Re: weakksa.c and wi_dat_len

>>>>> "mp" == Mathieu Perrin <[email protected]> writes:
>>>>> "h" == h1kari  <[email protected]> writes:
>>>>> "bj" == Brett Jones <[email protected]> writes:

     h> how many have you tried to crack?

Well, that would be ZERO, since I'm a NetBSD user.

     h> If you can provide some specific examples of how the code is
     h> nonfunctional, or any facts to back up your very elaborately
     h> illustrated argument for my software being a "nonfunctional
     h> example", please be my guest.

Thank you, gracious host.  I was thinking mostly of this post, to
which I never saw a response.  

    mp> Our problem was that we couldn't find the wepkey, even when we
    mp> had 7 or 8 millions packets ( using FMS attack ).

    mp> While trying to find why we always have too few weak packets,
    mp> we found that dwepcrack used only 3000 or 4000 packets ( from
    mp> a dumpfile of 400 000 packets ), and found only 200 weaks IVS.

    mp> So we investigated into weakksa.c, and found an average value
    mp> of wi_h-> wi_dat_len of about 50000, while you reject all
    mp> packets with wi_dat_len greater than 2304 ( MAX_PKT_SIZE ).

    mp> We are very surprised by this value of 50000.

A few weeks ago I quoted and summarized this post in a reply to
someone else's question, and no one addressed it the second time,
either.

     h> Time and time again, I see people post questions regarding not
     h> being able to crack wep and it's either related to:

     h> 1. not following the instructions

If ordinary, competent sysadmins like Mathieu can't get it to work,
and if their well-written appeals for help go ignored, then IMHO it is
broken.  Maybe the bug is in the documentation, or maybe it's in the
code, but it still doesn't fucking work, see?

    bj> when your done with the coding, I have some dishes and laundry
    bj> that need seeing to.......

oh, step off it, Brett.  I had a valid point and you know it.
Including support for *BSDs that you don't personally like or use is
kind of a big deal in the BSD community, and people get pissed all the
time when maintainers short-change their variant.  Is it a little
quirky?  Yes.  Does it challenge your ``I don't owe you jack---free
software means you fix it yourself'' dogma?  Yes.  But it's the
community's standard.  The Qlogic ISP driver and ipfilter are the
examples I've followed most.

In this case, I totally hear h1kari that he lacks the time, hardware
resources, and interest to do the traditional set of three ports.  And
I anticipated this understandable limitation in my post by mentioning
that h1kari contributed to solving the problem by starting once again
to actually distribute the current source code of his open-source
program.

But to imply I'm a lazy cur for reminding everyone of the deficiency
is totally out of line.

-- 
Le fascisme est la dictature ouverte de la bourgeoisie.
		-- Georg Dimitrov