Re: 5.51 - NSE assertion - timeout_msecs

[David Fifield <david () bamsoftware com>]

On Tue, Apr 05, 2011 at 03:30:10PM -0400, Patrick Donnelly wrote:
> On Tue, Apr 5, 2011 at 3:12 PM, David Fifield <david () bamsoftware com> wrote:
> > On Mon, Apr 04, 2011 at 04:04:15PM +0200, Vlatko Kosturjak wrote:
> > > Hello!
> > >
> > > It seems I stumbled upon another NSE assertion similar to negative
> > > socket timeout which is fixed in 5.51. Last Nmap lines:
> > >
> > > NSE Timing: About 84.64% done; ETC: 14:50 (0:01:02 remaining)
> > > nmap: nsock_event.c:375: msevent_new: Assertion `timeout_msecs >= 0' failed.
> > > Aborted
> > > #
> > >
> > > Will try to repeat it with debug flags. But from my experience with the socket
> > > timeout, probably I will not hit it again...
> >
> > I checked but it's not obvious to me how it may happen. There's a
> > negative check in nmap.set_timeout, and I don't think there's another
> > way to change the timeout that gets passed to msevent_new. If you manage
> > to get a stack trace I think we can solve it.
>
> Well the code says:
>
>   nu->timeout = luaL_checkint(L, 2);
>   if ((int) nu->timeout < -1)
>     return luaL_error(L, "Negative timeout: %d", nu->timeout);
>
> Should that be "<= -1" or "< 0"?

No, -1 means no timeout. It's a special case in msevent_new.

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