Print

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

[eliw]

True but had i left it at 5.1 it would have dropped by another db today then another the day after and a resync would have occured... a few of these and ip profile is affected (IMHO).

Can't wait for March 2013

[Bald_Eagle1]

True but had i left it at 5.1 it would have dropped by another db today then another the day after and a resync would have occured... a few of these and ip profile is affected (IMHO).

But wouldn't a prolonged test such as that actually confirm beyond any doubt that there is still an ongoing issue/fault that experienced engineers would be able to easily diagnose & hopefully repair?

Such similar symptoms as SNRM graudually dropping over a number of days, disconnections when using the phone etc. were resolved on my connection via a combination of replacing the SSFP (my own test with a dangly filter identified that it was faulty) & remaking a connection in the DP at the top of the pole across the road from my house.

TBH, I believe replacing the SSFP had the greater instant impact & remaking the DP connection has hopefully finally cured my connection's general, but ongoing intermittent instability issues.

Gradually dropping SNRM suggests (to me) that resistance is gradually building over a number of days, only being "released" by physically rebooting/disconnecting & reconnecting the modem.

My SNRM used to immediately increase by quite a few dB when whatever was causing the issue was "released" by the telephone ringing pulses when dialling in.
Sometimes this would also cause an immediate connection drop, with things returning to normal for a few hours at a time.

FWIW, see my attached 8 day graphs.
The before & after conditions are clear to see.
There has only been one connection resynsc that was not initiated by me.
That was when DLM decided that my connection is now sufficiently stable to turn off Interleaving, INP & delay (unless it is just lulling me into its regular false sense of security that it has had so much fun with over the last few months.

DS error seconds have increased (to be expected I suppose, due to Interleaving being turned off)
DS Output is also fluctuating more than what I would class as "normal".
Other than those, it all looks quite promising.

If my connection is indeed now permanently fixed (a couple of weeks monitoring will confirm one way or the other), it will just show how inadequete the fault finding efforts of the previous visiting engineers have been.

Have you tried the ongoing graphing scripts yet?
They can be a lot more revealing (& useful when trying to "prove" that fault conditions still exist than trying to study numbers in a chart/list.

[snadge]

@ eliw -  have you had this problem ongoing where the SNRM dips continually by say 1db per day too the point it drops a 3-4-5 days later? ,
- if not then I think you will find if it dropped to 5.1db that it would probably stay 'about' there OR/THEN eventually climb back up... but then again Iam unfamiliar to what has been going on in your case, so I apologise in advance if this is whats been happening in your case...

but normally, just because it dipped by say 1db one night from 6db to 5db... does not mean it will drop from 5db to 4db the next, then 4db to 3db the night after that etc..and continue to do so , this just means you have an average noise increase to force a 1db reduction in the SNRM - now you have rebooted and forced the SNRM back up then its likely that it may drop back down to 5db again, when the evening comes around the noise on the line increases and so the margin dips slightly, it is supposed to go back up but if it hasnt then it means the noise is still there at that amount OR it is simply 'stuck' - my SNRM seems to do it AND Bit-Loading seems to, I will have lots of tones missing in the Bit-Loading Map and my SNRM maybe 1-2db lower than default...when i reboot I get the tones back but the speed remains about the same because the SNRM is forced back up to 7db - later on it will drop back down to 5-6db , but it wont go any lower if I left it.

I would have left it... not rebooted, and see if it dropped further tonight, and again the next night... it will now that you have re-set it

p.s. I took out my SSFP and put in the FFP (made by same company) and results are the same for me, I will try a sky filter tomorrow

[eliw]

Cheers for the advice guys.

I have completely removed the SSFP and conected the router straight into test socket without any filter or phone - who needs customers... !! 

It is now back down to 5.3 and will most probably end up around 5.1 by the morning and stay there. And I agree I should have left it there to see if it will drop any lower the next day/evening.

Baldeagle- Yes I have started collecting continuous stats - have a look below. I am nearly convinced that the problem is at the top of the pole now .. if you read my first message in this thread you will see that I thought that was the case when the line was changed about 2 years ago now.

ABout the continous stats ... for some reason the max I can do is 20h even though I know I have about 26h or so of logged data.. but if I try anything above 20h I get : line 0: illegal day of month (I have deffo set the date and not rebooted at least around 27h ago, maybe I have an old version of the scripts ?). As you can see in the graphs I forced resync today around 15.30 - and the other thing is the higher than normal error count this evening ... quiet all day then suddenly a rush of errors (it was not the case yesterday)

I should have started to log the stats when I started this thread

[Bald_Eagle1]

ABout the continous stats ... for some reason the max I can do is 20h even though I know I have about 26h or so of logged data.. but if I try anything above 20h I get : line 0: illegal day of month (I have deffo set the date and not rebooted at least around 27h ago, maybe I have an old version of the scripts ?).

I will shortly be releasing updated versions (still tweaking them).

The illegal date message is usually only seen when there is insufficient data for the period requested.
If you wanted to post a copy of modem_stats.log (maybe zip it or rename it to a compatible format first), I'll have a look at my end for you.

[burakkucat]

I am nearly convinced that the problem is at the top of the pole now .. if you read my first message in this thread you will see that I thought that was the case when the line was changed about 2 years ago now.

@eliw -- Will you confirm the route of the last few lengths of your D-side pair for me, please? It appears from below ground to climb a pole on the north side of H . . . Road, as an aerial cable it crosses H . . . Road and enters P . . . Avenue to a pole top on the western side of the Avenue. All of the previous is quite clear. What follows is uncertain. Are you fed by a drop cable from a DP at the top of the second pole? Or does the cable climb down the pole to return underground and then is distributed radially from a DP in a joint chamber?

I'm trying to understand which cable was brought down by the falling bough from a tree . . . your individual drop cable, after the DP or the aerial D-side bundle linking the two pole tops.

[eliw]

Yes it goes underground between the second pole on P... avenue and the 3rd pole (our pole) then distributed radially from that 3rd pole ... the cable that was brought down is the cable that goes from the top of that pole to our property.. only our cable was brought down.

Our pole is at the intersection between our road (P... avenue) and Park Avenue.

[burakkucat]

Ah, I see. Got it.   So we need to factor-in the average height of three poles when calculating the length of your D-side pair.

did a TDR without me asking him ... used two types of machines one with knobs (which he called old school) and an exfo - both showed a huge spike at around 65m

I've been trying to judge where 65 metres from your NTE5/A would be.    Not too relevant -- just interesting to an inquisitive feline.

The "doings" with the knobs on must have been a Tester 301C . . . Yay! . . . 'cause Walter, Bald_Eagle1 and b*cat each possess one. Asbokid, being nonconformist, has something completely different. 

[asbokid]

BT Openreach engineers don't know how lucky they've got it with their dainty little test kits.  The Mole, the Hawk and the JDSU are all light enough that even a bantum-weight engineer can haul them about and, when called upon, could shimmy up a pole with them, too.

Unlike their colleagues in America who always go for the larger things in life.  Double BigMacs with extra fries and relish,  seven liter V8 Humvees,  84 inch plasma TVs to show Michael Moore in ultra widescreen,  etc..

If it's big and if it's fat -  it's American.

It's just the same when it comes to American telecoms test kit.   Until recently, the standard issue TDR tester supplied to AT&T linesmen was about as portable as a large car battery. Which is no exaggeration since the AT&T tester is actually powered by a heavy lead acid battery.   The nylon carrying strap behaves like cheesewire on the shoulderblades 

cheers, a

[snadge]

what are the purposes of these:
The Mole, the Hawk

i found this video about the JDSU, good for someone like me who knows nothing about it
http://article.wn.com/view/2010/01/27/JDSU_Selected_by_BT_to_Provide_HST3000_Broadband_Access_Test/

the JDSU is just like a massive SKY+ remote hehe

[Black Sheep]

Ah, I see. Got it.   So we need to factor-in the average height of three poles when calculating the length of your D-side pair.

did a TDR without me asking him ... used two types of machines one with knobs (which he called old school) and an exfo - both showed a huge spike at around 65m

I've been trying to judge where 65 metres from your NTE5/A would be.    Not too relevant -- just interesting to an inquisitive feline.

The "doings" with the knobs on must have been a Tester 301C . . . Yay! . . . 'cause Walter, Bald_Eagle1 and b*cat each possess one. Asbokid, being nonconformist, has something completely different. 

Just to clarify B*Cat, the 'twisting of knobs' (O-err missus) will not be done on a 301C Meter. It could have either been the old Ohmeter 18C, or the even older oscilloscope's we used to use ........ the latter pinpointing the smallest of an HR every time !!

 

[Black Sheep]

BT Openreach engineers don't know how lucky they've got it with their dainty little test kits.  The Mole, the Hawk and the JDSU are all light enough that even a bantum-weight engineer can haul them about and, when called upon, could shimmy up a pole with them, too.

Unlike their colleagues in America who always go for the larger things in life.  Double BigMacs with extra fries and relish,  seven liter V8 Humvees,  84 inch plasma TVs to show Michael Moore in ultra widescreen,  etc..

If it's big and if it's fat -  it's American.

It's just the same when it comes to American telecoms test kit.   Until recently, the standard issue TDR tester supplied to AT&T linesmen was about as portable as a large car battery. Which is no exaggeration since the AT&T tester is actually powered by a heavy lead acid battery.   The nylon carrying strap behaves like cheesewire on the shoulderblades 

cheers, a

Asbo, haven't got the time to 'Google', but isn't the JDSU an Americanised bit of kit ?? Seems odd that they're shuffling around with large wardrobes, for testers when they have these on their doorsteps ?? 

The only reason I have for believing they're American, is that all our literature regarding the testers talks about "Tip to Ring". This is a Yankeeeee term for our A and B wires. 

[snadge]

BT Openreach engineers don't know how lucky they've got it with their dainty little test kits.  The Mole, the Hawk and the JDSU are all light enough that even a bantum-weight engineer can haul them about and, when called upon, could shimmy up a pole with them, too.

Unlike their colleagues in America who always go for the larger things in life.  Double BigMacs with extra fries and relish,  seven liter V8 Humvees,  84 inch plasma TVs to show Michael Moore in ultra widescreen,  etc..

If it's big and if it's fat -  it's American.

It's just the same when it comes to American telecoms test kit.   Until recently, the standard issue TDR tester supplied to AT&T linesmen was about as portable as a large car battery. Which is no exaggeration since the AT&T tester is actually powered by a heavy lead acid battery.   The nylon carrying strap behaves like cheesewire on the shoulderblades 

cheers, a

Asbo, haven't got the time to 'Google', but isn't the JDSU an Americanised bit of kit ?? Seems odd that they're shuffling around with large wardrobes, for testers when they have these on their doorsteps ?? 

The only reason I have for believing they're American, is that all our literature regarding the testers talks about "Tip to Ring". This is a Yankeeeee term for our A and B wires. 

I was thinking that myself, having searched ebay for them they are all mostly American auctions

[asbokid]

Asbo, haven't got the time to 'Google', but isn't the JDSU an Americanised bit of kit ?? Seems odd that they're shuffling around with large wardrobes, for testers when they have these on their doorsteps ?? 

The only reason I have for believing they're American, is that all our literature regarding the testers talks about "Tip to Ring". This is a Yankeeeee term for our A and B wires. 

Stop spoiling the story, Black Sheep!

[Black Sheep]

Asbo, haven't got the time to 'Google', but isn't the JDSU an Americanised bit of kit ?? Seems odd that they're shuffling around with large wardrobes, for testers when they have these on their doorsteps ?? 

The only reason I have for believing they're American, is that all our literature regarding the testers talks about "Tip to Ring". This is a Yankeeeee term for our A and B wires. 

Stop spoiling the story, Black Sheep!

Whoops .....  I'll go back to my grazing.