Print

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

[kitz]

I pondered for a mo on UPBO based on the bit load graph, but I dont think so.  The SNR per tone overlay is way too spikey.  If it was a wrong PSD mask then the SNR per tone would still have a fairly smooth shape in U1.

[Chrysalis]

Yeah I have had another look, the QLN data shows a clear increase  on the upstream, a bit on the downstream also which suggests a new disturber went live, but the unstable nature from one tone to the next seems odd.

[tiffy]

The plot thickens:

Substituted my ZyXEL VMG1312-B10A for a Billion 8800NL-R2 I had on hand, known to be healthy when last used.
Certainly produced changes on my line:

Gained some more US synch speed but still half of original.
Lost close to 3 Mbps DS synch speed.
Good news, the US SNRM oscillations have disappeared.
Bad news, I now have DS SNRM oscillations with severe error rates, after 4 hours synch, > 10K. ES's and > 700 SES's, increasing all the time, will very likely get DLM intervention tonight.
G.Inp (ReTx Low) is still active.
I have never at any time seen DS errors of this magnitude on my line which has been monitored 24/7 since migration from ADSL to VDSL.

Graphs attached for reference.

Really not sure where to go from here.
Should I disconnect and "rest" again before the ES's/SES's get out of hand ?
Re-install the ZyXEL 1312A and live with the lesser impact of US speed loss and SNRM oscillations.

[burakkucat]

I am unable to think of any reason why you are seeing that result.   

I do not know what next to suggest.

[tiffy]

Yes, it just gets more and more "illogical" by my standards of logic anyway.

Removed the Billion modem/router again, was being crucified by DS errors, well over ILQ "red" now, will not go un-noticed by DLM I'am sure !

Did have a peek inside the ZyXEL as suggested, no obvious visual signs of swollen electrolytic or blown tantalum capacitors.

ZyXEL 1312A back on line again, after 30+ min's down time interval.
DS synch speed is back to original, 40 Mbps.
US synch speed is close to 3 Mbps, still half of original.
Believe it or not, US SNRM oscillations have disappeared, flat lining @ expected SNRM !!
DS error rates are back to expected very low values.

To be honest, I never tried the Billion modem/router on my line before, appears not to be best suited or faulty ?

Attached, current SNRM & bit loading graphs for reference.

This is likely as good as it gets for now, will have to live with lower US speed and hope the US SNRM oscillations don't return.

Still wondering why I'am now getting the destinct "blips" on US1 as seen on QLN graph which would appear to be the cause of the reduced US bit loading.
Will attach before & after graphs in next posts for comparison.

[tiffy]

For comparison, QLN and bit loading graphs from before the disturbance event.

[tiffy]

For comparison, QLN and bit loading graphs from after the disturbance event.

[burakkucat]

If it will not be too inconvenient, I suggest that you power down the CPE, disconnect the patch cable and allow the line-card port to "relax", once  again, over night. You might then just get away without any DLM action.

[tiffy]

If it will not be too inconvenient, I suggest that you power down the CPE, disconnect the patch cable and allow the line-card port to "relax", once  again, over night. You might then just get away without any DLM action.

Yes, will certainly do that when finished off for the day to try and avoid the "wrath" of DLM.
My brain could do with the rest also !

Still awaiting any response from Plusnet, won't be holding my breath, they now appear to be offering tech. support via telephone help line only which is to say the least difficult and time consuming to access with limited prospect of success for the very patient.
The PN forum until recently produced the best chance of tech. assistance and fault resolution, active staff members now appear to be few and far between, obviously not helped by the current crisis.

[ejs]

Perhaps someone else on your cabinet (on a shorter line than yours) has got a modem on which they've switched off UPBO.

[tiffy]

Obviously I'am no expert on UPBO considering I had to ask what it meant yesterday !

From my limited understanding of the parameter:
UPBO would be applied by the DSLAM on an individual line basis.
Yes, the absence of UPBO on a cabinet line could effect other lines in the cabinet, suppose that's why it's there in the first place.
Is it possible for an end user to switch off UPBO on their line, I would have though this would be a DSLAM function only and totally outwith individual user control ?

I'am sure some of the better informed forum patrons will be able to advise on the finer details of the subject.

[ejs]

Ultimately it's the transmitting end that is in control of what it transmits, and for the upstream, that's the modem. Yes there are those MediaTek modem based Asus models like the DSL-AC68U which have a setting to switch off UPBO.

[Alex Atkin UK]

Ultimately it's the transmitting end that is in control of what it transmits, and for the upstream, that's the modem. Yes there are those MediaTek modem based Asus models like the DSL-AC68U which have a setting to switch off UPBO.

I'd imagine this is like the setting to disable regulatory compliance (only less severe if you get caught).

If your telco then kicks you off the network or fines you for "disrupting other users" then you'd be SOL.

[tiffy]

Ultimately it's the transmitting end that is in control of what it transmits, and for the upstream, that's the modem. Yes there are those MediaTek modem based Asus models like the DSL-AC68U which have a setting to switch off UPBO.

Many thanks for the additional information, as I said, I'am certainly no expert but always keen to learn and very much appreciate the support from forum members.

Thankfully still stable tonight for whatever reason all be it with the 50% loss of US synch speed.
Just going to shutdown the system over night as suggested by B*cat to try and mitigate DLM action due to the very large error count accumulated today with modem swapping.

[tiffy]

After the over night rest period, re-connected this morning.
Have lost a wee bit more US synch speed and although DS synch speed is normal G.Inp Profile has changed to Re-Tx high, if that's my only DLM punishment I can live with that, should eventually change back to Re-Tx low as DS error rate is now zero again.

Glad to say the US SNRM oscillations have not returned, DSLStats graph practically flat lining again.

Would still really like to know the source of the line disruption resulting in loss of US synch speed, change in QLN U1 profile & much decreased bit loading (all of which still remain) and the strange US SNRM oscillations which thankfully have now gone away.

Again, many thanks to all for the help and support offered to date, much appreciated.