>> I need to schedule site download at night time and
>> I need to STOP all downloads at 6:00.

HTTrack has an "scheduler" built in. Well, in fact it's not that, but you can
manage to more or less have one:

TO START a project: you could use --advanced-wait=N, where N is the number of
seconds that the robot waits until it starts downloading. Note it counts from
midnight, local time. That is, let's say it's 7:18 pm and you want your
project to start at 00:30 (half an hour past midnight); you will add
--advanced-wait=1800 (1800 seconds past midnight). I've read it's an option
specially created for the graphical frontend, but don't know how to smartly
use it there. To use this command-line option inside winhttrack I think it has
to be added into filters box.

TO STOP a project: you could use --max-time=N, where N is the number of second
to the end of the process, so --max-time=3600 will stop the project in 1 hour.
This is a command-line, too. I'm not completely sure if the project is killed
or instead of that all queries are stopped and the robot waits until all
queued files are downloaded.

>> How this issue is handled by HTTrack?>> does it fix-up page links and
everything immediately

To my knowledge, HTTrack modifies html links as soon as a file has been
downloaded, so an external link becomes local link.

>> waits until whole content is downloaded?
You could try and see with a small project.

Hope this helps