Re: Choosing unique passwords - how paranoid is too paranoid?

["Johann MacDonagh" <johann () macdonaghs com>]

I've since moved to a password from passphrase system. I choose an
easy to remember passphrase for each system I want to keep secure, and
then take the initials and punctuation of each word to create a
password. I usually try to choose a passphrase where numbers can be
substituted in as well.

Seems to be the best way so far.

Thanks for all the replies!

Johann

On 6/26/08, Chris LoVerme <chris.loverme () sl-tech net> wrote:
> Great topic... For years I've always struggled with a good password rotation and selection theme. I think everyone 
> has their own methodology about it and there's the issue at heart; someone figures out the method or the routine and 
> cracks away. I've found some good themes and I've tested them against password crackers and dictionaries.  The 
> following is on my blog as well (if reading this here is a pain) and hoping not to break any rules here.  I want to 
> test against linux too, but haven't done so yet, that will make a good follow up post I suppose :)
>
> (http://loverme.net/wordpress/?p=20)
> Make your passwords more difficult to crack:
> These days it's actually quite difficult to pick a decent password that's difficult to crack; the future is going to 
> present us more options around biometrics or algorithmic keys.   More than likely, passwords will never go away but 
> will continue to be combined with a device (card, key, etc) which is called two-factor authentication (something you 
> know such as a password, and then something you have such as a card or fingerprint).  If you've done technical 
> support or network administration, no doubt you've seen all the password combinations from numbers, to letters and 
> numbers, mixes, and so on.   You also know that social engineering often gets someone to give up their password, so 
> it is difficult to guard against that as well.
>
> Crackers are basically hackers that use toolsets designed to defeat either the password or the method that encrypts 
> the password.  There's lots of commercial software available because people forget password, even system 
> administrators, so you'll find a password cracker for just about every system out there.   Most of these applications 
> start with what's called a dictionary attack, which means it has a large text files with just words in them and it 
> tests them.   Some of these dictionaries are giant, in other languages, and focus on specialty niches like movies, 
> Shakespeare, etc.   The other common attack is a brute force attack in which the program will try to logically guess 
> your password in a sequence such as "pasp", "pasq", "pasr", "pass".   Brute force can take a long, long time because 
> your password might be 16 characters long, contain symbols (*&^%$#), capital letters, numbers, and so on.   Brute 
> force attacks will eventually find your password no matter what, plus the time to find your password decreases as 
> crackers take advantage of powerful hardware like distribution farms (lots of computers) with multiple, multi-core 64 
> bit processors.   A bank of 10 servers with dual 64 bit quad core processors could crack the password 
> "g00dn1gHt3very1″ in less than 10 seconds if encrypted with a 40 bit algorithm.
>
> Anyway, let discuss 5 good password techniques that will throw off the dictionary attacks and brute force methods.
>
> 1. Use a pass-phrase - Pass-phrases are much better, basically use a sentence rather than your child's first name.  
> Unfortunately, most programs limit your password length.  If a program doesn't limit password length pick a 
> pass-phrase because it's easy to remember, won't be in a dictionary, and hard to brute force.  An example would be "I 
> love 2 drink sugar free lemonade!".  That's a 35 character password with a capital letter, a symbol, spaces, and a 
> number.   (I tested this password with 128 bit AES encryption and a cracking tool was unable to crack the password 
> after working on it for 12+ hours on a single dual core 3.2Ghz processor.
>
> Example test (I love 2 drink sugar free lemonade!):
> Works on Windows Vista
> Works with WinRAR 3.70
> Works with WinZIP 11.1
> Works with Excel 2007
> Does not work with Word 2007 (character limitation)
> Passed Basic Passware Audit*
>
> *Password Audit Notes: Passware 8.0 has a 27 character limit on brute force attacks for the Office Recovery tool.  
> The RAR & Zip Recovery tool has a limit of 12 characters.  In all my tests, I used the defaults. This is one of the 
> top password recovery and audit tools on the commercial market and retails for $495.   If you're an IT specialist, I 
> suggest a copy.  (No, I'm not getting paid for referring them)
>
> 2. Math - These are easy to remember, won't be in a dictionary, and hard to brute force.  Use a word, symbol, and a 
> number.  Here are some examples of passwords: 12*Twelve=144, Ten*10=100!, Eighteen-1=17.  Combine this tip with the 
> previous one for a super strong password: "Ted said 2*2=4″.  When I spoke with Microsoft consultants a couple of 
> years ago, they fell in love with this method.
>
> Example test (Ted said 2*2=4):
> Works with Windows Vista
> Works with Word 2007
> Works with Excel 2007
> Works with WinRAR 3.70
> Works with WinZIP 11.1
> Passed Basic Passware Audit
>
> 3.  Extended ASCII (Grpahics) - Some password crackers don't have options for Extended ASCII, in fact, it's rarely 
> used anymore within the Windows world due to fonts and graphics.  They aren't preloaded into cracking tools, they 
> aren't well known, and they're not in dictionaries.  Someday this might change, but until then, a passphrase like 
> "451°F will burn paper" is a platinum-class passphrase.  Easy to remember, is 21 characters, has extended ascii, 
> numbers, and a capital letter.
>
> The whole extended ASCII set is 127 through 255, 255 is fantastic because it looks like a blank space but it isn't! 
> Imagine a password that's intertwined with 255's and spaces.  Even if a password cracker cracks the password and is 
> able to display it, it's going to show as a bunch of blank spaces looking as if it failed to crack it correctly.   
> It's even better if someone is using a sniffer and not looking at the hex codes.  Some web browsers won't be able to 
> display this but here's what it looks like: "     " That's two 255's a space, then two more 255's.
>
> All the extended ASCII sets make good passwords (for programs that support them) or add one symbol to your current 
> password.  Here's a another slightly artistic example: "░▒▓▒░"  A poorly written cracking program may not be able to 
> display these characters and may crash or display other symbols in an interpreted font set.  For example, ▒ may show 
> up as "_" in another font, but the underlying value is ASCII 177 not "_".  Remeber to add words to make it a phrase 
> and make it even stronger.
>
> Example test (451°F will burn paper):
> Works with Windows Vista
> Works with Word 2007
> Works with Excel 2007 (not Mac compatible)
> Works with WinRAR 3.70
> Works with WinZIP 11.1 (not DOS compatible)
> Passed Basic Passware Audit
>
> 4. Common Set (capital letter, longer than 6 characters, and number or a symbol) - Common set passwords involve a 
> pattern and are great to use because they're not in a dictionary, brute force will take some time, but these are not 
> always easy to remember.  Examples are: Tropicana9, Battlestar3, !Starbucks!, Goldfrapp$, etc.
>
> Bold example:
> Works with Windows Vista
> Works with Word 2007
> Works with Excel 2007
> Works with WinRAR 3.70
> Works with WinZIP 11.1
> Passed Basic Passware Audit
>
> 5. When possible save your documents in a higher encryption such as 128bit AES or RC4 RSA encryption.  128bit AES is 
> set by default in Office 2007.
>
> People are now using passwords that they think are secure but really aren't anymore because password crackers have 
> picked up on these ones. Password methods to now avoid:
>
> 1.  Leet (or Hax0r) - This was clever, but the brute force crackers picked up on it quickly and coded for common 
> substitutes.  There's even a dictionary now for it, so avoid these passwords.   Examples of these now well used 
> passwords are: l33t, r0xx0rz, n00b, etc.
>
> 2.  Foreign Language - Language dictionary sets are common now, so forget the Russian password you came up with.   A 
> brute force is going to crack this very fast anyway.
>
> 3.  Qwerty - Keyboard patterns all well known now and in dictionaries as well as some brute force options.  Examples 
> include: qwerty, asdfg, poiuy, and zxcvb.
>
> Not sure if you password is secure?  Try cracking it yourself (elcomsoft, lostpassword.com) provide tools for 
> dictionary and brute force password recovery.
>
>
> -Chris
> P.S.
> There's a good interesting article here too: http://www.codinghorror.com/blog/archives/000949.html
>
> -----Original Message-----
> From: listbounce () securityfocus com [mailto:listbounce () securityfocus com] On Behalf Of Kurt Buff
> Sent: Wednesday, June 25, 2008 6:08 PM
> To: Johann MacDonagh
> Cc: security-basics () securityfocus com
> Subject: Re: Choosing unique passwords - how paranoid is too paranoid?
>
> On Tue, Jun 24, 2008 at 8:27 PM, Johann MacDonagh <johann () macdonaghs com> wrote:
> > Hi all,
> >
> > I've recently began a full on password change process where I'm
> > increasing the security of passwords I use for various systems I use
> > (computer systems, websites, etc...). In the past I've only used a few
> > different passwords and hoped for the best. I'd like to start working
> > on a new system that allows me to create easy to remember passwords
> > for each unique system. I don't want to create completely random ones
> > and rely on a password manager, because I use these systems at home,
> > at work, and on my iPhone. They need to be something I can easily type.
> >
> > So my first scheme involved coming up with a rather long base
> > password, choosing a 4 character acronym for each system, mixing it up
> > in a certain way, and inputting those jumbled characters in predefined
> > locations. This solved one issue:
> > 1. If someone where to compromise one password, it's unlikely they
> > would be able to deduce the same pattern for other systems.
> >
> > Then, I got paranoid. What if they had two passwords? The differences
> > could be found, and analyzing the 24 different permutations (4!) of
> > the differences could quickly find a pattern.
> >
> > So, I modified it a little. I took the name of each system, padded and
> > mixed in yet *another* master password (this time much shorter), and
> > ran it through this (on OS X):
> >
> > echo -n mypaddedstring | openssl dgst -md5 -binary | openssl enc
> > -base64
> >
> > and took the first few characters. I put that in a certain location of
> > my master password. The reason to use a hash function is pretty
> > obvious, and
> > base64 allows me to add in additional bits to brute force with the
> > same number of keys.
> >
> > This has worked out better. I've started using mnemonics to remember
> > each system's unique part. Muscle memory!
> >
> > Now, I'm up against a wall. I can't possibly remember a different
> > password for *each* system. So I came up with the (final) idea of
> > classifying systems as high or low, depending on the problems a
> > compromise would create. For example, my registration on some random
> > forum is low, whereas my PGP passphrase is high.
> >
> > I know this is looking like there will never be a question, but there is.
> > What does everyone think of this system? Would you classify sites that
> > hold somewhat private information (such as Amazon.com without any
> > saved payment
> > methods) as high or low? Is there another way?
> >
> > Let me close by saying that the day I can use a smarcard for 3 factor
> > authentication (PIN, physical access to card, and biometrics) to
> > access ALL systems (hey, web developers, you can ask for x.509 certs
> > you know!) is the day that I stop worrying about all this. Or should I
> > be worried about that too? :)
> >
> > Thanks!
>
> Way too complex
>
> Get a USB key or PDA, and put your favorite password manager on it. I like PasswordSafe, but Keepass is derived from 
> it and many like it, though I haven't tried it. I'm sure there are commercial alternatives as well, but these are OSS 
> - and available on sourceforge.
>
> I like my passwords to be pass sentences. 20+ characters, with all of the punctuation and other non-alpha characters 
> they deserve. Much easier to remember and to type. Think up a sentence, commit it to your password management 
> application, have it ready. One site, one pass sentence. Very easy.
>
> Even if the site/server/application doesn't take passwords that long, having it in your password management database 
> is better than trying to remember it, and your password management software will generate passwords if your brain is 
> non-functional for whatever reason.
>
> Back up your password management database, too.
>
> Kurt