Nmap Development mailing list archives

Re: [PATCH] Match line update for Famatech Radmin remote control


From: Brandon Enright <bmenrigh () ucsd edu>
Date: Wed, 10 Jun 2009 16:42:42 +0000

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Wed, 10 Jun 2009 06:36:58 -0500 or thereabouts Tom Sellers
<nmap () fadedcode net> wrote:

Brandon Enright wrote:
Hey Tom, this patch looks really good but before I check it in, I
have a question about the softmatch.

Is "m|^\x01\x00\x00\x00\x25|" broken?  Your change to
"m|^\x01\x00\x00\x00.{3}\x02\x12\x08\x02|" is a lot more specific.
Is there a case where the original would match but your change
won't?


Thanks for catching the mistake Brandon!

How embarrassing to attempt to fix a match line only to botch it.
The original softmatch line would fail on Radmin 3.x, the new one
would not, had it been correct.

Don't worry about it, all of us who work on nmap-service-probes make
a lot of them. Something about trying to construct fast, generic regular
expressions from limited data seems to encourage mistakes ;-)


A better softmatch would be:

softmatch radmin m|^\x01\x00\x00\x00.{5}\x08.\x00| p/Famatech Radmin/
o/Windows/

I started counting bytes and trying to decide if this was a good
match.  It seems to work it would need to be ".{4}", not 5.


If you think that the line is too generic I can alter it to include
options for the two values I have seen at position 8 (x01 and x02), 9
(x10 and x12) and 11 (x01 and x02).  My concern is that a new minor
release would come out and change those fields, breaking the
softmatch.

I wouldn't say it's too generic, it just doesn't capture the structure
to what is changing very well.  I'd propose changing it to:

softmatch radmin m|^\x01\x00\x00\x00\x25.\x00..\x08.\x00..| p/Famatech Radmin/ o/Windows/

Also, I noticed your new match lines use ASCII where possible even
though the old matches do things like \x79 instead of 'y'.  I converted
your match lines over to escaped bytes because I think it makes the
field numbers a lot more obvious between different matches.

Attached is a new patch with my proposed changes.  If you think it will
work well I'll go ahead and check it in.

Regards,

Brandon

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.11 (GNU/Linux)

iEYEARECAAYFAkov4pMACgkQqaGPzAsl94LHSACgiR3DK/8155VPvpTKZBljGUO2
hmkAn3wIPfiz+3eie1fWMWbjH8nlew+C
=9zcR
-----END PGP SIGNATURE-----

Attachment: radmin.diff
Description:


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

Current thread: