Appears to all be working at 07:07 now I'm awake and could see something is adrift
The Apache server process crashed at 00:47:50.789463 (approximately! by which time I was asleep), the first time in nine months it's failed. There is virtually nothing in the Server logs to say why though, the only entry that indicates this is in Apache's own log : httpd.pid overwritten -- Unclean shutdown of previous Apache run? when it was restarted this morning. However, the Windows event log shows a fatal call to Apache2 for an incorrect function so maybe a bug in Apache?
Local processing continued okay except for anything that needed http access of course.
I don't believe there's a direct connection between this and the previous problem mark saw as there's a large gap between them and restarting Apache has got it working again with no other changes. However, the timings of some messages in this thread seem to indicate no access between the two times? There are no 403 errors recorded in the log since restart so far except for the usual hackers from Europe and Asia.
At the moment, uploads are running fine, as are those people running the web pages.
One consequence of this is that as there were no uploads for 6 hours or so, the ES counts and so on for today will be out with no chance of recovery.