Erm, yes that URL was a little confusing. Now back on track.

The interesting think is that you are coming in from the loopback address whereas I get a real IP address at that point. I wonder whether you have some funny web server internal firewalling problem.

If you have access to the server itself (I assume it is linux) can you log in and run
`wget http://aaaaa.bbb?feed=make-pdf-newspaper`
and fire up an editor on the file it creates and see if it contains articles or is just some html template stuff. Also check the tail of the log file again and check the wget is coming in via

We can do this exchange by email for a while and post back the outome. I’m webmaster at bruntsfield dot org.