RE: UNIX password auditing tool and the search for dictionaries too

["Tim Heagarty" <tim () heagarty com>]

Please pardon me if this has already been covered in this thread, I didn't
see the earlier posts on the subject.

First the PIN algorithm is widely known and there really is no need for much
of a PIN cracker program. Like DES the algorithm is published but the keys
are kept secret. If an institution uses a simplistic key, which some do,
then it is trivial to derive the natural PIN. However most systems don't use
the natural PIN but create an offset that is mod 10 added to the natural PIN
to create the number that you remember and don't write down anywhere.

Bruteforce is handled in one of two ways on almost all ATM systems. On a
"track 3 write" system the PIN retry count is decremented and written onto
the card so it counts down and when it hits 0 the bank has the option to
perform a card retain or just give it back to you with instructions to go
into the bank and take care of the problem.

The second method simply records the retry count at the host and the same
retain/return decision is made when the retry count is exceeded. Either way
there is no bruteforce available beyond three tries. Now if you have a card
writer and can keep resetting the retry count or bump it up to 99 to start
with you have a slightly greater advantage. You have to properly calculate
the LRC on the card too without fudging up the data and having the modified
card retained on the first insertion.

Thank you,

Tim Heagarty CISSP, MCSE
Tim at TheaSecure dot com
http://www.TheaSecure.com/
"There are only 10 kinds of people in the world, those that understand
binary, and those that don't."

> -----Original Message-----
> From: Tomas Wolf [mailto:tomas () skip cz] 
> Sent: Saturday, August 09, 2003 4:04 AM
> To: Michael Martinez
> Cc: security-basics () securityfocus com
> Subject: RE: UNIX password auditing tool and the search for 
> dictionaries too
>
>
> I would like to note a little about the security of 4 digit 
> pin... I believe that author wanted to point out that thanks 
> to the fact, that you can't try the 4 digit number more than 
> three times at a time - which makes it a pretty strong 
> system, not that 4 digit is a strong password. 
> Of course who has the time, can go from ATM to ATM and try 
> two passwords at the time to bruteforce it, but that is 
> almost impossible to achieve (since anybody responsible who 
> looses any type of financial card usualy reports it the same 
> day). We have four digits with possible ten variants = 10 on 
> the fourth power = 10.000 possibilities that is (divided by 
> two tries per card insertion) 5.000 maximum tries, which 
> gives us 2.500 average tries to get the right pin (approx. 
> 1.250 card insertion of two tries)... And let's get the 
> theory little further, let say that each insertion takes 15 
> seconds, that is 1.250x15 =  18.750 second = 312.5 minutes = 
> 5.208~ hrs. of actuall interaction with ATM... Well maybe for 
> a student :-) I believe that security is always a trade off. 
> To have top noch security one has to count with a lot of 
> expenses with training people to understand and use the craft 
> of security. Not many end users a willing to authenticate 
> more than one time, they need to work and not to worry about 
> IT stuff, that is why we are here, or am I wrong? But there 
> is always more, isn't there? :-)
>
> Good luck to you all...
> Tomas
>
>
> > > Before you go too far with strong passwords, remember, they do more
> > harm
> > > than good in most cases. You trust your money to a four 
> digit pin so 
> > > think about strong authentication, not strong passwords. 
> Two factor 
> > > can be done with a variety of inexpensive technologies.
> >
> > Are you kidding me, you are under the impression that a 4 
> digit pin is 
> > secure?  I for one have no illusions about how insecure a 4 
> digit pin 
> > actually is!  Whatever security is provided by said 4 digit pin is 
> > more related to that fact that there are not freely available pin 
> > cracking tools for ATM machines...as there are password cracking 
> > tools.
> >
> > > Strong passwords are the number one source of denial of service in 
> > > most environments due to the frequent false reject problem that 
> > > occurs when users can't keep up with frequent changes and strong 
> > > password. They're also one of the highest costs for security since 
> > > it's the number one task for help desks and sys admins to support.
> >
> > As a help desk supervisor, I assure you that the related 
> cost of time 
> > and money supporting the reset of passwords is minimal and 
> therefore a 
> > small price to pay for increased security.
> >
> > ...
> >
> > > In terms of dictionaries, I think the aggressive approach would 
> > > include concatenations and number and special character injections 
> > > into the words. In more secure environments, were users 
> are battered 
> > > with
> > monthly
> > > password changes they usually inject the numeric value for 
> the month 
> > > somewhere in a common word. But the point is, it's not too 
> difficult 
> > > to build a really big database of words with special character and 
> > > numeric injections, run them through the hash algorithm and have a 
> > > table to check for matches.
> >
> > If someone were in an environment where they must change their 
> > password monthly...they are probably using the wrong technology.  
> > Perhaps a combination of different layers would be a better 
> solution 
> > to monthly changes.
> >
> > ...
> >
> > -----Original Message-----
> > From: Shane Lahey [mailto:s.lahey () roadrunner nf net]
> > Sent: Monday, August 04, 2003 7:38 PM
> > To: james.easterling () ed gov; security-basics () securityfocus com
> > Subject: RE: UNIX password auditing tool
> >
> > Alec Muffett Crack :: http://www.crypticide.org/users/alecm/
> >
> > > -----Original Message-----
> > > From: james.easterling () ed gov [mailto:james.easterling () ed gov]
> > > Sent: Monday, August 04, 2003 4:39 PM
> > > To: security-basics () securityfocus com
> > > Subject: UNIX password auditing tool
> > >
> > >
> > >
> > > I have tried searches for UNIX password cracking tools and I have 
> > > come
> > up
> > > with little value.  Can someone direct me to passwd 
> auditing tools 
> > > besides "John The Ripper" that are free or cost?
> > >
> > > Regards,
> > > James
> ----------------------------------------------------------------------
> > --
> > --
> > > -
> ----------------------------------------------------------------------
> > --
> > --
> > > --
> ----------------------------------------------------------------------
> > --
> > ---
> --------------------------------------------------------------
> ----------
> > ----
> ----------------------------------------------------------------------
> > --
> > ---
> --------------------------------------------------------------
> ----------
> > ----
> ----------------------------------------------------------------------
> > -----
> --------------------------------------------------------------
> --------------
> >
>
>
> --------------------------------------------------------------
> -------------
> --------------------------------------------------------------
> --------------


---------------------------------------------------------------------------
----------------------------------------------------------------------------