Re: Port scanning behind proxies: my experiments with making ultra_scan work with nsock

[Jacek Wielemborek <d33tah () gmail com>]

W dniu 01.07.2015 o 17:54, David Fifield pisze:
> I looked over the patch and it looks like a nice reasonable approach. I
> didn't try it in depth.

Thanks! I'm still trying to track down one annoying segmentation fault
that happens once every few -p- scans with the following command:

while true; do
  valgrind ./command 2>&1 | tee command
  grep -q Invalid && break
done

The problem is related to deleting a probe while a pointer to it is
still in the connectHandler - possibly because I'm trying to handle an
event that isn't really related to the probe (like NSOCK_STATUS_KILL
that I already caught).

> On Tue, Jun 30, 2015 at 08:07:28PM +0200, Jacek Wielemborek wrote:
> > +/* Sets connection lingering options to apply before connect().  It makes a copy
> > + * of the options, so you can free() yours if necessary.  This copy is freed
> > + * when the iod is destroyed */
> > +int nsock_iod_set_linger(nsock_iod nsi, struct linger* lingeropts);
> > +
> > ...
> > +  /* structure used to trigger connection lingering */
> > +  struct linger *lingeropts;
>
> It looks like struct linger is just two ints. Maybe you could just
> statically allocate it and not need to allocate and free memory
> dynamically.

Is it guaranteed to be two ints on all supported platforms?

Attachment: signature.asc
Description: OpenPGP digital signature

_______________________________________________
Sent through the dev mailing list
https://nmap.org/mailman/listinfo/dev
Archived at http://seclists.org/nmap-dev/