Author

Topic: Great explanation about 51% attack on Alts. (Read 841 times)

legendary
Activity: 1008
Merit: 1000
Accidental forks are created from bugs in software making two distinct forks with clients from one version only seeing one valid, while clients from the other version only see the other as valid (see BIP 0050).

If a pool has 51% of the hash rate (or more) it's not a problem for the network unless the pool owner (or a hacker with administrative privileges to that pool) decides he wants to start hiding his chain from the network.

PoW 51% + doublespend in a few words:
1) Everyone else has <50% of hash and is mining a chain.
2) Attacker mines a longer chain in secret, not reporting it to the network.  Attacker can do this because he has more hash power.
3) Attacker spends on the <50% fork, waits for 6 (or whatever) confs to get it to spend at the exchange, then exchanges it for cash or whatever.
4) Attacker dumps his chain onto the network.  Entire network invalidates the <50% chain, replacing it with the attacker's chain.  Attacker's coins are now returned to him to respend at will.  Not only this, but all the blocks mined in the <50% chain are now invalidated, so everyone not the attacker loses their blockrewards.

So, end result is that attacker gets 100% of the network coins from block reward and can double spend freely.

How the attacker could have money to spend in step 3 if he s holding a secret chain? His coins wouldnt be valid until he dumps his chain?

He has to have the coins beforehand.

So if you want to 51%, first buy up; then mine your chain while selling your stock; then overwrite and again send the coins somewhere.

The coins earned by mining the blocks in your longer chain is no more than what you would get by mining on the normal chain anyway. The only profit is the double spend. Which makes the attack on FTC quite baffling as there was no significant double spend.
hero member
Activity: 826
Merit: 1001
@Bit_John
Great article I highly encourage folks to read through every line. Digital currency in general is risk business. Big risk Big rewards
full member
Activity: 210
Merit: 100
Accidental forks are created from bugs in software making two distinct forks with clients from one version only seeing one valid, while clients from the other version only see the other as valid (see BIP 0050).

If a pool has 51% of the hash rate (or more) it's not a problem for the network unless the pool owner (or a hacker with administrative privileges to that pool) decides he wants to start hiding his chain from the network.

PoW 51% + doublespend in a few words:
1) Everyone else has <50% of hash and is mining a chain.
2) Attacker mines a longer chain in secret, not reporting it to the network.  Attacker can do this because he has more hash power.
3) Attacker spends on the <50% fork, waits for 6 (or whatever) confs to get it to spend at the exchange, then exchanges it for cash or whatever.
4) Attacker dumps his chain onto the network.  Entire network invalidates the <50% chain, replacing it with the attacker's chain.  Attacker's coins are now returned to him to respend at will.  Not only this, but all the blocks mined in the <50% chain are now invalidated, so everyone not the attacker loses their blockrewards.

So, end result is that attacker gets 100% of the network coins from block reward and can double spend freely.

How the attacker could have money to spend in step 3 if he s holding a secret chain? His coins wouldnt be valid until he dumps his chain?
member
Activity: 84
Merit: 10
Accidental forks are created from bugs in software making two distinct forks with clients from one version only seeing one valid, while clients from the other version only see the other as valid (see BIP 0050).

If a pool has 51% of the hash rate (or more) it's not a problem for the network unless the pool owner (or a hacker with administrative privileges to that pool) decides he wants to start hiding his chain from the network.

PoW 51% + doublespend in a few words:
1) Everyone else has <50% of hash and is mining a chain.
2) Attacker mines a longer chain in secret, not reporting it to the network.  Attacker can do this because he has more hash power.
3) Attacker spends on the <50% fork, waits for 6 (or whatever) confs to get it to spend at the exchange, then exchanges it for cash or whatever.
4) Attacker dumps his chain onto the network.  Entire network invalidates the <50% chain, replacing it with the attacker's chain.  Attacker's coins are now returned to him to respend at will.  Not only this, but all the blocks mined in the <50% chain are now invalidated, so everyone not the attacker loses their blockrewards.

So, end result is that attacker gets 100% of the network coins from block reward and can double spend freely.

The link I shared it tells something different . It says that the attackers coins get invalidated after he has converted them to BTC and other currencies and thus the exchange loses the coins not the normal user. But your story is opposite. Don't know which one to believe.

No, it is the same story.  The <50% chain has the coins spent, attacker's chain has coins unspent.  Exchange sees <50% chain, so it sees them spent (sent to exchange).  Attacker then swaps these coins for something else, then dumps their chain onto the network.  The transaction in which the coins are spent on the <50% chain become invalidated because the whole chain is invalidated.  Thus the attacker now regains his coins (because they are unspent on his chain) while the exchange loses these coins (because they came from a now invalid transaction).

Taco you seem quite busy but I've posted a request for quotes , Willingboro pay a dev forma simple dynamic checkpointing system for sCrypt , and a possible blockchain " audit system"   - for the whole community , but will pay for work , topic is on page 1 .
sr. member
Activity: 462
Merit: 250
PXC Research Team

No, it is the same story.  The <50% chain has the coins spent, attacker's chain has coins unspent.  Exchange sees <50% chain, so it sees them spent (sent to exchange).  Attacker then swaps these coins for something else, then dumps their chain onto the network.  The transaction in which the coins are spent on the <50% chain become invalidated because the whole chain is invalidated.  Thus the attacker now regains his coins (because they are unspent on his chain) while the exchange loses these coins (because they came from a now invalid transaction).

Now I got it perfectly. It means that the exchange and normal miners both loose their coins. Smiley
legendary
Activity: 1484
Merit: 1005
Accidental forks are created from bugs in software making two distinct forks with clients from one version only seeing one valid, while clients from the other version only see the other as valid (see BIP 0050).

If a pool has 51% of the hash rate (or more) it's not a problem for the network unless the pool owner (or a hacker with administrative privileges to that pool) decides he wants to start hiding his chain from the network.

PoW 51% + doublespend in a few words:
1) Everyone else has <50% of hash and is mining a chain.
2) Attacker mines a longer chain in secret, not reporting it to the network.  Attacker can do this because he has more hash power.
3) Attacker spends on the <50% fork, waits for 6 (or whatever) confs to get it to spend at the exchange, then exchanges it for cash or whatever.
4) Attacker dumps his chain onto the network.  Entire network invalidates the <50% chain, replacing it with the attacker's chain.  Attacker's coins are now returned to him to respend at will.  Not only this, but all the blocks mined in the <50% chain are now invalidated, so everyone not the attacker loses their blockrewards.

So, end result is that attacker gets 100% of the network coins from block reward and can double spend freely.

The link I shared it tells something different . It says that the attackers coins get invalidated after he has converted them to BTC and other currencies and thus the exchange loses the coins not the normal user. But your story is opposite. Don't know which one to believe.

No, it is the same story.  The <50% chain has the coins spent, attacker's chain has coins unspent.  Exchange sees <50% chain, so it sees them spent (sent to exchange).  Attacker then swaps these coins for something else, then dumps their chain onto the network.  The transaction in which the coins are spent on the <50% chain become invalidated because the whole chain is invalidated.  Thus the attacker now regains his coins (because they are unspent on his chain) while the exchange loses these coins (because they came from a now invalid transaction).
sr. member
Activity: 462
Merit: 250
PXC Research Team
Accidental forks are created from bugs in software making two distinct forks with clients from one version only seeing one valid, while clients from the other version only see the other as valid (see BIP 0050).

If a pool has 51% of the hash rate (or more) it's not a problem for the network unless the pool owner (or a hacker with administrative privileges to that pool) decides he wants to start hiding his chain from the network.

PoW 51% + doublespend in a few words:
1) Everyone else has <50% of hash and is mining a chain.
2) Attacker mines a longer chain in secret, not reporting it to the network.  Attacker can do this because he has more hash power.
3) Attacker spends on the <50% fork, waits for 6 (or whatever) confs to get it to spend at the exchange, then exchanges it for cash or whatever.
4) Attacker dumps his chain onto the network.  Entire network invalidates the <50% chain, replacing it with the attacker's chain.  Attacker's coins are now returned to him to respend at will.  Not only this, but all the blocks mined in the <50% chain are now invalidated, so everyone not the attacker loses their blockrewards.

So, end result is that attacker gets 100% of the network coins from block reward and can double spend freely.

The link I shared it tells something different . It says that the attackers coins get invalidated after he has converted them to BTC and other currencies and thus the exchange loses the coins not the normal user. But your story is opposite. Don't know which one to believe.
legendary
Activity: 1022
Merit: 1001
Accidental forks are created from bugs in software making two distinct forks with clients from one version only seeing one valid, while clients from the other version only see the other as valid (see BIP 0050).

If a pool has 51% of the hash rate (or more) it's not a problem for the network unless the pool owner (or a hacker with administrative privileges to that pool) decides he wants to start hiding his chain from the network.

PoW 51% + doublespend in a few words:
1) Everyone else has <50% of hash and is mining a chain.
2) Attacker mines a longer chain in secret, not reporting it to the network.  Attacker can do this because he has more hash power.
3) Attacker spends on the <50% fork, waits for 6 (or whatever) confs to get it to spend at the exchange, then exchanges it for cash or whatever.
4) Attacker dumps his chain onto the network.  Entire network invalidates the <50% chain, replacing it with the attacker's chain.  Attacker's coins are now returned to him to respend at will.  Not only this, but all the blocks mined in the <50% chain are now invalidated, so everyone not the attacker loses their blockrewards.

So, end result is that attacker gets 100% of the network coins from block reward and can double spend freely.

Great explanation. Thankyou  Smiley
legendary
Activity: 1484
Merit: 1005
Accidental forks are created from bugs in software making two distinct forks with clients from one version only seeing one valid, while clients from the other version only see the other as valid (see BIP 0050).

If a pool has 51% of the hash rate (or more) it's not a problem for the network unless the pool owner (or a hacker with administrative privileges to that pool) decides he wants to start hiding his chain from the network.

PoW 51% + doublespend in a few words:
1) Everyone else has <50% of hash and is mining a chain.
2) Attacker mines a longer chain in secret, not reporting it to the network.  Attacker can do this because he has more hash power.
3) Attacker spends on the <50% fork, waits for 6 (or whatever) confs to get it to spend at the exchange, then exchanges it for cash or whatever.
4) Attacker dumps his chain onto the network.  Entire network invalidates the <50% chain, replacing it with the attacker's chain.  Attacker's coins are now returned to him to respend at will.  Not only this, but all the blocks mined in the <50% chain are now invalidated, so everyone not the attacker loses their blockrewards.

So, end result is that attacker gets 100% of the network coins from block reward and can double spend freely.
hero member
Activity: 798
Merit: 531
Crypto is King.
Good read, but this specifically refers to an intentional attack. What about 'accidental' forks due to someone or a pool having more than 51% of the net hash rate? Is this possible?
Most certainly Watson. Although, a responsible pool owner will put in an auto script to halt new members from registering and overpowering the network.
legendary
Activity: 1022
Merit: 1001
Good read, but this specifically refers to an intentional attack. What about 'accidental' forks due to someone or a pool having more than 51% of the net hash rate? Is this possible?
sr. member
Activity: 462
Merit: 250
PXC Research Team
Hello Guys,

Here is a really good explanation of 51% attack on altcoin for those who want to know what 51% attack is.


http://www.bitcoinmoney.com/post/53207712103/your-bitcoins-are-not-safe-at-alt-coin-exchanges

Edit: If I helped someone feel free to share some wealth. Grin
Jump to: