Re: Scans failing on Win32 - failures differ depending on args len(maybe)

[jah <jah () zadkiel plus com>]

On 11/10/2011 02:39, David Fifield wrote:
> On Fri, Oct 07, 2011 at 03:26:07PM +0100, jah wrote:
> > On 07/10/2011 14:52, Gisle Vanem wrote:
> > > > DEV  (SHORT) IP/MASK                     TYPE        UP   MTU  MAC
> > > > eth0 (eth0)  192.168.1.15/24             ethernet    up   1500 00:19:21:3D:3F:CE
> > > > eth0 (eth0)  fe80::219:21ff:fe3d:3fce/64 ethernet    up   1500 00:19:21:3D:3F:CE
> > > You can try to add "-e eth0 --privileged" to the command-line.
> > > See the 'o.isr00t' stuff in mswin32/winfix.cc. Or disable the
> > > __HrLoadAllImportsForDll() sillyness therein.
> > These results differ based on the position of
> > "-e eth0 --privileged" in the args list:
> >
> > nmap -F -v -v -d -n 195.166.128.231
> > MS-DOS - FAIL - route_dst_generic: Failed to obtain system routes:
> >                 getsysroutes_dnet: sysroutes_dnet_find_interfaces()
> >                 failed
> >
> > nmap -F -v -v -d -n 195.166.128.231 -e eth0 --privileged
> > MS-DOS - FAIL - pcap_open_live(eth0, 256, 0, 2) FAILED.
> >                 Reported error: Error opening adapter:
> >                 The system cannot find the device specified. (20).
> >                 Will wait 5 seconds then retry.
> >                 ... Call to pcap_open_live() failed three times...
> >
> > nmap -e eth0 --privileged -F -v -v -d -n 195.166.128.231
> > MS-DOS - SUCCESS
> >
> > I bet there's a really simple explanation and I'm just not seeing the
> > wood for the trees.
> That's a good clue. This may be a long shot, but please try this patch.
> It defers calling win_init until option processing is finished,
> including the options that have to wait to be processed until after the
> whole command line has been seen.
>
> The delayed option processing is a new thing added by Shinnok this
> summer, which would explain why we haven't seen this error before.
>
> David Fifield

Thanks David,

After applying your patch to the original source (not modified by the patch
I submitted), all of the failing scans now complete successfully.

I'm still looking into why the addition of a single '-v' to the
arguments '-F -v -d -n 195.166.128.231' made a difference to the
number of times _refresh_tables was called before a call to
getAdapterAddresses returned ERROR_INVALID_PARAMETER.

Do you have any thoughts on what might be happening?

jah


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