Bitcoin Forum
November 12, 2024, 11:32:08 AM *
News: Check out the artwork 1Dq created to commemorate this forum's 15th anniversary
 
   Home   Help Search Login Register More  
Pages: [1]
  Print  
Author Topic: Bitcoin and its Malleability? Important Questions!  (Read 689 times)
Trance (OP)
Hero Member
*****
Offline Offline

Activity: 551
Merit: 500


View Profile
February 12, 2014, 03:11:20 PM
 #1

I have been taking a cryptography class and I'm wondering if someone can assist with my question in regards to understanding the "transaction malleability" BTCitcoin vulnerability that was supposedly exploited by Mt Gox customers to double-spend their BTCitcoin' (see press release below)? I'm surprised to learn that the SHA256 signature of the block chain is "malleable". I thought any secure MAC protocol would prevent this sort of 'attack'. And I don't see where the security proof is broken for the BTCitcoin protocol.

https://www.mtgox.com/press_release_20140210.html

Some people are so poor ALL they have is money
cr1776
Legendary
*
Offline Offline

Activity: 4214
Merit: 1313


View Profile
February 12, 2014, 03:13:35 PM
 #2

I have been taking a cryptography class and I'm wondering if someone can assist with my question in regards to understanding the "transaction malleability" BTCitcoin vulnerability that was supposedly exploited by Mt Gox customers to double-spend their BTCitcoin' (see press release below)? I'm surprised to learn that the SHA256 signature of the block chain is "malleable". I thought any secure MAC protocol would prevent this sort of 'attack'. And I don't see where the security proof is broken for the BTCitcoin protocol.

https://www.mtgox.com/press_release_20140210.html

See:
https://bitcointalk.org/index.php?topic=460944.0
un_ordinateur
Full Member
***
Offline Offline

Activity: 157
Merit: 100


View Profile
February 12, 2014, 03:46:28 PM
 #3

The SHA256 signature of a transaction serves as it txid. It is NOT malleable. However, the data inside a transaction message is somewhat arbitraty. Although the inputs, the outputs, and the amouts are signed by the sender (and are thus not changeable without invalidating the transaction), one can append some other arbitrary data to the transaction, while keeping it valid. Althoug the effect of the transaction is going to be the same (a certain amount chosen by the sender is going to be sent to the receiver), the associated txid is going to be different.

Thus it is NOT a double spend attact. One CANNOT take your coins without knowing your private key, nor one CANNOT send the same coin to two different persons.

However one should NOT use the txid as a way to uniquely identify a transaction. It should instead use a combinaison of inputs, outputs and amount to acertain that the transaction went trought (or not).

MtGox did not do that, they used txid. Some users exploited that, and claimed their withdrawal failed to their costumer service, when in fact it went trough but with a different txid than what MtGox tought. MtGox then wrongly resend a transaction.

It was not a double spend, it was a double send.
Trance (OP)
Hero Member
*****
Offline Offline

Activity: 551
Merit: 500


View Profile
February 12, 2014, 04:35:31 PM
 #4

I have been taking a cryptography class and I'm wondering if someone can assist with my question in regards to understanding the "transaction malleability" BTCitcoin vulnerability that was supposedly exploited by Mt Gox customers to double-spend their BTCitcoin' (see press release below)? I'm surprised to learn that the SHA256 signature of the block chain is "malleable". I thought any secure MAC protocol would prevent this sort of 'attack'. And I don't see where the security proof is broken for the BTCitcoin protocol.

https://www.mtgox.com/press_release_20140210.html

See:
https://bitcointalk.org/index.php?topic=460944.0


+1

Thanks!

Some people are so poor ALL they have is money
Pages: [1]
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.19 | SMF © 2006-2009, Simple Machines Valid XHTML 1.0! Valid CSS!