Re: Missing getaddrinfo on Windows 2000--almost solved

["Kris Katterjohn" <katterjohn () gmail com>]

Hey guys,

On Wed, Aug 27, 2008 at 10:32 PM, David Fifield <david () bamsoftware com> wrote:
> > > So, does _WIN32_WINNT *have* to be set to _WIN32_WINNT_WIN2K, or can
> > > it be something slightly newer, like maybe _WIN32_WINNT_WIN2KSP1?
> > >
> > > Of course I could hack a little in6addr_any for Ncat, but if Nmap or
> > > anything else using Nbase wants any of these really common IPv6
> > > definitions, this will be a problem.
> >
> > We compile Nmap using Visual C++ 2008, and the respective runtime components
> > specify Windows 2000 SP4 as a minimum requirement*, so you can probably get
> > away with WIN2KSP1 (although people may still be able to compile our source
> > themselves using another compiler). I'd also be surprised if many people are
> > still using WIN2K RTM anywhere, but I'm sure we'll manage to annoy someone
> > somewhere if we leave them out ;)
>
> Setting it to SP4 sounds fine. The important thing is that it be set to
> *something* otherwise it defaults to Vista or Longhorn.

*sigh*

I was looking at the "Using the Windows Headers" MSDN page you posted
which lists the NTDDI macros and values (including NTDDI_WIN2KSP*).
The thing is I was in the mindset of _WIN32_WINNT so I must've just
glazed over the NTDDI prefix when looking at this and thought I was
looking at something different.  There are only 5 _WIN32_WINNT values,
corresponding to major versions (2K, XP, Longhorn, etc) and without
SPs.

Since I had already modified nbase_winunix.h on my box to use
_WIN32_WINNT_WINXP so I could work on Ncat before looking around for
any permanent solution to this, I just left it there and continued on
thinking that it would work fine.

So.. since it just has to be something before Vista, XP works fine for
what I need and I guess will work for this other stuff too.  But I
don't want to do anything without an OK on this.

> Sorry for the trouble.

No worries, as I seem to be causing as much trouble as I'm trying to solve.

> David Fifield

Thanks,
Kris Katterjohn

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