to indicate that no SEF defect has occurred and shall be coded 0 to indicate that
an SEF defect has occurred since the last previous RDI indicator bit transmission. An RDI defect
occurs when a received RDI indicator bit is set to 0. An RDI defect terminates when a received
RDI indicator bit is set to 1.

. . .

A PDF copy of ITU-T Rec. G.992.3, which is entitled “Digital Line Systems – Access Networks (ADSL2)”, may be downloaded from this location.

(Although it was written for ADSL2, the basic principals are valid for VDSL2. One just needs to replace all mention of NSC=256 with NSC=4096. (NSC is the Number of Sub-Carriers.))

burakkucat,

Thanks for the information. I really appreciate it. But, where does it say this is something to do with DLM or not? I am still confused as to if this was a DLM action or a LOS?

[burakkucat]

The DLM process is just a software entity whose sole purpose it to try to maintain a circuit in a stable and operative state.

G.992.3 makes no reference to the DLM process, for the latter is a higher level protocol which operates "on top" of the lower level procedures to which G.992.3 will apply.

From the above document it can be seen that a RDI is just a report of a SEF at the "far end" from that at which the observation has been made.

[William Grimsley]

The DLM process is just a software entity whose sole purpose it to try to maintain a circuit in a stable and operative state.

G.992.3 makes no reference to the DLM process, for the latter is a higher level protocol which operates "on top" of the lower level procedures to which G.992.3 will apply.

From the above document it can be seen that a RDI is just a report of a SEF at the "far end" from that at which the observation has been made.

Ah right, so basically my line just dropped out twice on two consecutive days... Odd.

[NewtronStar]

Ah right, so basically my line just dropped out twice on two consecutive days... Odd.

How healthy was your DS & US SNRm dB when the line dropped out ?

[William Grimsley]

How healthy was your DS & US SNRm dB when the line dropped out ?

Healthy. I think Downstream SNR Margin was 9 dB and Upstream SNR Margin was 6 dB.

[loonylion]

I've been banded at 43999 for over a year and DLM seems to have no intention of lifting it even though my SNR varies between 8 and 9

[William Grimsley]

I've been banded at 43999 for over a year and DLM seems to have no intention of lifting it even though my SNR varies between 8 and 9

Was this a fault? If so, did you get an engineer visit and get him to reset DLM?

[William Grimsley]

Just had 4 more Upstream SES on my line.