> WinHTTrack (3.33 and 3.42) not only restarts
> non-html files, it also restarts html files. 

3.33 will check non-html files for changes but doesn't download them if
there's no changes.  Your timeout of 6 is very low. One little glich and it
will abandon the file and try again next time.  Try timeout=60 retry=3

> I've enabled "Do not purge old files" and 
> "Store ALL files in cache". 
> 
> My understanding of WinHTTrack is everytime it needs
> to update/resume a download, it looks at the cache
> (new.zip) and decide which directory (of the remote
> site) to continue donwload. 

Dont need store all. httrack stores the orginal html file in the cache and the
modified file on disk. On continue (not update) it reads the cache to get any
original links not yet DLed

> However when i tried to resume a download, the
> filesize of new.zip didn't continue to grow, instead
> it was renamed to old.zip and a new new.zip is
> created (Is this what its suppose to happen?) And
> therefore download re-starts from the beginning. 

That is what is supposed to happen.  If you cancel an mirror the next time
that Action should be Continue not Update.