I found where the problems was (I guess). I selected "not robot.txt rules" in
the "Spider" tab. This way all the pages and resources have been downloaded
and all links are now relative to the mirror. 

Now I have a new different problem. Some documents (doc, pdf) whose URLS are
available via javascript have been downloaded to the mirror but cannot be
accessed by clicking on the link (it says that the document cannot be found).
Any idea on how to solve the problem?
Thanks for your help