Beta
×

Welcome to the Slashdot Beta site -- learn more here. Use the link in the footer or click here to return to the Classic version of Slashdot.

Thank you!

Before you choose to head back to the Classic look of the site, we'd appreciate it if you share your thoughts on the Beta; your feedback is what drives our ongoing development.

Beta is different and we value you taking the time to try it out. Please take a look at the changes we've made in Beta and  learn more about it. Thanks for reading, and for making the site better!

A Competition To Replace SHA-1

kdawson posted more than 7 years ago | from the securing-government-bits dept.

Encryption 159

SHA who? writes "In light of recent attacks on SHA-1, NIST is preparing for a competition to augment and revise the current Secure Hash Standard. The public competition will be run much like the development process for the Advance Encryption Standard, and is expected to take 3 years. As a first step, NIST is publishing draft minimum acceptability requirements, submission requirements, and evaluation criteria for candidate algorithms, and requests public comment by April 27, 2007. NIST has ordered Federal agencies to stop using SHA-1 and instead to use the SHA-2 family of hash functions."

cancel ×

159 comments

Sorry! There are no comments related to the filter you selected.

Generic hashing is impractical (0)

LiquidCoooled (634315) | more than 7 years ago | (#17736548)

When you digest a message and obtain a hash it is obvious that there will be collisions.
Unless the hash key length is equal to the data you are hashing there will be problems.

Whenever you are throwing data away you must decide which is important, do you remove the grand overall detail of the data or the fine grain details?

As an equivilent, your ID card will hold a hash of you.

If I show you some pictures of people can you tell which one is me? Would you let me on a plane with just a grainy picture?

Maybe secure hashing needs to store a mixture of the low level and the high level details but in a context specific way - the face picture example should also store the detailed iris pattern as well as an overall face picture, both should match to allow this person through. It might be easy to find someone who looks like me, but the specific portion cannot be modified without surgery.

A hash of a zip file may contain the overall hash plus a specific portion of the zip root structure (its virtual FAT table), something like a word doc would need its document information, an executable would need a breakdown of its segments, other formats would require other extensions.

You keep the details specific to the format instead of trying to generalise everything (unsupported formats would of course just use the general algorithm.

Re:Generic hashing is impractical (1)

petermgreen (876956) | more than 7 years ago | (#17736690)

When you digest a message and obtain a hash it is obvious that there will be collisions.
It is obvious that there will be many possible inputs that produce the same output.

however the actual chance of encountering two inputs that hash to the same value by accident is vanishingly small.

with SHA1 even finding two inputs that hash to the same value deliberately is very hard and finding a second input to match an existing output is considered virtually impossible.

If I show you some pictures of people can you tell which one is me? Would you let me on a plane with just a grainy picture?
that is a very different situation because two photos of you will be far from identical. Secure hash functions are only usefull in the case where things are supposed to be identical (two copies of the same file for example).

Re:Generic hashing is impractical (0)

LiquidCoooled (634315) | more than 7 years ago | (#17736756)

If I take a HTML document and produce a hash for it, I can very easily modify that file and then reproduce the same hash by simply extending that document with an offscreen DIV with white text until I find a match.
Sure, it might take a while, but it would be possible.

If I store information specific to that type of document (node count, word count or something) then the job becomes MUCH more involved.

Re:Generic hashing is impractical (3, Insightful)

simm1701 (835424) | more than 7 years ago | (#17736836)

No you can't very easily modify it - thats the point.

You can exhaustively search for a collision, but the time requirement is very much non trivial.

Feel free to prove me wrong - unless you have a huge botnet or a supercomputer available I dont give you much chance of finding a collision that way for md5 let alone SHA-1

Re:Generic hashing is impractical (0, Troll)

LiquidCoooled (634315) | more than 7 years ago | (#17736930)

The whole point of the fucking article is that some researcher found a method that was 2000 times less work to find a collision in SHA-1, hence making it feasible to do.
If that had not been done then I would agree with you and we wouldn't even be having this discussion.

Recent years have seen a stream of ever-more-refined attacks on MD5 and SHA-1--including, notably, Wang's team's results on SHA-1, which permit finding collisions in SHA-1 about 2,000 times more quickly than brute-force guessing. Wang's technique makes attacking SHA-1 efficient enough to be feasible.

I was simply considering an alternative method which was content specific hence making it impractical to extend a document to insert extra data and get a match.

Re:Generic hashing is impractical (3, Informative)

simm1701 (835424) | more than 7 years ago | (#17737014)

2000 times quicker than brute force (where brute force is average time 2^159 attempts) means the algorithm is not as secure as it used to be thought.

This has demonstrated a cryptographic weakness, there could quite well be more, look at the research over the years on weakening md5, therefore moving to different algorithm would be advisable.

Its doesn't mean that you are going to be able to find a collision in non trivial time, but it did lower the bar. Lowering it enough that people wanting high grade protection should switch to a more secure algorithm.

Context specific data has no place in a hash, it would only weaken it.

Re:Generic hashing is impractical (3, Informative)

RAMMS+EIN (578166) | more than 7 years ago | (#17736692)

``Maybe secure hashing needs to store a mixture of the low level and the high level details but in a context specific way - the face picture example should also store the detailed iris pattern as well as an overall face picture, both should match to allow this person through. It might be easy to find someone who looks like me, but the specific portion cannot be modified without surgery.''

The idea is that, in a good hash function, each input bit affects all the output bits more or less equally. This is especially true of cryptographic hashes, and for a good reason. The stronger the correlations between input and output, the weaker the hash function.

Re:Generic hashing is impractical (1)

lordtagoh (1003233) | more than 7 years ago | (#17736698)

Collision always exist,

what u don't want is an easy way to locate them!

Know that there is a collision is a normal thing,
be able to fabricate Hundred of them in a couple of minute
(It's NOT possible now with SHA1, but it's becoming just a bit easier..)

So better start now to replace it so if even more serius
problems are founded we will be in a good shiny new boat!
(It will probably also be sunked but we buy time in this way)

Re:Generic hashing is impractical (5, Informative)

delt0r (999393) | more than 7 years ago | (#17736716)

You clearly don't know what a crytographic hash is about. And this is not what is ment by collisions resitant. What it means is that there is minum amount of work needed to produce a collision.

There are a number of different type of collisions as well. Lets assume we have a 256-bit hash. There is the kind of colision where you just find *any* 2 strings that produce the same hash, which should require on avarage 2**128 "operations". A harder task is given a string and its hash find another string with the same hash. For a secure hash 256-bit hash function this will require on avarage 2**256 "operations".

There are other properties that are important as well. Its a well established idea. Hashes are very very usefull and are used for a lot more that file verification and we know what properties they need. We are just not very good at producing very good hashes yet.

Re:Generic hashing is impractical (2, Informative)

simm1701 (835424) | more than 7 years ago | (#17736808)

I think that you are missing the point of a hash.

A hash is a signature of the file, its designed to give a good confidence that a given file that you have been supplied matches the one that you think has been supplied.

The theory being that being able to create a file that is of the same length as the orignal, is not corrupt (eg a zip file still unzips, an executable still runs, a pdf still displays) and is different from the original but still hash should be infeasable (not impossible, most cryptography doesn't look for impossible, not practical within a given time frame is sufficient for most needs)

Another use of hashes is on data storage systems, especailly with backup systems, where two files with the same hash and length are treated as the same file (so no need to write it to tape twice) this way you only have to sort the list of hashs and look for matches, rathering than having to diff every file against every other one.

Personally I think I'd rather binary diff matches hashes just to be safe - but thats time intensive. The chances of two files each having the same size and SHA-256 hash and being different is less than the chance of your sotrage device being destoryed (meteroite, fire, flood, plane) before you are able to back up either file

Re:Generic hashing is impractical (2, Informative)

PDAllen (709106) | more than 7 years ago | (#17737326)

Sure there must be collisions, but that's not the point.

The point is that you can verify that data is correct with a good amount of confidence, from a relatively small hash code. So I can download a lot of data through, say, bittorrent, and despite the fact that I don't necessarily trust the people I actually download from, I can verify that the hash is right and therefore I am confident that the data I receive is what the original seeder put out: no-one's decided to play games and (say) sneak their CC number grabber into the data.

So what you want is an algorithm which is reasonably easy to run, which SHA-1 is, but where it is not easy to find a collision. For example, if my hash code was simply to give the total byte sum modulo 1000, then while it would almost certainly catch accidental errors in data, it would be very easy for an attacker to stick in his CC number grabber to your data then fiddle the byte sum back to where it should be.

Your idea pretty clearly shows you have no idea of what hashes are used for: there is no point preserving the data structure, it takes a lot of extra space and gives virtually no security. For example, SHA-1 produces a 20 byte hash. I can put something that size up on my personal website without getting huge bandwidth charges even if millions of people want to download it - and then I can distribute my 1GB zipfile by way of people I don't necessarily trust (but who have more bandwidth than I) and still the eventual recipients can be confident that what they receive is what I sent out. If I include the virtual FAT table of this zipfile, my hash size goes up by about 500,000 percent (literally), and so do my bandwidth charges. And I get virtually no extra security, because all that an attacker has to do above finding an SHA-1 collision is ensure that the change doesn't affect the FAT table: i.e. he replaces some suitable virtual file of mine with one of his, keeps the name and size the same and he's done.

Draft location (5, Informative)

ErGalvao (843384) | more than 7 years ago | (#17736580)

The draft can be found (in PDF) here [nist.gov] .

How long before we get (3, Funny)

Anonymous Coward | more than 7 years ago | (#17736588)

...the magical SHA-24M?

Interesting.... (0, Funny)

Anonymous Coward | more than 7 years ago | (#17736606)

NIST is preparing for a competition to augment and revise the current Secure Hash Standard.

I guess society is getting more liberal about drug use. I mean, a competition to secure hash? Cool!

Now, we just need a competition for securing pot, coke, meth, etc....

Leadtime for security: Is it too late? (2, Insightful)

G4from128k (686170) | more than 7 years ago | (#17736622)

The security of a given hash/encryption would seem to be a function of how much effort has gone into breaking it. Lots of algorithms can look good on paper, but until people really tear into the math and code, it's true level of unbreakability is undecidable. A 3 year competition is not likely to bring enough IQ, theorems, malevolence, or brute CPU cycles to bear against any candidate.

The point is that any attempt to quickly create a new algorithm is likely to create an insecure one. Shouldn't we be trying to create candidate algorithms for the year 2050 to give the algorithms time to withstand attack? Or do we plan to keep creating new algorithms as a serial security-by-obscurity strategy.

Re:Leadtime for security: Is it too late? (2, Interesting)

bhima (46039) | more than 7 years ago | (#17736738)

The general consensus among the experts in cryptology is that a competition is far more effective than other methods of designing algorithms. Presumably the 3 years is a function of how long the world can wait as compared to how the experts need to crack it. The thing that makes me wonder is why they waited so long to begin it.

Characterizing this process as a "serial security-by-obscurity strategy" is completely wrong because due to the very nature of the competition the algorithm is known from the start.

Re:Leadtime for security: Is it too late? (5, Insightful)

suv4x4 (956391) | more than 7 years ago | (#17736766)

Shouldn't we be trying to create candidate algorithms for the year 2050 to give the algorithms time to withstand attack? Or do we plan to keep creating new algorithms as a serial security-by-obscurity strategy.

This is what a hash is by design: obscurity. For mathematical reasons alone, you can't have a unique hash for your megabyte message crammed in (say) 256 bytes. Or 512, or 1024 bytes.

And with a public algorithm spec, it's all about whether there's a determined group to turn it inside-out and make it easy to crack.

That said, the ability to hack SHA/MD5 given the time and tools, doesn't make hashes useless. A hash by itself can be useless, but coupled with a good procedure that incorporates it, it can raise the security level just enough so it's not reachable by 99.99999...% of the potential hackers out there that will try to break you.

Security is just an endless race on both sides, and will always be.

Re:Leadtime for security: Is it too late? (1)

Sublmnl (868393) | more than 7 years ago | (#17736952)

A professor once told me..."Knowledge is an infinite resource; we tap it when needed." Much the same as, "Necessity is the mother of invention."

It's difficult enough to figure out an encryption model that will be secure for the next few years, let alone to work out an encryption model for 50 years from now.

Immediacy and viability will be plenty to create encryption that is, "good enough for now." If you desire stronger encryption you might fair better asking hackers to break the current code.

Re:Leadtime for security: Is it too late? (1)

evilviper (135110) | more than 7 years ago | (#17737308)

This is what a hash is by design: obscurity.

Not unless "obscurity" has been entirely redefined, recently.

And with a public algorithm spec, it's all about whether there's a determined group to turn it inside-out and make it easy to crack.

A (mathematically) good algorithm can stand up to such scrutiny. a "determined group" wouldn't make it any weaker. They can only (potentially) expose weaknesses in the algorithm, that allow it to be circumented faster than brute-force alone.

Security is just an endless race on both sides, and will always be.

No, it isn't.

Re:Leadtime for security: Is it too late? (2, Informative)

Beryllium Sphere(tm) (193358) | more than 7 years ago | (#17737528)

>This is what a hash is by design: obscurity.

"Security through obscurity" means trying to depend on indefensible secrets. The classic example from 19th century crypto theory is that it's stupid to try to keep your crypto algorithm secret, so you should keep keys secret instead.

Security through obscurity leads to worldwide breaks when it fails.

The existing secure hashes have nothing obscure about them. The algorithms are published and open for review. The fact that they're vulnerable to brute force is not being hidden and is the same problem that all the workhorse encryption algorithms have.

"Security through obscurity" would be trying to hide the fact that there's a work factor reduction attack and hoping that nobody rediscovered it.

Re:Leadtime for security: Is it too late? (0)

Anonymous Coward | more than 7 years ago | (#17737656)

For mathematical reasons alone, you can't have a unique hash for your megabyte message crammed in (say) 256 bytes.

Yes you can. The point is that there will not be 2^(8*256) different megabyte messages in the lifetime of the universe. Probability (of collision) is truly "beyond astronomical".

And with a public algorithm spec, it's all about whether there's a determined group to turn it inside-out and make it easy to crack.

Wrong. MD5, SHA-1, etc are designed withstand attacks by determined groups. The fact that MD5 and SHA-1 failed does not mean all hash functions must fail. Breaking e.g. Whirlpool would almost certainly mean breaking AES. Not impossible, but very unlikely.

not reachable by 99.99999...% of the potential hackers out there that will try to break you.

Unfortunately there are more than 10'000'000 potential hackers trying to break me. And one succesfull is too many.

Not obscurity (1)

sacrilicious (316896) | more than 7 years ago | (#17738586)

This is what a hash is by design: obscurity. For mathematical reasons alone, you can't have a unique hash for your megabyte message crammed in (say) 256 bytes.

Your point about the impossibility of producing unique M-byte hashes for every N-byte message (where N>M) is of course mathematically correct. But instead of thinking of hashes as working via obscurity, think of the function of the ideal hash to be: the impossibility of finding data with a matching hash without so radically changing the input data that the change is obvious to anyone who sees it. For example, if someone can produce a page of text that has the same hash value as garbage, or as a video of a monkey, the value of the hash function is not diminished. Whereas if someone can produce two license agreements that differ only in the phrases "I accept the following terms" and "I do *not* accept the following terms", the hash function is considered broken.

A hash function seeks to re-map a mathematical space in such a way that previously "adjacent" places in the input space are far apart in the output space. An ideal hash function would do this

  • (a) uniformly: for all points in the space; and
  • (b) unpredictably: H(A)-H(B) != H(A+X)-H(B+X); (so for example, by this criteria a simple checksum is not a good hash)

Re:Leadtime for security: Is it too late? (1)

mrmeval (662166) | more than 7 years ago | (#17738838)

sha1sum sha224sum sha256sum sha384sum sha512sum

I have those on my system.

Re:Leadtime for security: Is it too late? (1)

cperciva (102828) | more than 7 years ago | (#17736804)

The point is that any attempt to quickly create a new algorithm is likely to create an insecure one. Shouldn't we be trying to create candidate algorithms for the year 2050...

Competitions like this and the AES competition aren't about inventing new cipher designs; they're about taking the state of the art and creating a standard. The ideas underlying Rijndael are essentially the same as those in Square, which was published back in 1997; while nearly all of the ciphers submitted to the AES competition were "new", the ideas behind them had been studied long before the competition started.

Of course, no competition will ever be able to anticipate future developments: If people had been paying attention to non-constant table lookup timing leaks (as Bernstein and Osvik/Shamir/Tromer demonstrate) it is unlikely that Rijndael would have won the competition. But unless we want to wait until cryptography is a dead field, with no new research being performed, this will occur whenever there is a new standard, and regardless of how long the process is which constructs that standard.

Re:Leadtime for security: Is it too late? (2, Informative)

duffbeer703 (177751) | more than 7 years ago | (#17736866)

Its not like everyone is starting from a blank slate on the first day of the contest. It's basically a call for the math geeks who design this stuff to polish up whatever they are working on.

Re:Leadtime for security: Is it too late? (3, Informative)

Kjella (173770) | more than 7 years ago | (#17737416)

Let's start with the facts: SHA1 is cryptographically "broken" in the sense there's a "better than brute force" attack which takes about 2^63 operations instead of 2^80 of finding a colliding pair of two random strings.

It's not a practical attack because 2^63 is still a huge number.
It's not a "find a collision to a known string" attack which would be stage 2.
It's not a "find a collision to a known string by appending to a fixed string" attack which would be stage 3.
It is a sratch in the armor which creates doubt if there are more powerful attacks, nothing more.

There are strong alternatives like SHA-512 and Whirlpool (AES-based) which it is possible to use today, if you're paranoid more is better. Is it urgent? Not really, even a practical stage 1 and 2 attack would just be "stuff breaks, files corrupt, migrate away". The only one with really nasty consequences is stage three with code injection attacks in software and such.

Re:Leadtime for security: Is it too late? (1)

Plutonite (999141) | more than 7 years ago | (#17738850)

If that's correct, I wouldnt consider it much of a threat at all. It is scientifically correct for NIST to seek a hash that, in principle, is random enough to be brute-force-equal in terms of no. of attempts required, but practically it will make little difference.

There should be no password entry scheme in the world, in whatever application, that allows this number of attempts without delaying the attacker for years using various methods. But why did you say that

"find a collision to a known string" is different than the problem here? Aren't we talking about the input strings? If so, will it not take approx 2^63 attempts to find a string producing the same hash whatever the input?

Re:Leadtime for security: Is it too late? (1)

Fahrenheit 450 (765492) | more than 7 years ago | (#17739194)

"find a collision to a known string" is different than the problem here? Aren't we talking about the input strings?

No. There are a number of different properties that a cryptographic hash function should have. The first is what is referred to as collision resistance. That is, given a hash function H, the probability of finding any two strings x and y (x <> y) such that H(x) = H(y) should not be significantly greater than that of an ideal hash function (i.e. 2^-(|H(x)|/2)). The second (what the the GP was referring to) is one-wayness or non-invertability. With that the problem is, given H and input x, find an input y (x <> y) such that H(x) = H(y). The probability of this happening are not to be significantly greater than 2^-|H(x)|.

There are others, of course that have come to light of late, multi-collision resistance (i.e. finding k collisions should take k times as long as finding one collision -- most hash functions today only require log(k) times the amount of work), preservation of random oracle indistinguishability, length extension attack resistance, and so on. But the first two listed have always been of paramount importance.

Re:Leadtime for security: Is it too late? (1)

PDAllen (709106) | more than 7 years ago | (#17737472)

Hash insecurity is nothing like as much of an issue as cipher insecurity.

If I use a hash algorithm now, and I am confident that it's secure enough that no-one's going to find a useful collision within a month, then I can happily distribute my data for a couple of weeks, then I maybe need to find another hash algorithm, and the eventual recipients of my data can check their data against the hash I produce and be confident they got the right stuff; the possibility of an attacker coming along a few weeks later with some malware with an identical hash isn't an issue, because I'm no longer using the old hash as a certificate.

Whereas if I want to send secrets around, then I probably don't want anyone to be able to read them for a long time, so my cipher has to be much more secure against being broken.

Re:Leadtime for security: Is it too late? (3, Interesting)

Fahrenheit 450 (765492) | more than 7 years ago | (#17737992)

That's great. Except for one thing...
Hashes are used all over the place in cryptography. That digital signature you generated? You didn't sign the message, you signed a hash of the message. That key you just exchanged? There was likely a hash involved in that process. Hashes are one of the basic building blocks of cryptographic protocols and systems, and while the recent weaknesses aren't too much to worry about yet as they aren't really practical or directly applicable, their presence is troubling.

And far more interesting (to me at least) are the attacks like Joux's multicollisions and Kelsey and Kohno's Hash Herding/Nostradamus attacks.

Re:Leadtime for security: Is it too late? (1)

bahstud1 (1055206) | more than 7 years ago | (#17737668)

The thing is, this has been in motion since august 2005, when the proof of concept was announced @ the annual crypto con in Santa Barbara, Ca. Since then, NIST has hosted 2 HASH workshops and has entertained ideas from all kinds of experts.

There is nothing new going on....

Re:Leadtime for security: Is it too late? (1)

certain death (947081) | more than 7 years ago | (#17738938)

It could be that they will enlist the help of people who have been developing a new hash for years, and now will allow them to bring it to light. I personally have one I have been working on for about 3 years, and given another 3 years, do you think that will be long enough to prove/disprove it?

ROT-7 (1, Funny)

Chapter80 (926879) | more than 7 years ago | (#17736640)

Rot-7. Everyone's doing ROT-13. I'm going to suggest Rot-7.

Think about it. You walk into a video store and you see Rot-13 and right next to it you see Rot-7 --which one you gonna spring for?

Not 13. Seven. Seven Little monkeys sitting on a fence...

Re:ROT-7 (0)

Anonymous Coward | more than 7 years ago | (#17736748)

When did they release Rise of the Triad 7, is it better than 13?

Re:ROT-7 (0)

Anonymous Coward | more than 7 years ago | (#17736758)

But what if someone comes up with Rot-6?

Re:ROT-7 (1)

somersault (912633) | more than 7 years ago | (#17737734)

What about a scheme based on ternary operators - ROT-6.5? 4 times and you're back to where you started, but anyone who is expecting ROT-13 will give up after the first try!

Double ROT13 (0, Redundant)

MountainMan101 (714389) | more than 7 years ago | (#17736646)

Subject says it all. Use Double ROT13 and prosecute anyone who breaks it.

Schneier Proposed this in 2005 (5, Informative)

RAMMS+EIN (578166) | more than 7 years ago | (#17736656)

Schneier proposed such a competition in March 2005: http://www.schneier.com/crypto-gram-0503.html#1 [schneier.com]

Re:Schneier Proposed this in 2005 (-1, Offtopic)

trifish (826353) | more than 7 years ago | (#17736880)

I proposed it several months earlier. How's that +5 Interesting?

Re:Schneier Proposed this in 2005 (0, Offtopic)

Kadin2048 (468275) | more than 7 years ago | (#17737680)

You're evidently just less interesting than Bruce Schneier.

Don't feel bad; same goes for most of us.

Re:Schneier Proposed this in 2005 (-1, Troll)

trifish (826353) | more than 7 years ago | (#17738892)

The topic was "Schneier proposed it". I replyied to the topic by saying I proposed it too, but months before him. What's suddenly so off-topic about it? Meta-mods will take care of you, don't worry, "moderator".

Re:Schneier Proposed this in 2005 (1, Interesting)

Anonymous Coward | more than 7 years ago | (#17737058)

Yeah, and it probably takes a year or two for a project like this to be fully thought-out, organized, blessed by the appropriate governmental approval agencies, &c, &c, &c. It's entirely possible that the origins of this contest were with Schneier's suggestion.

Re:Schneier Proposed this in 2005 (1, Informative)

Anonymous Coward | more than 7 years ago | (#17737594)

Yeah, but other people actually did something about it. The first workshop was in 2005.

Re:Schneier Proposed this in 2005 (2, Informative)

Fahrenheit 450 (765492) | more than 7 years ago | (#17738156)

Yeah. 80% of the crypto world called for one too, they're just not as loud.

The thing is these kinds of contests take money and time to get running and (at least initially) NIST didn't have the resources to get a competition going. So what they did is organize a hash workshop for Halloween 2005, and had a second one last August following the Crypto conference where initial planning for the contest took place (a work shop that Schneier didn't bother to attend -- I guess he had yet another book to sell).

Good News (3, Interesting)

Ckwop (707653) | more than 7 years ago | (#17736662)

The amount of research done in to hash functions is nothing like the amount that goes in to ciphers. I'm not really sure why this is the case because hashes are much more important than ciphers. Hashes are used in MACs to protect the integrity and authenticity of a message.

Ask yourself this, is it more important that somebody can read your SSH connection or that somebody can hijack the channel? The reasons for wanting a good hash function suddenly become very clear.

It's true that hashes are becoming less important as a result of AEAD modes. But they have uses far beyond MACs and it's good to see a competition from NIST to stoke research in to those primitives.

Simon.

Re:Good News (1)

steelfood (895457) | more than 7 years ago | (#17738224)

Hashes are more important than ciphers. But hashes can only be secured so far. Beyond that, the return is minimal. All hash algorithms will eventually be cracked. It's the nature of hashing that the signature is not necessarily unique. Otherwise, it'd be called compression rather than hashing. The goal is to find an algorithm that will produce unique results under the most common conditions, and be least likely to produce the same result for two messages with purely algorithmic differences.

On the other hand, a good cipher can potentially be technologically unbreakable. So everyone is trying to find this holy grail (especially because once quantum computing comes along, today's strongest ciphers will amount to nothing) because it is theoretically possible, while for hash algorithms, this holy grail by definition doesn't exist, so researchers will only put enough time and effort to get to the next level of security when the previous level is threatened.

Re:Good News (1)

Paul Crowley (837) | more than 7 years ago | (#17739388)

This doesn't make any sense to me at all. Studied as theoretical objects, hash functions are "keyed" just as ciphers are. Just like ciphers, they are strong only against an attacker with limited computing power, not against an infinitely powerful adversary. Just like ciphers, we have no way of proving them secure, but through crypatanalysis we gain confidence in them.

And quantum computing has nothing to do with it; there are algorithms that are believed to be strong even in the face of quantum computing, and for symmetric primitives like block ciphers and hash functions, quantum computing is not a big problem.

Hash functions in common protocols (3, Interesting)

Srin Tuar (147269) | more than 7 years ago | (#17736702)


Does anyone know whether or not common protocols and formats such as TLS, ssh, X.509 certs, etc are being updated to use newer hash functions?

Its easy to change parts of a self-contained system, such as password hashes, but common protocols require interoperability and standards compliance.

This is actually fairly interesting situation, where NIST certification and platform interoperability may actually be at odds with each other.
   

Re:Hash functions in common protocols (3, Informative)

cpuh0g (839926) | more than 7 years ago | (#17736746)

Most modern protocols and standards are designed to be agile. Basically, this means that they don't mandate any one particular algorithm, but rather are designed such that alternatives can be used. Otherwise, many specs would be woefully out-of-date every few years as computing power and cryptographic algorithms advance. The 3 examples you give above are all considered "agile", read the specs and note that they use algorithm identifiers and allow for a wide variety of different algorithms to be used, none of the above are strictly bound to use SHA-1 or MD5.

Re:Hash functions in common protocols (1)

Srin Tuar (147269) | more than 7 years ago | (#17736870)


That doesnt seem to be the case.

Looking at the RFC for TLS:

http://www.ietf.org/rfc/rfc2246.txt [ietf.org]
It seems sha-1 and md5 are the only options for hashes in 1.0.

Not to mention that the vast majority of existing implemtations would not be interoperable, even if it is technically possible to update the protocol to support newer hash algorithms. (there are asn.1 id's allocated, but the fixed sized buffers for the output of various hash functions may be different, etc, so protocol changes seem mandatory)

Re:Hash functions in common protocols (1)

Marillion (33728) | more than 7 years ago | (#17737284)

While I agree that TLS and SSL and the like are flexible, the real barrier is not the specification but how long it take for a critical mass of adoption to make a revised specification useful.

Re:Hash functions in common protocols (1)

Kjella (173770) | more than 7 years ago | (#17736948)

Typically, they send some sort of algorithm list, choose the best algorithm they both have, and then use a hash to make sure the algorithm list was transferred successfully (so you can't downgrade security by doing a man-in-the-middle on the algorithm list). So basicly replacing SHA1 starts the day one "better than SHA1" client connects to a "better than SHA1" server, without any backward compatibility issues.

Re:Hash functions in common protocols (1)

Bert64 (520050) | more than 7 years ago | (#17737270)

But if you consider the length of time between AES encryption being available for SSL/TLS use, and microsoft actually supporting it (i believe they still don't) it's going to be years before these new hashing algorithms appear in microsoft products.

Re:Hash functions in common protocols (1)

durdur (252098) | more than 7 years ago | (#17739458)

My understanding is that, due to the way TLS/SSL works, the weaknesses in SHA-1 do not really affect TLS transport-layer security. Hash-based digital signatures are used to validate certificates, though, so the possibility of forging a cert indirectly weakens TLS.

How about SHA-512? (3, Interesting)

ngunton (460215) | more than 7 years ago | (#17736770)

Anybody know if SHA-512 is mathematically vulnerable to the same kind of attack as SHA-1 (only presumably requiring more computing power)? Or is it really a different kind of beast?

Re:How about SHA-512? (1)

delt0r (999393) | more than 7 years ago | (#17737044)

I beleive SHA-256/512 are fine. Also IIRC SHA-256 is a truncuted SHA-512 so there is little point using SHA-256 unless you have tight mesage size constraints. Also remember that SHA-1 is "broken" is a abstract type of way. In terms of real security its not really "broken" in the wild. Its just a little weaker than we hoped.

Re:How about SHA-512? (2, Informative)

tomstdenis (446163) | more than 7 years ago | (#17737092)

No, SHA-224 is truncated SHA-256 and SHA-384 is a truncated SHA-512.

SHA-256 and SHA-512 are different hash functions (same basic design though). On 32-bit boxes SHA-256 is faster, and on 64-bit boxes SHA-512 is faster.

There is no point in 224 or 384, but they're there just for completeness (e.g. to comply with some specs that don't allow the arbitrary truncatage of a hash).

Tom

Re:How about SHA-512? (1)

delt0r (999393) | more than 7 years ago | (#17737112)

My bad. Thanks for the correction.

Re:How about SHA-512? (1)

tomstdenis (446163) | more than 7 years ago | (#17737168)

No worries. Common mixup if you're not waist deep in it all day (most customers don't quite know what SHA-512 is or why they can't pair it up with AES-512 hehehehe).

Tom

Multiple Hash Functions (1, Informative)

RAMMS+EIN (578166) | more than 7 years ago | (#17736794)

I always wonder about what would happen if we used multiple hash functions together. E.g. you provide an SHA-1 hash, an MD5 hash, and an RMD-160 hash, all for the same message. Would that be harder to fool (i.e. make the system think you got the original, but it's actually a forgery) than one hash function that generated as many bits? What about weaknesses in the individual hash functions; would you be worse off because a flaw in any one of your hash functions affects you, or better off, because you have more hash functions that need to be fooled?

By the way, IIRC, OpenBSD and NetBSD include multiple hashes per archive in their ports trees, but use only one for verification.

Re:Multiple Hash Functions (4, Informative)

rbarreira (836272) | more than 7 years ago | (#17736984)

Doesn't work very well. Read this:

http://www.mail-archive.com/cryptography@metzdowd. com/msg02611.html [mail-archive.com]

Re:Multiple Hash Functions (3, Informative)

RAMMS+EIN (578166) | more than 7 years ago | (#17737286)

Thanks. The post you linked to precisely answers both my questions. I'll restate the questions and copy the answers from the post for /.ers' convenience.

1) Would multiple hash functions be harder to fool (i.e. make the system think you got the original, but it's actually a forgery) than one hash function that generated as many bits?

No. In fact, the multiple hash functions perform worse:

``Joux then extended this argument to point out that attempts to increase
the security of hash functions by concatenating the outputs of two
independent functions don't actually increase their theoretical security.
For example, defining H(x) = SHA1(x) || RIPEMD160(x) still gives you only
about 160 bits of strength, not 320 as you might have hoped. The reason
is because you can find a 2^80 multicollision in SHA1 using only 80*2^80
work at most, by the previous paragraph. And among all of these 2^80
values you have a good chance that two of them will collide in RIPEMD160.
So that is the total work to find a collision in the construction.''

2) Does using multiple hash functions protect you against the case where one of them gets broken?

Basically, yes. Just note that your total security is no better than the security of the best hash function (as explained in point 1).

Re:Multiple Hash Functions (1, Informative)

bfields (66644) | more than 7 years ago | (#17737376)

You could take that as a warning against feeding the output of hash functions to each other in series; the OP however was asking about calculating hashes in parallel, and concatenating the output of the different hash functions. Seems to me that that's trivially at least as strong as the strongest of the individual components, but whether it's likely to be worse or better than a single hash of comparable output size sounds like a crapshoot.

Re:Multiple Hash Functions (1)

rbarreira (836272) | more than 7 years ago | (#17737906)

Read the link properly, that's not what it says or talks about.

Re:Multiple Hash Functions (1)

finkployd (12902) | more than 7 years ago | (#17737150)

In a nutshell, stacking hash functions makes it worse.

Finkployd

Re:Multiple Hash Functions (2, Informative)

grumbel (592662) | more than 7 years ago | (#17737458)

I don't think he wants to stack them, but instead simply concat them:

md5sum foo -> 4f1cbee4972934c3beccc902f18242a7
sha1sum foo -> 3c92a387f898a31d2e8af31caff27c0f8f7a5a3a
md5sha1sum foo -> 4f1cbee4972934c3beccc902f18242a73c92a387f898a31d2e 8af31caff27c0f8f7a5a3a

That should definitely not weaken anything, it will require some more CPU and storage, but thats it.

Re:Multiple Hash Functions (1)

RAMMS+EIN (578166) | more than 7 years ago | (#17739018)

``That should definitely not weaken anything, it will require some more CPU and storage, but thats it.''

First of all, you're looking at quite a bit more CPU. At least with the hash functions I've implemented, as well as with MD5 and SHA-1, the time needed to compute the hash depends on the amount of data processed, not really on the size of the hash (some hash functions actually run quicker if you compute larger hashes). If you want to use three different hash functions, rather than one that is triple the size, you will need about three times as much time.

Secondly, as the post linked to earlier explains, multiple hashes together _are_ weaker than one large hash. The reason is that, once you've found a collision in a single hash function, it's easy to generate more of them. The example it gives is for 160-bit hash functions. It takes 2^80 attempts to brute-force a collision in such a function. Now, you might think it takes 2^80 * 2^80, i.e. 2^160 attempts to brute-force a message that collides in both of them (which would be the case for a single 320-bit hash function), but it doesn't. The reason is that it only takes you 2^80 * 80 attempts to get 2^80 colliding messages for the first 160-bit hash function...which gives you enough of them to brute-force a collision in the second. 2^80 * 80 is obviously a lot less that 2^80 * 2^80...

Re:Multiple Hash Functions (1)

PDAllen (709106) | more than 7 years ago | (#17737506)

It would almost certainly be much harder, but it would mean your hash overheads would triple (bandwidth and processing), and that is very important for hash codes.

How frustrating! (2, Interesting)

Paul Crowley (837) | more than 7 years ago | (#17736798)

The unfortunate thing is that in order to win this competition, the hash function submitted will have to be pretty conservative - very much like the SHA-512 family. There isn't time to analyze anything really new and have enough confidence in its security to bless it as the new standard for ever on. But if these attacks (and especially the recent attacks on the whole Merkle-Damgard structure) show us anything, it is that the old way turns out not to be the best way to build hash functions, and that more innovative ideas (eg Daemen et al's "belt and mill" proposal RADIOGATUN) should be the way forward.

What we need is for NIST to pull the rug under everyone near the end, and say "thanks for putting huge amounts of energy and hard work into designing and attacking all these hash functions, now you can all make new proposals based on what we've all learned and we'll start over again!"

Think of it as synchronization. (1)

Kadin2048 (468275) | more than 7 years ago | (#17738362)

As other people have pointed out, I'm not necessarily sure that these competitions really result in a whole lot of new development work per se. Rather, they serve as encouragement to researchers in the field, to take whatever they've been working on for the past few years, tidy it up and make it publishable, and submit it as a candidate for standardization.

The research into new functions progresses more or less on its own in the academic world most of the time. These competitions basically seek to tap into what's going on there, and re-synchronize the commercial/governmental world with whatever the state-of-the-art is in academic cryptography.

Competitions change the community (1)

Paul Crowley (837) | more than 7 years ago | (#17738530)

I disagree. The AES competition really galvanized the research community and resulted in some exciting innovations in cryptography - eg the application of bitslicing to the construction of ordinary primitives a la Serpent - and cryptanalsysis - eg Lucks's "saturation attack". And we're seeing similar effects with the eSTREAM competition, which in turn results from the way that all the stream ciphers proposed for NESSIE were broken.

One Word.... (4, Interesting)

tomstdenis (446163) | more than 7 years ago | (#17736814)

WHIRLPOOL.

It's a balanced design, an SPN to boot.

The big problem with the SHA's [and their elk] is that they're all UFN [unbalanced feistel networks], in particular they're source heavy. Which means the the branch/diffusion is minimal (e.g. it's possible to make inputs collide and cancel out differences).

SPN [substitution permutation networks] like WHIRLPOOL are balanced in their branch/diffusion.

Best of all, WHIRLPOOL is already out there. just a sign the paper!

Tom

Re:One Word.... (1)

G-Licious! (822746) | more than 7 years ago | (#17737048)

Amen!

And many, many thanks for your public domain implementation. :)

Re:One Word.... (1)

tomstdenis (446163) | more than 7 years ago | (#17737306)

You're welcome.

Glad to be of service.

Tom

Re:One Word.... (1)

delt0r (999393) | more than 7 years ago | (#17737144)

I assume it could be entered as a contender?

Re:One Word.... (1)

tomstdenis (446163) | more than 7 years ago | (#17737274)

I definitely hope so. I certainly don't look forward to the deluge of non-mathematical FROGHASH style [or HPC] submissions which IMHO are a total f'ing waste of time.

I'd say if you can't prove the branch of your primitive then you need a REALLY COMPELLING reason to submit it. Otherwise, be gone!

Tom

Re:One Word.... (1)

Paul Crowley (837) | more than 7 years ago | (#17737366)

Whirlpool is pretty slow. What do you think of Radiogatun?

Re:One Word.... (1)

tomstdenis (446163) | more than 7 years ago | (#17737420)

Whirlpool is slower than SHA ... but SHA is insecure so it's a meaningless comparison. memcpy() is faster than AES-CTR!!! :-)

Never heard of Radiogatun. To be honest i'm not really into crypto as much as I used to be. The whole debacle with the LT projects threw me off. I'd much rather play a Sonata on my piano than read a dry boring paper about cryptography. :-)

Tom

Re:One Word.... (1)

Fahrenheit 450 (765492) | more than 7 years ago | (#17738624)

Of course it's also slower than SHA-256 and SHA-512 which have no reported weaknesses, so ... not no meaningless.

Re:One Word.... (1)

tomstdenis (446163) | more than 7 years ago | (#17738958)

Presumably this is because people lost faith in the UFN approach. If that's the case, comparing new designs to the faster, but generally accepted as insecure designs, is not wise nor prudent.

I'm not saying WHIRLPOOL is perfect, but it's definitely a step in the right direction. Maybe this contest will spur a faster hash which is also as mathematically sound.

Tom

Moo... (0)

Anonymous Coward | more than 7 years ago | (#17737722)

the big problem with the SHA's [and their elk]

...or whatever sound that Elk make ;-)

I think you probably meant "ilk" as I don't think SHA's have any elk (or deer or moose or even caribou for that matter).

Sorry Tom, I know spelling flames suck, but I couldn't resist!

Re:Moo... (1)

tomstdenis (446163) | more than 7 years ago | (#17738254)

This is why we have editors...

I don't know what's scarrier, my loose grasp of formal language, or that I'm the author of two comp.sci text books :-(

hehehehe

Whatever, I do what I want!

Tom

Re:Moo... (2, Funny)

swillden (191260) | more than 7 years ago | (#17738464)

...or whatever sound that Elk make ;-)

They make a variety of sounds, most of them surprisingly high and squeaky for such large animals.

What about multi-hashing? (1)

jcr (53032) | more than 7 years ago | (#17736832)

So, I haven't studied this matter at all, but it seems to me that if you use more than one has algorithm on the same message, the chances of a different message generating the same has from both algorithms should be vanishingly small. Any cryptographers here care to fill me in?

-jcr

Re:What about multi-hashing? (1)

jcr (53032) | more than 7 years ago | (#17736848)

Please s/has/hash/ in the message above. I guess my finger-memory to type "has" is pretty strong.

-jcr

Re:What about multi-hashing? (1)

madcow_bg (969477) | more than 7 years ago | (#17737096)

Dunno about cryptographers, but Gentoo uses that already. Just see:
* portage-2.1.1.tar.bz2 MD5
* portage-2.1.1.tar.bz2 RMD160
* portage-2.1.1.tar.bz2 SHA1
* portage-2.1.1.tar.bz2 SHA256
* portage-2.1.1.tar.bz2 size

By the way, this is not a bad solution, since it uses industrial standards. The problem is that if you can't use this solution: i.e. want to store the hash of a password to check, you can't use this method, as brute forcing will still wield the same result.

Re:What about multi-hashing? (1)

pipatron (966506) | more than 7 years ago | (#17737190)

The chance of a different message generating the same hash basically only depends on the number of bits used in the hash. Sure, a combination of hash functions would give more bits. However, I strongly suspect that the combination of two hash functions to create one final hash would always be worse in that respect than a carefully designed hash function with the same number of bits.

No hashing algorithm can care for more than 2^bits number of different documents.

Not mathematically advantageous, but still useful. (2, Interesting)

Kadin2048 (468275) | more than 7 years ago | (#17738592)

I was wondering the same thing, and apparently so were a few other people besides. There's another discussion of it further up in the thread, and the quote which seems to be the final answer doesn't seem to be too hot on the idea. Here it is [slashdot.org] (quoting here from another source):
"...attempts to increase the security of hash functions by concatenating the outputs of two independent functions don't actually increase their theoretical security. For example, defining H(x) = SHA1(x) || RIPEMD160(x) still gives you only about 160 bits of strength, not 320 as you might have hoped. The reason is because you can find a 2^80 multicollision in SHA1 using only 80*2^80 work at most, by the previous paragraph. And among all of these 2^80 values you have a good chance that two of them will collide in RIPEMD160. So that is the total work to find a collision in the construction."
What this means to me is both 'yes,' and 'no.' Yes, using multiple hash algorithms protects against the failure of one algorithm. It avoids putting all your eggs in one basket. However, using multiple algorithms doesn't, in itself, offer any greater security than just using a single algorithm and a longer hash, assuming the algorithm is good. (By 'good,' I mean that it doesn't offer any ways of finding collisions that are significantly faster than brute force.)

Mathematically, using multiple algorithms may not offer much of an advantage, but practically, where you may by necessity have to work with algorithms that have flaws (because you have to pick from algorithms that are well-agreed-upon standards), or that may be discovered to have flaws in the future, it seems like a good way to hedge one's bets. Aside from the added complexity, there doesn't seem to be any compelling reasons not to do it, if time and computational power allow.

FYI (1, Offtopic)

trifish (826353) | more than 7 years ago | (#17736908)

This "news" is several months old.

Oh well I know, it's Slashdot.

Re:FYI (0, Offtopic)

Goaway (82658) | more than 7 years ago | (#17737358)

But Slashdot just recently duped the original news of the crack from several years back, so it's totally topical!

Re:FYI (1)

Fahrenheit 450 (765492) | more than 7 years ago | (#17738286)

No, it's not. The draft requirements and evaluation criteria were announced just yesterday.
Unless you live in a place where January 23, 2007 is several months ago....

Re:FYI (1)

trifish (826353) | more than 7 years ago | (#17738976)

Wrong. If you read the first line of the summary, it says: "NIST is preparing for a competition to augment and revise the current Secure Hash Standard." THAT is several months old news.

Wrong (1)

trifish (826353) | more than 7 years ago | (#17737304)

TF title is wrong. It says: "A Competition To Replace SHA-1". But, it's to replace the whole SHA family, which includes both SHA-1 and SHA-2.

SHA-2 includes SHA-256 and SHA-512. Why the whole SHA family? Because its design is not very trustworthy anymore since the "Chinese" attacks in 2005.

Re:Wrong (3, Informative)

Fahrenheit 450 (765492) | more than 7 years ago | (#17738410)

Again you are wrong (and somewhat right about the incorrect title at the same time, iI suppose). The point of this workshop is to revise and amend FIPS 180-2. Now, while the SHA-2 line of hashes are laid out in FIPS 180-2, it is not the case that SHA-2 and the like will be thrown out. They meet the requirements laid out in the call, and frankly NIST would be insane to not make it one of the workshop's submissions. It may very well fall out that the SHA-2 is just fine and indeed the best candidate submission.

As for the Chinese attacks, they haven't shown any real applicability to SHA-2 as of yet.

Re:Wrong (1)

trifish (826353) | more than 7 years ago | (#17739028)

As for the Chinese attacks, they haven't shown any real applicability to SHA-2 as of yet.

The keyword is "yet". The only thing that "protects" SHA-2 from the attacks are a bunch of extra rotations. If you think NIST and NSA are going to rely on that as a strong form or protection, you're a bit naive.

Some suggestions (1)

UnknowingFool (672806) | more than 7 years ago | (#17737402)

Can we make a real competition and call it Hashing Idol where every week another function gets voted out? Or they could compete in a head to head. Two functions enter ring. One function leaves.
...
Have I been watching too much TV?

Perfect Solution... (3, Funny)

evilviper (135110) | more than 7 years ago | (#17737444)

I have a perfect solution to the hashing problem, for verifying the data integrity between two points...

You simply have to find autistic twins. The one at the source looks through the MB file, then writes a hash, explaining that it "smells like 5 green triangles". If the twin at the destination agrees, you know you have a match.

It's nearly impossible, even to brute-force this method... I mean, you need to covertly aquire a sample of their DNA, and wait several years for the clone to mature.

Of course, this method's weakness is that it doesn't scale-up effectively. There are only so many autistic twins out there, and human cloning technology is still quite expensive.

Well... (1)

Jugalator (259273) | more than 7 years ago | (#17738758)

My expert advice is that now that we've seen what happened to the SHA-1 family, I think they should just skip the inevitable upcoming round of exploits for the SHA-2 family and go straight for a new SHA-3 family.

Are they going to pick a shit algorithm again? (-1, Troll)

Anonymous Coward | more than 7 years ago | (#17739016)

Like they picked AES when serpent and twofish are clearly better?

Perhaps you would better understand (1)

certain death (947081) | more than 7 years ago | (#17739076)

If you attend the conference (http://www.cwi.nl/projects/crypto/tcc07/) this year. It is in Amsterdam, so it might be a bit far for some to travel, but definitely worth attending. I hope to see some of you there.
Load More Comments
Slashdot Login

Need an Account?

Forgot your password?

Submission Text Formatting Tips

We support a small subset of HTML, namely these tags:

  • b
  • i
  • p
  • br
  • a
  • ol
  • ul
  • li
  • dl
  • dt
  • dd
  • em
  • strong
  • tt
  • blockquote
  • div
  • quote
  • ecode

"ecode" can be used for code snippets, for example:

<ecode>    while(1) { do_something(); } </ecode>