- cypherpunks-legacy - lists.cpunks.org

Re: Brute Force DES
by Ryan Russell/SYBASE 17 Dec '03

17 Dec '03

How about harder logistical problem? I had considered the possibility of cracking DES once and for all (I was specifically thinking of crypt(3), but it applies just as well for DES in general..) and instead of trying up a bunch of computers for however many months it took to crack a single key... Let's tie up everyone's extra storage and store the results as each key is generated.. Yes, I realize that it's a rather large amount of storage... Then key lookups could be done at will, reverse DNS style.. Ryan ---------- Previous Message ---------- To: frissell, cypherpunks, trei cc: From: trei @ process.com ("Peter Trei") @ smtp Date: 07/22/96 04:55:17 PM Subject: Re: Brute Force DES > Peter wrote: > >Any one up for a distributed brute force attack on single DES? My > >back-of-the-envelope calculations and guesstimates put this on the > >hairy edge of doability (the critical factor is how many machines can > >be recruited - a non-trivial cash prize would help). Duncan wrote: > I volunteer my 120 MHZ Pentium. A lot more Pentiums are out there now than > a year ago. That makes it more feasible. A lot more people with full net > connections. Like most Americans, I have a flat rate net connection and a > flat rate local phone connection so could run a cracking session permanently > (as long as no one tells my ISP). We need a full test of the Winsock > cracking client in any case. It wasn't working very well last time. > > DCF <back-of-envelope> In my terminology, 'hairy edge of doability" means we have a shot at success, but I wouldn't bet the farm on it. I thought that I might bet a couple hundred bucks, though. Sadly, after further calculation, I'm not so sure if it's doable just yet. What I'm looking at is a known plaintext attack on single ECB DES, using a brute-force test to cycle through the key space. People would get chunks of keyspace to test from a central server or servers, and would be motivated to take part by a cash prize for the lucky person who finds the key. Lets do the numbers: Single DES has the security of 56 bits of key - there are 64 bits in the keys, but 8 of them are parity bits which add nothing to security. 2^56 = 7.205e16 keys (which is a whopping big number) Let's guess that we can recruit the equivalent of full-time on 1000 machines. 7.205e13 keys/machine. Let's guess that we have about a month before people start to lose interest - so we want to be more than 1/2 done by then. Lets say we want to sweep the whole space in 40 days. 1.8e12 keys/machine/day ~21,000,000 keys/machine/second The fastest general purpose, freely available des implementation I'm aware of is libdes. by Eric Young. With this, I can do a set_key in 15.8 us, and an ecb_encrypt in 95 us/block. That adds up to about 9,000 keytests/sec (this is on a 90 MHz P5, running NT). I'm looking at ideas to speed up DES - if I'm willing to use honking great lookup tables, the permutation steps can be done more quickly than libdes. I'm also looking at implementing the algolrithm in hand-optimized P5 assembler. (It's been years since I've done a major assembler project - the P5 has some truely weird features to be considered, but also has (some) internal 64 bit registers to play with). Let's guess that I can speed up a key-test up by a factor of 10. (This is not a slur on Eric's code - it's extremely clever, but not optimized for any particular processor, or for key-testing. Note that the keytest described above takes about 10,000 cycles/test.) That gets my workstation up to about 90,000 keys a second, which is still almost a factor of 250 too slow. I'm going ahead with my work on a faster DES keytester, but unless optimizing gives an astounding win, I now think a distributed bruting effort is a bit pre-mature. What will make this brute doable, if not now, then in the near future? 1. Faster Processors - Moore's Law is still holding. A year ago, my 90 MHz Pentium was one of the faster machines taking part in the 40-bit RC4 crack. Now, it's passe. 2. More processors. The number of people on the internet continues to grow rapidly. 3. More interest - Crypto awareness has greatly increased in the last year, and a real cash prize (say, over $500) will generate both publicity and interest. These factors all multiply together. The number of cycles that could probably be recruited is increasing at a fast rate. A major part of the work will be a keyspace distribution mechanism which can handle the load (this was a major stumbling block last year). </back-of-envelope> Peter Trei trei(a)process.com Disclaimer: This has nothing to do with my employer.

1 0

Lawsuit over publishing how-to-murder book
by Deranged Mutant 17 Dec '03

17 Dec '03

>From LI Newsday today, p. 18: Suit Follows Muder by the Book Publisher of how-to 'Hit Man' manual is blamed in 3 killings The Associated Press (Greebvelt, Md.) - James E. Perry committed muder by the book. Now the book's publisher is accused of aiding and abetting his crimes. In a case that legal scholars say could set a precedent in First Amendment law if allowed to proceed, a federal judge yesterday said he would rule in 30 days on a motion against the publisher of "Hit Man: A Technical Manual for Independent Contractors". [..] The $10, 130-page book has sold 13,000 copies since it was published in 1983 by Paladin Press of Boulder, Colo., a small company that sells mostly through mail orders from its catalog. [..] No mention of the Internet or four-horseman in the article.

1 0

Re: [Noise] was Re: Giving 6 year old kids Uzi's
by Duncan Frissell 17 Dec '03

17 Dec '03

>At 5:14 PM -0700 7/22/96, Ernest Hua wrote: >>Or the way many blacks were lynched (physically and socially) in the South. >>Or the way many asians were segregated. Or the way many ethnic groups >>fought each other in inner cities. >> >>These are cultural relics of the good ol' days I simply can do without. Few armed blacks were lynched. Like--none. In any case, posession of machined metal is in no way comparable to lynching. Massed armed attacks on people simply because they are alleged to have possessed machined pieces of metal the size and shape of fifty-cent pieces (the reason for the BATF attack on the religious community outside of Waco) *is* comparable to lynching. DCF

2 1

Re: NSA Lawyers Believe ITARs Would be Overturned if Tested in Court (fwd)
by Vincent Cate 17 Dec '03

17 Dec '03

My arms trafficker page has been up since April 26th. We have 838 arms traffickers and a public file with a list of 360 "known arms traffickers". It was mentioned on CNN. So far, nobody from the NSA or anywhere else has complained. I am sure if this went to court they would loose. They have to claim that clicking a mouse button on a web form so that 3 lines of text go back to where they just came from, makes someone a criminal. "May it please the court, the charge against these 838 criminal clickers is international arms-trafficking." In a country where a double murderer can walk? I am sure they realize this, which is why I feel it is safe for me to do this (also I believe that EFF etc would so love to have my case as a test case that they would pay my legal bills). -- Vince Cate http://online.offshore.com.ai/arms-trafficker/ http://online.offshore.com.ai/publicity/cnn.html

1 0

Re: Bare fibers
by frantz＠netcom.com 17 Dec '03

17 Dec '03

At 12:30 AM 7/23/96 -0500, Douglas R. Floyd wrote: >> >> * Carbons sent to: In: jimbell(a)pacifier.com >> >> -=> Quoting In:jimbell@pacifier.com to Harka <=- >> >> In> The fiber is usually coated with a very thin layer of clear plastic to >> In> protect against moisture and abrasion, and the diameter is around 0.5 >> In> to 1.0 millimeters in diameter. >> >> >> Doesn't that make it vulnerable (detectable) to Tempest attacks? > >Not really sure how. I have had heard of ways to tap a fibre optic link >noninvasively, but its not related to Van Eck or anything like that. You could break the fiber and add a repeater (if you know enough about the light protocol). Plastic fiber can be cut with a pocket knife, glass requires a machine which will make a square cut and polish the end. Those machines are not yet cheap. ------------------------------------------------------------------------- Bill Frantz | Cave ab homine unius lebri | Periwinkle -- Consulting (408)356-8506 | [Beware the man of one | 16345 Englewood Ave. frantz(a)netcom.com | book] - Anonymous Latin | Los Gatos, CA 95032, USA

1 0

Re: Digital Watermarks for copy protection in recent Billbo
by Alex F 17 Dec '03

17 Dec '03

> Sniffers aren't much effort, and if I sniff your tagged purchases and > put them out over the net anonymously, they are traced to you. Actually I was thinking more along the lines of physically stealing, but if someone sniffed an electronic transfer of a record then the laws would become even more useless as far as enforcement goes. > Probably, but the Billboard article discussed using CC numbers as an > ID in the online watermarked transactions. Doesn't mean they were > correct, of course. Considering that their sources are probably more reliable (even though less knowledgable), and considering the idiocy of such an idea, I would risk saying that they are right :) > AFAIK, most "bootlegging" is of unreleased concerts or out-takes. > Digital watermarks would be of little use. There are solutions to this that work. 1) The Greatful Dead approach - let everyone bootleg live shows. Who cares? 2) The Frank Zappa Approach - take the bootlegged copies, use better equipment, and possibly your own soundboard recordings of the same show, and put them out yourself. Since you are capable of putting out a better product sonically, then beat them at their own game ("Beat the Boots") Alex F =-=-=-=-=-=-=-=-=-=-=-=-=- Alex F alexf(a)iss.net Marketing Specialist Internet Security Systems =-=-=-=-=-=-=-=-=-=-=-=-=-

1 0

Re: Netscape
by shamrock＠netcom.com 17 Dec '03

17 Dec '03

At 16:07 7/23/96, Tom Weinstein wrote: >The Deviant wrote: >> >>> You should only break rules of style if you can | Tom Weinstein >>> coherently explain what you gain by so doing. | tomw(a)netscape.com >> >> Style is standing up for what you beleive in. Netscape obviously has >> none, or they would be activly fighting the ITAR. > >Anyone who believes that Netscape is not actively fighting ITAR is a >fool. Amen. -- Lucky Green <mailto:shamrock@netcom.com> PGP encrypted mail preferred. Defeat the Demopublican Unity Party. Vote no on Clinton/Dole in November. Vote Harry Browne for President.

1 0

Re: random numbers reverse-engineering
by Bill Stewart 17 Dec '03

17 Dec '03

At 08:28 PM 7/15/96 +0300, Juri wrote: >Is there somewhere where I could find more information on finding out RNG >algorithms or reverse-engineering RNG's, once you have some quantity of >random numbers generated by some RNG? > >For example a local bank is giving each customer a list with 600 one-time >passwords (6-digit decimal numbers), and I believe they use the account >number as (one of the) seeds for the RNG. Is there some program that I >could use, together with the numbers and possible seed, to try to break >the RNG? If you have some guess about the algorithm being used, you can try it, and if they've chosen a weak algorithm, you may be successful. For instance, if they use a simple Linear Congruential Multiplicative PRNG, X[n+1] = ( a * X[n] + c ) mod m and if you've got a list of 600 one-time passwords, you can get a good approximation to m by taking the largest password and looking for prime numbers slightly higher than it. You can then try solving for a and c. On the other hand, if the account number is large, and each X[n+1] is the low-order 6 digits of Y'[n+1] = MD5(Y[n]) and Y[0] = account number, then it'll be much harder to reverse-engineer. # Thanks; Bill # Bill Stewart +1-415-442-2215 stewarts(a)ix.netcom.com # http://www.idiom.com/~wcs # Confuse Authority!

1 0

Re: Defend Mail Bomb
by Bill Stewart 17 Dec '03

17 Dec '03

At 11:25 PM 7/14/96 +0800, you wrote: >Is it possible to defend mail bomb? If not, detect who they are? Mail bombs are hard to defend against. If your company or ISP has a firewall or mail server that handles their incoming email, it can potentially be configured to block mail from known harassers, but it's much harder to do if automatically - ten megabytes of binary data might be mailbombs, or might be the CAD/CAM file that your engineering department wanted. If the mailbombs are all coming from one place, over a period of time, you can often look at the headers and track down where they came from, and contact the administrators of the machines the mail came from to ask them to stop the problem. But if the attacker is good, this requires looking at large numbers of log-files, and many administrators aren't willing to do this except for serious on-going problems. If the mailbombs are coming from anonymous remailers, most remailer operators are happy to put you on their block list, to block further anonymous mailbombs. # Thanks; Bill # Bill Stewart +1-415-442-2215 stewarts(a)ix.netcom.com # http://www.idiom.com/~wcs # Confuse Authority!

1 0

Re: Sternlight on C'punks
by shamrock＠netcom.com 17 Dec '03

17 Dec '03

At 14:42 7/16/96, Peter D. Junger wrote: >How does one set up a kill-file for a mailing list? I run a Linux box >with sendmail and use the MH mail system. > >My best guess is that I will have to install procmail, but would like >your advice before going to a lot of labor. Procmail is the way to go. -- Lucky Green <mailto:shamrock@netcom.com> PGP encrypted mail preferred. Defeat the Demopublican Unity Party. Vote no on Clinton/Dole in November. Vote Harry Browne for President.

2 1