Re: New to Nmap

[David Fifield <david () bamsoftware com>]

On Wed, Jan 12, 2011 at 02:19:17AM +0530, ambarisha b wrote:
> Hi,
> I know I am a bit late(out of station).I have been studying Nmap
> Network Discovery like you guys said.Will finish it in 2 days.While
> referring the code ,I saw this in the todo folder(david.txt).
>
> Nsock:
> Add a buffer to each iod, so that you can ask for a certain number of
> bytes
>   or lines and get exactly that many, no more. Venkat wrote a proposal
> at
>   http://seclists.org/nmap-dev/2009/q3/0600.html.
>
> I read the proposal, but I couldn't figure out the reason to add a
> buffer to iod to serve this specific purpose.The facility is already
> built into networking sockets.But may be do_actual_read() is not
> availing it.In the recvfrom() (line:637) call instead of passing
> sizeof(buf) as an argument, if max_chunk is passed, then it should
> do.Anyway max_chunk is being set to the requested no.of bytes if the
> call is readbytes().Would this work or am I missing something?
> If this has specifically been assigned to David , I am sorry of nosing in.

Ambarisha, this is mostly finished but it is in a private branch. You
can check it out with

svn --username guest --password "" svn://svn.insecure.org/nmap-exp/david/nmap-readbuf

You can see the logs of what is changed with "svn log --stop-on-copy".
Most changes will be in the nsock directory.

Since December 6, I have had a message almost ready to send regarding
what the API should be. I didn't come to a conclusion and didn't have
enough time to think about it. But since you have expressed an interest
I'll finish it with the current status.

There is more to this job than just sending a smaller buffer size to the
operating system. For example, we also wish to have a function to return
exactly one line, and keep extra bytes for later reading. This is an
extra buffer on top of the OS socket buffer. Please see my message at
http://seclists.org/nmap-dev/2010/q4/385, which also summarizes the
state of the branch.

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