> 'Extract URLs from the logfile.' 
> That's exactly what I don't know how do quickly. I 
> know how to do it one url after the other (copy and 
> paste in a new text file). But as there are more 
than 
> 500 '.rm' files (each one bigger than 10 Mo), it 
will 
> take me a lot of time. ;-( 

Ah.. on Un*x, one line similar to:
cat hts-log.txt | grep "File too big" | grep -
E "http:\/\/.*\.rm" | sed -e 's/.* \(http:\/
\/.*\.rm\) .*/\1/g'

But I don't know of any similar tools on Windows - 
except if you install standard gnu utils on DOS