Print

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

[adslmax]

I got hit by the DLM when testing the TP-link,  I just thought 'grrrrr tough luck'. Theres nothing either I nor Plusnet can do about it so I just waited for it to go back to normal.

Bingo....I think it down to that TP Link put DLM on my line. Thanks Kitz. I am now on day 1 DLM. Another 13 more days to go (hopefully DLM removed) as long my line stay connection (PPP)

[kitz]

kitz - plusnet had a gateway problem last week on bng gateway. Sometimes the ping is higher on some gateway, that's why i change to different gateway to get lower ping. Plusnet told me to stop doing this as any PPP drop will trigger DLM even worse on the line.

I did request plusnet for a connection log plus GEA FTTC Test result too.

I think DLM hit because of me doing this:

Testing BT Home Hub 5 (VDSL2 Modem)
Testing Netgear DGND3700v1 (VDSL2 Modem)
Testing TP-Link TD-W9980 (VDSL2 Modem)
Testing Billion 8800NL (VDSL2 Modem)
Back to normal BT Openreach Modem (VDSL2 Modem)

I know about the PPP issue on Monday.. Its the 11 new sessions on Wednesday I was on about.   5 looked to be a full resync but I couldnt tell for the others as they were too quick.  How did you pipe hop?  Did you specifically only change PPP session from within the router..  or did you do a full resync?

[adslmax]

I know about the PPP issue on Monday.. Its the 11 new sessions on Wednesday I was on about.   5 looked to be a full resync but I couldnt tell for the others as they were too quick.  How did you pipe hop?  Did you specifically only change PPP session from within the router..  or did you do a full resync?

I just do a quick reconnection to gateway using router PPP session soft disconnection. Not the VDSL2 Modem resync.

[JustAnother]

I would be very interested in a reliable source for the statement that ppp disconnects are part of the DLM logic - as mentioned earlier, its completely the wrong layer for a 'line quality' assessment.

[burakkucat]

Plusnet told me to stop doing this as any PPP drop will trigger DLM even worse on the line.

It will not. Whoever you were talking to at Plusnet was using her/his anal orifice rather than the conjunction of her/his brain & mouth. The Openreach DLM operates at the lower level (of synchronisation) between the CO and CPE (i.e. DSLAM and EU's modem) and not at the higher level of the PPPoE session!

If a member of Plusnet support staff still persists in disseminating that utter hogwash, please make a note of the persons name and report it back here. It could well be that you were talking to a new Plusnet employee who has not yet finished the training course.

[les-70]

  I had forgotten that I am out this evening so I have cut short the tests of the 8800NL with the original 038 dsldriver.   The results obtained in about 21 hours are similar to those I obtained in early tests when I first had the 8800NL.  in reply 104 above I noted that

The HG612 gave 10.5 errored sec/hour  c.f. 15.5 with the 8800NL (with 039 dsl driver) -- and --For CRC and SES with difference is not just a factor of ~1.5 but a factor of  ~4.

  I now have 16.4 errored secs/hour with the 8800NL (with the 038 driver) i.e. basically the same as with the 039 driver . The CRC and SES are however much less than with the 039 driver and more like the HG612. In one 24 hour period this may be luck but it is consistent with past tests. The result with the 039 dsldriver for CRC and SES could equally be bad luck but again they like past tests.  When I am really interested I usually like to do the 24 hours test alternately at least 5 times for each device.

  As in the precious post 104 the  abbreviated "info --stats" is attached

[adslmax]

Great news DLM had now been removed overnight. It's took me just 2 days from DLM to Non-DLM. So, I better be careful next time. My speed still 67 Meg as I can see Plusnet Profile still on old one 70.6 Mb as BTw profile report 77.4 Mb. I guess I have to wait for Plusnet profile to be updated.

Updated: Plusnet profile had been updated now match BT 77.4 Mb





[attachment deleted by admin]

[kitz]

 

[adslmax]

Thanks Kitz. From now must remember power down the vdsl2 modem for 30 minutes. Power down VDSL2 first then wait for few seconds, then unplugged modem cable. So, DLM won't be affected. Mine took just 2 days before DLM was removed. That's was quick.

[les-70]

 I will shortly be repeating my 8800NL tests to see if the results are the same on my second try but it will be harder to judge now as one of my cross talk sources that turns on and off from time to time seems to have ended a 3 week absence!!  I attach a fairly recent plot of attainable speeds on my line starting when only about 2 other people had FTTC.  It misses my initial attainable of 107. All stats are with the HG612 as when testing other devices I did not collect the data.  It will probably give some insight into why I have been capping my speed, as without capping, rather than vary from 10 down to 7 my snrm might vary from 6 to 3.  On many days the variation is a bit less but often at least 2db and sometimes 3db but the changes are not very predictable. 

 The new test will be done with a Mark 1 SSFP as I have ben testing SSFP's over the last few says and concluded that the Mark 1 or old NTE-2005 work best for me.  These two SSFP's are esentially identical to a Mark 2 and Mark 3  IF the built in "RF3" type toroid of the Mark 2 and 3 is removed.  I gain an extra ~1Mb/s with no extra errors and possibly one ot two less.  (see the Kitz SSFP Mark3 thread for details)

[kitz]

Thanks for the update les-70.   Im interested in seeing your findings.   Anything that can throw light on why some chipsets behave better on some lines for some and not on others would be interesting..  as would any observations regarding crosstalk.   

[adslmax]

unfortunately I cannot test mine with billion 8800NL as it already gone back for a full refunded into my account today.

[Bald_Eagle1]

I attach a fairly recent plot of attainable speeds on my line starting when only about 2 other people had FTTC.  It misses my initial attainable of 107. All stats are with the HG612 as when testing other devices I did not collect the data.  It will probably give some insight into why I have been capping my speed, as without capping, rather than vary from 10 down to 7 my snrm might vary from 6 to 3.  On many days the variation is a bit less but often at least 2db and sometimes 3db but the changes are not very predictable. 

Do any or all of those reductions in attainable rates tie in with actual sync speeds or are they purely due to reductions in SNRM?

If purely due to SNRM reductions, do they tie in with phone calls (incoming or outgoing)?

[les-70]

  They don't tie in with any fault condition and I am certain that it just varying cross talk with people turning modems on and off.  Look at today's snrm.  The attainable matches it.

Lately I tend to cap my sync a bit below the minimum that occurs every week or so. That way the connection remains stable and on fast path.   

[kitz]

I am certain that it just varying cross talk with people turning modems on and off

Thats the problem with fttc - we dont know if it is x-talk.. or not   Once upon a time we would have said it was a small amount of REIN.