Ethernet network not phone network.

In reply to a post by Realalemadrid:

No wrong again, the problem is not the master socket, but upstream of the modem in structured cabling to the router.

If its structured cabling and the structured cabling tests OK then have they tested the patch leads.

Yes I've tested the patch leads.

Have you got any other leads that are known to be broken, that you can test against the tester?

In reply to a post by joconnell:

Yes I've tested the patch leads.

I can't understand why it wouldn't work then, structured cabling test OK, patch leads test OK, what else could be wrong?

Edited by PCJM40 (Mon 04-Sep-23 17:19:50)

Where are you located?

In reply to a post by ionic:

Something like this: https://www.flukenetworks.com/enterprise-network/net...

Though often certain managed network switches and network cards have some cable diagnostic functionality which may be a good indicator

LinkIQ are OK but they are fairly limited in the actual physical cable tests that they can carry out. They are more of a general purpose tester that give a helicopter indication whether a cable passes or fails (Ethernet speed tests) but not the underlying physical cable properties.

A much better test instrument from Fluke Networks is a DSX-5000 or DSX-8000 CableAnalyzer. This its what a decent commercial structured cabling installer will use to certify the cabling is up to standard, or indeed meets the EiA/TIA Category or ISO Class for the installed cabling.

Here is sample test page. In order for permanent cabling which has been installed to meet Cat5e (or Class D) or Cat6a (Class E) the cabling must meet certain parameters including:

- Wiremap
- Insertion Loss
- Return Loss
- NearEnd CrossTalk (NEXT)
- FarEnd Cross Talk (FEXT)
- Attenuation to Cross Talk Ratio (ACR)
- PowerSum ACR (PS-ACR)

The link could fail on any one of those aspects and it would result in an overall test fail for the installed cable. A DSX will not only tell you why it failed but where. Typically its the termination at the ends, but there could be a n issue along the length of the cable - sometimes these don't immediately manifest.

Hove you tested from the plug you plug into your router all the way to the incoming line point through all ports.

you could also check for straight through testing to see if the pins all go to the correct pins at the other end and there is no crossover or short.

In reply to a post by joconnell:

Yes I've tested the patch leads.

Your original post is contradictory. You are saying you've narrowed it down to a particular cable but then you're saying that cable has checked out as OK.

Unless it has degraded in some sort of unusual way (not enough for the basic cable to tester to pick up on it) then it is unlikely that is actually the cause.

Even if the (more expensive) cable tester does find an obscure issue, what do you hope to do about it?

Have you tried re-crimping both ends as a matter of course? This takes seconds to do and is of no cost to you.

I would re-trace all your diagnostic steps carefully and consider again what may be the issue. Consider buying a long reel of cable and bypass that structured cabling on a temporary basis. This will be much quicker and more conclusive than buying cable testers.

My guess is a simple diagnostic has been used when the tester showed all as green. Replace each part until it works - in this case presumably they used an alternate route to that cable which works whereas when using that cable it doesn't. Assuming all else is the same then it would suggest something wrong with that cable even if the tester says all is fine.