HappyNomad wrote:

Nine hours? One imagines that whatever was on the end of a bad connection would show itself as having 'lost communication with' or some such once the car had been plugged in?

Cars are certainly very electrically and electronically complex but the tools to see into that complexity are available or is that just naivety on my part?

I don't know about our cars but I can give you an example from my own experience. This is for the Volvo commercial chassis I used to work on, we used a Volvo diagnostic tool known as VCADS. when adjusting any engine parameters if any other ECU's on the vehicles Can system had been replaced and not properly programmed to the system (i.e. just fitted with pre loaded software installed) when it came to adjustments of parameters on other systems it was a case of "The computer said no"

The way round this was to look for any new ECU's fitted and simply unplug the Can connection, resulting in the system not seeing it and allowing the diagnostics full access to the rest of the system. So It makes sense to me that a faulty connection may not be picked up by a diagnostics tool.

9 hours does sound a lot but I can well understand how difficult it may well have been to track down the fault ?

One I remember quite well at work took about a week to trace and was initially diagnosed as an engine ecu failure but was still the same after replacement. It turned out to be corrupted software .

Not easy to get to the bottom of these types of faults sometimes  2007 TD4 XS The work horse that earns its up keep
2013 SD4 HSE The posh one for towing the caravan
1973 Triumph Stag I bought in 2009 and restored.

I am just happy they idd find the problem, i would never found it  'per aspera ad astra'

The danger is that garages without the correct diagnostic tools and/or training (plus experience) will try to find fault by swapping out components and units. This will result in extra cost, or as I have experienced once, swapping units with other customers cars. (not JLR)

Not everything is digital, the FL2 has lots of sensors and analogue signals floating around. And worse of all if there are bad connections some of these signals will be floating making the outcome random.

I've not even mentioned the software which may become corrupt, or even contain unintentional errors right from the start.

As a regional support engineer for a major computer manufacturer I have sometimes spent days, working round the clock, solving and finding obscure faults. (Hardware and Software)
In hindsite the faults are always easy to see why they occured, they just are difficult to find.

Perhaps, in the future, we will just lease, or hire, a (working) car, from a company, and let them worry about how long it will last (servicable) on the road. FL2 XS SD4 Auto 2010 2012-2017 (21k - 91k miles) (MY2011)
FL2 Metropolis SD4 Auto 2014 2017- (16k - 77k+ miles) (MY2015)
Metro in its 11th Year of (Extended) LR Warranty / Full LR Service History
(Expensive, but Trouble/Worry free - hopefully?)

Do you know which connection it was? The one in the footwell is usually problematic. 
Hung like Einstein, smart as a horse.

I'm sorry, i have no idea witch connection is was. Somewhere near the PCU.
Alltough they spend a lot of hours, had to pay 'only' €560.-  'per aspera ad astra'