Print

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

[tickmike]

What tests can a ISP do on a adsl2 line, list them and what they can test for.

[Weaver]

The tests listed on my ISP's control panel control.aa.net.uk are, amongst others:

• Traffic dump - captures all traffic on the link for so many sec, so it can be analysed, displayed and  downloaded
• Sync status - queries modem current status and displays some cumulative stats
  
• Copper line test - BT physical layer tests performed live by connecting test h/w
  
• Kill PPP - kills the current PPP session
  
• KBD diagnostics - BT knowledge-based diagnosis: mixture of question-and-answer session, history and various database queries
• Run PSTN test - BT hardware test and diagnostics
• Kill BT - don't know what that one is. Drops the DSL link (??)
  

I've only ever hit the buttons for a few of these

[burakkucat]

The Copper Integrated Demand Testing (CIDT) was rolled out in 2011 and should be available to all CPs/ISPs.

FAQ: Copper Integrated Demand Testing (CIDT)

The purpose of this document is to help to answer CPs questions about CIDT.

Contents:

Section 1.0 - Overview

1.1 What is Copper Integrated Demand Testing (CIDT)?

CIDT is an upgrade to the existing Basic Line Test Service which offers customers enhanced benefits over
the existing product.

Information can be found on the CIDT page of the Openreach Portal. This can also be accessed via the
Key Information section of the LLU product pages.

1.2 What enhancements does CIDT offer over the Basic Line Test?

CIDT improves the ability of Openreach and CPs to identify and locate end users’ service problems by
finding High Resistance (HR) faults (small and large) which can significantly impact broadband performance.
HR faults are not currently detected by the existing BAU Line Test capability as it tests through lots of
exchange equipment which masks and blocks the diagnostic signals on the line. EVOTAM and LLUTAM
provide an unrestricted view of the line and enable CIDT to detect HR faults. CIDT technical trials proved a
significant increase in certainty where the CIDT test showed Line Test OK (LTOK).

<snip>

Section 2.0 - Systems and processes

2.1 How does it work?

CIDT requires the end user’s modem to be attached to the end users computer, connected to the line being
tested and switched on as it uses the tones from the modem to measure the difference between wires in a
copper service pair.

Currently CPs submit a line test via Dialogue Services, as the basic line test cannot detect High Resistance
(HR) faults CP’s could receive a Line Test OK (LTOK) result on a line which has an HR fault. If a CP
receives a LTOK result and there are still problems with the line CPs can push through a visit request for a
repeat fault or request an SF12 visit which would be chargeable if no fault was found in Openreach’s
network.

With CIDT a LTOK result tells the CPs that there is either no fault on the line or that the problem lies in their
network or in the customers’ premises and that they should investigate further. CPs will still be able to
request SFI2 visits when required.

A Line Test Not OK (LTNOK) result means that the problem is probably in Openreach’s network, however,
CIDT can detect faults which may be in the customer’s premises and CPs should check that their customers’
kit and wiring are working properly before reporting the fault to avoid incurring unexpected charges.

CPs will need to advise end users that the test can take up to 90 seconds. This is longer than the Basic Line
Test but CIDT will detect HR faults and many underlying broadband performance faults.

<snip>

[I have two very basic Openreach PDF documents related to CIDT in The Cattery's library; sections of the first have been quoted, above. If anyone wishes to have a copy of both, please send me a PM containing a valid e-mail address that I may use to forward the two files.]

[tickmike]

PM sent.

[Weaver]

PM sent too

[burakkucat]

Acknowledging both requests have been received and copies of the documents have been dispatched, by return. 

[tickmike]

Thank you, interesting.

I'm going to ask my ISP if it was a 'Copper Integrated Demand Testing (CIDT)' test they have done this week and last week.

[tickmike]

I want to expand this question.

Can any test be done say in the PCP or other points a long E / D side or the exchange without disconnecting my connection ?.

[burakkucat]

Hmm . . . I don't know.

I would presume some tests could be performed by just "tapping onto" the pair. My feeling is that for a valid, reproducible, result the "extraneous" parts of the circuit should be disconnected.

Perhaps Black Sheep (or licquorice or . . . ) would please comment?

[tickmike]

Perhaps Black Sheep (or licquorice or . . . ) would please comment?

Yes I hope so.

I was thinking about the REIN team how they are monitoring my line while they are doing tests.
You can see that they did two copper line tests on this graph maybe from the exchange.

[kitz]

Perhaps Black Sheep (or licquorice or . . . ) would please comment?

Yes I hope so.

I was thinking about the REIN team how they are monitoring my line while they are doing tests.
You can see that they did two copper line tests on this graph maybe from the exchange.

Only just seen this thread.

Hmmm.  That looks a bit odd to me.   Not heard of them doing tests such as that .... especially out of hours.  Why or even how would they do something to cause a dip in the snrm.?

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

The suite of tools available to isps was called woosh.   Some of the tests are pretty basic with the latest addition being the  CIDT and part of the newer KBT suite.  I'm not at the pc atm as i have a copy of some kbt tests that i got plusnet to run for me so i coukd see what they look like.......but I'm sure there's a copy of the old version of the woosh handbook available if you do a google search which should give you an idea of what was available.   From memory I think it was called the one shot test which isps used most frequently.    Some tests may be what they call intrusive and will disconnect the EU,  but most aren't.

AFAIK the info is mostly about line info and dlm type info.  It checks if the line is OK between the SP and the ras.  And also to the dsl am.   I'm sure they base anticipated speed on attenuation.... in fact most of the info is based on info obtained from the modem.  Ie hlog and qln to detect hr and crosstalk.

--

Sorry really battling to type this on a mobile device as it's wanting to correct a lot of words and abbreviations which it doesn't recognise.

[licquorice]

Hmm . . . I don't know.

I would presume some tests could be performed by just "tapping onto" the pair. My feeling is that for a valid, reproducible, result the "extraneous" parts of the circuit should be disconnected.

Perhaps Black Sheep (or licquorice or . . . ) would please comment?

Sorry I haven't replied sooner, missed the thread for some reason. Not that it would have helped if I had seen it, as I'm afraid I can be of no help, my experience of working in the local network finished in the mid seventies when all tests were carried out from the local exchange by the test clerk. Hopefully BS can shed more light.

[Weaver]

Burakkucat wrote:
> My feeling is that for a valid, reproducible, result the "extraneous" parts of the circuit should be disconnected.

You would think so, wouldn't you? Perhaps using some clever Magic the external unknowns can be compensated for as part of the electronic design of the tests, or as part of post processing the outputs of multiple tests as in eliminating some unknown constants when solving simultaneous equations.

I can just hit "copper line test" on the Andrews and Arnold control server 'clueless', and it just gets on and does its thing.

[tickmike]

Today my isp will do an escalation with BT to try and get more information, they haven't had an update on the REIN team's activities but they are still looking into it.
My isp said they did not do the two cu line tests, he also said he had never seen a graph like that.