On Wednesday 24 December 2003 15:07, Tom Mitchell wrote: > > To get the full path it may be necessary to visit the site by > hand with a browser and drill down a little. Good guesses may > work once the structure is understood. > > > So far the mirrors I have selected are faster... so this is worth > doing. Yes, I did the same. I just used the URL that pointed to the directory that _contained_ the "headers" directory. Once I validated it with a --dry-run it was good to go. Regards, Mike Klinke