Just a bump to let anyone interested know what happened.
I ended up taking the easy option - a weekend in Canberra (300 km away) seeing old friends and re-installing the backup server (HP ML-350, SME and AFFA).
There were too many obstacles to solving the problem remotely - compounded by having no local staff willing to work on a Linux machine following my phone intructions.
Since I broke the system, it was up to me to fix it.
That was easily done and everything works well now.
It is a testament to the usefullnes of AFFA.
The server I broke was the production server. I had run an AFFA backup before I started work, so the AFFA server was "up-to-date".
When I deleted the home directory (Idiots can be so powerful, can't we?), I managed to halt the production server. I performed a "RISE" on the AFFA server to be the new production server, which took about 20 minutes with 200 Gb data. The users were off the air for about half-an-hour, but at 22:00 at night, no-one was at work.
So, by 22:30 or so, the system was running, all the data was there and the next day everyone could do their work.
I could not remotely restart the old production server, which would have been a silly thing to do, anyway, since that would have caused an IP clash (two systems with the same IP). This is where I would have needed local hands on the keyboard and that was not going to happen.
Anyway, it was good to see old friends and Canberra is a lovely place to visit.
Thank you, Jean-Phillippe for your help. I am sure what you suggested would work but the difficulties meant a re-install was simpler.
Cheers
Ian