> When I view the temp file, this contains the
> contents of the html file from that site at the end
> of the file, containing a german company address. 
HTT's tmp files are temporary binary files with NO HTML, gone once the actual
file has been processed. I don't know what you were looking at.

> After about 18 hours of scanning, the capture is
> done, with about 600 errors, the tmp files are gone
> with no html files and the fd_* (and most other)
> directories are empty.
How can you know about the errors when you can't even find the log file.

> I have tried various settings (ignore robots.txt, do
> not delete files), this has not changed the
> behavior. 
> Limiting the download to max 8 sessions reduced the
> number of errors drastically.
Overloading the site resulting in a denial of service attack countermeasures?
try Connections=4 connections/sec=1

> 1) Is there a setting the prevents that these file
> are deleted after the scan?no and nothing useful there anywhere.

> 2) Is there a way to convert the temp files to the
no and nothing useful there anywhere.

> 3) After scan, I can review the error log. Where is
> this saved locally (to review with a different
> utility) ? Viewing it inside httrack (file is
> extreamly large) is very slow.
top directory of the mirror (<base path>\<mirror director>

Next time post the command line used (log file line two)
Some sites do not like a HTT browser ID (I only use msie6)