Bitcoin Forum

Bitcoin => Development & Technical Discussion => Topic started by: bearex on October 24, 2015, 11:36:43 AM



Title: The cost of Bitcoin attacks
Post by: bearex on October 24, 2015, 11:36:43 AM
How much does it costs, for an attacker to double spend a coin at:

0 confirmations
1 confirmation
...
?

Trying to get some borders for buyers to wait, so the double spend doesnt happen.

Thanks


Title: Re: The cost of Bitcoin attacks
Post by: redsn0w on October 24, 2015, 11:45:17 AM
How much does it costs, for an attacker to double spend a coin at:

0 confirmations
1 confirmation
...
?

Trying to get some borders for buyers to wait, so the double spend doesnt happen.

Thanks

To double spend (send two times the 'same' TX) you have only to use a custom client (the RbF -replace by fee | bitcoinCore). However, you can't double spend a tx once it has 1 or more confirmation.


Title: Re: The cost of Bitcoin attacks
Post by: bearex on October 24, 2015, 11:48:30 AM
So you can double spend a 0 confirmation transaction with no cost? :o


Title: Re: The cost of Bitcoin attacks
Post by: redsn0w on October 24, 2015, 11:52:37 AM
So you can double spend a 0 confirmation transaction with no cost? :o

You have to put & spend the correct fee(s), and choose the 'right' inputs.


Title: Re: The cost of Bitcoin attacks
Post by: bearex on October 24, 2015, 11:57:07 AM
Is this something that could be possible when a person is buying goods <100$? In a physical store.


Title: Re: The cost of Bitcoin attacks
Post by: achow101 on October 24, 2015, 01:39:24 PM
Is this something that could be possible when a person is buying goods <100$? In a physical store.
Theoretically, yes. They would need custom software on their mobile device that would allow them to spend the same inputs again and with a different fee.


Title: Re: The cost of Bitcoin attacks
Post by: shorena on October 24, 2015, 04:51:52 PM
Is this something that could be possible when a person is buying goods <100$? In a physical store.

Its possible, but you should read this thread -> https://bitcointalk.org/index.php?topic=1197567.0


Title: Re: The cost of Bitcoin attacks
Post by: Hugroll on October 24, 2015, 08:26:55 PM
How much does it costs, for an attacker to double spend a coin at:

0 confirmations
1 confirmation
...
?

Trying to get some borders for buyers to wait, so the double spend doesnt happen.

Thanks

To double spend (send two times the 'same' TX) you have only to use a custom client (the RbF -replace by fee | bitcoinCore). However, you can't double spend a tx once it has 1 or more confirmation.
i remember seeing this video on youtube that explained that double spending can have more than 1 confirmation but the two tx would have to be confirmed at the same time? but the chances are very low of that happening. Does it work like that or if you double spend theres a 0% chance of both of them getting 2 confirmations each?


Title: Re: The cost of Bitcoin attacks
Post by: redsn0w on October 24, 2015, 09:04:19 PM
How much does it costs, for an attacker to double spend a coin at:

0 confirmations
1 confirmation
...
?

Trying to get some borders for buyers to wait, so the double spend doesnt happen.

Thanks

To double spend (send two times the 'same' TX) you have only to use a custom client (the RbF -replace by fee | bitcoinCore). However, you can't double spend a tx once it has 1 or more confirmation.
i remember seeing this video on youtube that explained that double spending can have more than 1 confirmation but the two tx would have to be confirmed at the same time? but the chances are very low of that happening. Does it work like that or if you double spend theres a 0% chance of both of them getting 2 confirmations each?


It is impossible to get confirmed the 2 transactions (double spend). Because for a successful double spend one of the two tx must not be confirmed, this is how it works.


Title: Re: The cost of Bitcoin attacks
Post by: amaclin on October 25, 2015, 12:21:16 PM
However, you can't double spend a tx once it has 1 or more confirmation.
wrong


Title: Re: The cost of Bitcoin attacks
Post by: redsn0w on October 25, 2015, 12:26:39 PM
However, you can't double spend a tx once it has 1 or more confirmation.
wrong

Can you make a test?
This is my address bitcoin 1GkGt3218AM8FU7ZGDCeZS4LWZR8xeZcjf  send me a PM once you'll start.



Good luck !


Title: Re: The cost of Bitcoin attacks
Post by: amaclin on October 25, 2015, 01:34:35 PM
Can you make a test?
This is my address bitcoin 1GkGt3218AM8FU7ZGDCeZS4LWZR8xeZcjf  send me a PM once you'll start.
Not today, man.
On mainnet.
We can test this on testnet or on altcoin network.


Title: Re: The cost of Bitcoin attacks
Post by: redsn0w on October 25, 2015, 01:37:05 PM
Can you make a test?
This is my address bitcoin 1GkGt3218AM8FU7ZGDCeZS4LWZR8xeZcjf  send me a PM once you'll start.
Not today, man.
On mainnet.
We can test this on testnet or on altcoin network.


Why not the real BTC blockchain?


Title: Re: The cost of Bitcoin attacks
Post by: yakuza699 on October 25, 2015, 01:46:14 PM
Can you make a test?
This is my address bitcoin 1GkGt3218AM8FU7ZGDCeZS4LWZR8xeZcjf  send me a PM once you'll start.
Not today, man.
On mainnet.
We can test this on testnet or on altcoin network.
Do it on litecoin/dogecoin network.Are you going to orphan block that included your transaction?


Title: Re: The cost of Bitcoin attacks
Post by: criptix on October 25, 2015, 02:04:37 PM
Can you make a test?
This is my address bitcoin 1GkGt3218AM8FU7ZGDCeZS4LWZR8xeZcjf  send me a PM once you'll start.
Not today, man.
On mainnet.
We can test this on testnet or on altcoin network.


Why not the real BTC blockchain?

because he cant.


Title: Re: The cost of Bitcoin attacks
Post by: redsn0w on October 25, 2015, 02:21:55 PM
Can you make a test?
This is my address bitcoin 1GkGt3218AM8FU7ZGDCeZS4LWZR8xeZcjf  send me a PM once you'll start.
Not today, man.
On mainnet.
We can test this on testnet or on altcoin network.


Why not the real BTC blockchain?

because he cant.

He said I'm wrong, but you can't double spend a 1 conf. BTC transaction. It's real impossible  ::).


Title: Re: The cost of Bitcoin attacks
Post by: criptix on October 25, 2015, 02:23:56 PM
Can you make a test?
This is my address bitcoin 1GkGt3218AM8FU7ZGDCeZS4LWZR8xeZcjf  send me a PM once you'll start.
Not today, man.
On mainnet.
We can test this on testnet or on altcoin network.


Why not the real BTC blockchain?

because he cant.

He said I'm wrong, but you can't double spend a 1 conf. BTC transaction. It's real impossible  ::).

you would need to fork the blockchain. i'm sure he cant do it.


Title: Re: The cost of Bitcoin attacks
Post by: amaclin on October 25, 2015, 02:46:37 PM
you would need to fork the blockchain. i'm sure he cant do it.
I can not do it today on mainnet.
I haven't said that it is impossible always.


Title: Re: The cost of Bitcoin attacks
Post by: yakuza699 on October 25, 2015, 02:51:47 PM
you would need to fork the blockchain. i'm sure he cant do it.
I can not do it today on mainnet.
I haven't said that it is impossible always.
All you need is just a simple orphan block.


Title: Re: The cost of Bitcoin attacks
Post by: redsn0w on October 25, 2015, 03:03:54 PM
you would need to fork the blockchain. i'm sure he cant do it.
I can not do it today on mainnet.
I haven't said that it is impossible always.

Maybe tomorrow?


Title: Re: The cost of Bitcoin attacks
Post by: Decoded on October 27, 2015, 11:13:57 AM
It's near impossible to get a double transaction with even 1 confo.

Just think about the usual 3 confos... Jeez!


Title: Re: The cost of Bitcoin attacks
Post by: amaclin on October 27, 2015, 01:47:20 PM
It's near impossible to get a double transaction with even 1 confo.
yes. in mainnet. today


Title: Re: The cost of Bitcoin attacks
Post by: fbueller on October 27, 2015, 03:27:49 PM
Check the blockchain.info page on blockchain forks. You'll see one-block-deep forks happen all the time. Your application needs to be able to undo a 'successful payment' because reorgs occur.

Should a certain someone ;) have his signature malleator running, a transaction with TXID A that confirmed in block A might have a different TXID if block B replaces block A.

So you have to be able to totally undo a successful transaction if the tip changes and orphans any block you have that includes a payment.