Working my way through nmap, one function at a time.

[James Rogers <jamesmrogers () gmail com>]

Added gdb stack traces to the strace results I posted here the other day.

http://mystry-geek.blogspot.com/2012/05/analyzing-nmap-using-strace.html

Going to do this several more times with different command line
arguments to work out exactly what the program is telling the OS to do
when different options change the behavior of the program, and then
checking the stack against gdb at those points in the program.

Also planning on doing gdb stack traces of where the config files are
being loaded in, because this is not clear in looking at the code.

--

With a better understanding of what is happening and where it is being
called I began going through the program one function at a time,
understanding what each does in turn.  I am going to accumulate all my
thoughts about the program here:

http://mystry-geek.blogspot.com/2012/05/analysis-of-nmap-program-flow.html

Going to work on this for the entire summer (as I figure out a new
part of the program).  Try to get a complete view of the entire
program before I am done.

--

A question I have about ultra_scan() in scan_engine.cc  is that it
appears that it does one scan at a time, and then waits for the
response. Is that right?

Or does it send out all the requests and then waits for all the responses?

Does that behavior depend on the type of scan?

Is scripting done through ultra_scan() as well?

-- 

I have some suggestions after my initial look at the program.  I'll
bring them up in my weekly meeting tomorrow.

Thanks,
James
_______________________________________________
Sent through the nmap-dev mailing list
http://cgi.insecure.org/mailman/listinfo/nmap-dev
Archived at http://seclists.org/nmap-dev/