Print

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

[orainsear]

>>Vendor ID: P----<<

Which manufacturer does this code represent?

[kitz]

Looks like the 2 lines are on different dslams/msans then if they are going via different RASes.

Vendor ID: P----

Ive no idea what that is Ive never seen 'P' before

[waltergmw]

@ Kitz,

It's not a DMT display bug as the 585V7 Link Information gives the same
Vendor ID (Local/Remote): TMMB/P
which presumably is where DMT is getting the same information in any case.

I do wish BT weren't quite so secretive !

Kind regards,
Walter

[waltergmw]

@ Kitz,

Do you know you might just be right re your suspicions !

I've just captured the attached noise margin graph which still shows bursts of noise but at much reduced levels.
I've asked James whether they fitted an AR3 but he's yet to reply.

From BT O's viewpoint they have made the line quite usable so unless I can grab the engineer again I don't suppose we'll ever find out.

Kind regards,
Walter

[attachment deleted by admin]

[kitz]

Ive done a search and cant find anything at all about who 'P' is.

There are a couple of 'P's that crop up but there is so few of them - one of them is here.  Revs per min is one of those names that crops up as generally knowing what he's talking about.

One thing that could be a longshot and this is a very wild guess... but the CLI switch for configuring the Vendor ID and Product ID is

-o vendor id [ -p product id ]

Wonder if its just a simple case someone has not configured the vendor id correctly when setting up the dslam/msan?

[waltergmw]

Hi Kitz,

I'd believe anything's possible in sunny Ewhurst !

We've just had the modem drop out and resync at 576 kbps and the noise margin variations are worse than the Blackpool roller-coaster !

Kind regards,
Walter

[orainsear]

Wonder if its just a simple case someone has not configured the vendor id correctly when setting up the dslam/msan?

Could well be.  According to this thread P designates unknown, so it looks like it's either the hardware ATUC has not been configured properly or it's a new manufacturer.


Walter:  Have you had any reports of problems from other users of cabinets 19&20?

[waltergmw]

@ orainsear,

I've got one other which I suspect is on cabinet 19. However I am about to circulate my distribution list of about 130 homes to see if anyone else is suffering.

My guess is that there's something around the BT parts of THCN which is causing the noise problem as we see similar problems on different DSLAMs.
I expect all the E side cables run down the same set of ducts for the first 3.5 km at least, but as they must be screened it seems unlikely that they're causing the grief.
The user I'm currently helping has another LLU with TalkTalk but that seems O.K.

I've attached a picture of the last drop which mirrors the types of noise margin graph I've seen many times before.

Kind regards,
Walter

[attachment deleted by admin]

[waltergmw]

@ kitz

How about this for a nightmare ? Do you know of any software that would let us split the noise margin down into single or groups of tones?
Would that give us a better idea of what's going on?

Kind regards,
Walter

[attachment deleted by admin]

[roseway]

With supported routers, DMT displays the SNR per tone in the second graph. But in practice (on my system anyway) the second graph is almost exactly the same shape as the first graph (bit loading per tone), so maybe you can infer what you're looking for from that.

[kitz]

>> Do you know of any software that would let us split the noise margin down into single or groups of tones?

Unless anyone knows better - I dont Im afraid - well not that would graph it.

DMTtool graphs the tones and if youre monitoring it live - and you have the right router such as a ST585 v6 - then you can sometimes see the shift in the SNRM for the rough area of the tone.


------------------------

/sorry eric I'd started to type this - then got distracted by something else Im trying to sort, but I posted anyhow.

"Warning - while you were typing a new reply has been posted. You may wish to review your post. "

[orainsear]

A 585v6 would be a good aquisition - it's a good bit of kit and with the right firmware would probably serve you better than the 585v7 for fault finding with DMTtool.

Do you have a have a DMTtool graph of the frequency spectrum for the line that you are currently monitoring?

[waltergmw]

Herewith pictures of my current DMT system, viewed remotely.

I had, perhaps wrongly, assumed that I was limited by using the 585V7 and "allow untested firmware versions"

All suggestions gratefully received.

I will certainly look for a 585V6 and will be pleased for any advice on suitable firmware.

Kind regards,
Walter

[attachment deleted by admin]

[orainsear]

I will certainly look for a 585V6 and will be pleased for any advice on suitable firmware.

There's usually a few on ebay.  If you do decide to buy from ebay get a non-ISP branded one because some of the ISP supplied units are locked to their own services and won't easily flash with standard firmware.

In their wisdom (under pressure from certain ISPs) Thomson crippled the 'expert modes' in their firmwares (commands in the CLI were disabled - the same commands DMTtool uses).  The last best firmware for the 585v6 is release 6.1.4.3 and if you check the sticky thread in this forum you can find a link to it.

[waltergmw]

Thanks orainseer,

I shall persue this activity shortly but I have to keep on top of the immediate problem at present.

@Kitz

I've now discovered that the first BT O investigation did not actually change the line card but they did a lift and shift which effectivly uses a different line card.
We thought that had cured the problem due to the sporadic nature of fluctuating noise.

The next BT O visit changed the E side pair and by coincidence another BT O team cured a REIN fault in the cabinet which was affecting many users.
This had the appearance of no real change on our lines.

Just to be absolutely sure I changed the 585V7 for a DG834V4 and am using routerstats only to monitor the noise margin and provide sync speed.
The noise curves are slightly different as the Netgear reports chages down to 0.1 dB but the general appearance is very similar with long periods of calm followed by wild fluctuations and usually causing the modem to re-sync eventually - several times.

This afternoon my laptop was left on running Routerstats lite, but otherwise untouched by human hand, I noticed that noise line had become almost absolutely flat.
I am suspicious as I don't believe that the real noise margin could possibly remain so for long periods.
I am therfore questioning what causes these major changes in operating parameters and subsequent re-syncs?
Could it be with zero real traffic the Tone noise margin summation logic goes into sleep mode as there's nothing there to detect ?
I also note the latest disturbance happened shortly after I had used my remote access software to examine Routerstats, which obviously generated some real internet traffic.

Thankfully this householder has two ADSL lines, the other being directly connected to TalkTalk's LLU system.
That noise margin seems to behave in a similar but less-pronounced way.

We are asking BT to continue their investigations and changing the D side pair, if others are available.

I've attached pictures to illustrate this very strange situation, together with one of my 585V7 connected on the TalkTalk line.
Any ideas would be most welcome.

Kind regards,
Walter

[attachment deleted by admin]