Topic: alternative for a bitcoin fork/rollback (Read 1347 times)

theoretically it would be a fork with consensus .
i assume something like this situation needs to happen: https://bitcointalksearch.org/topic/m.9531 (not to be mistaken with what OP is asking, this was  a bug but practically they invalidated the block 74638 hence the transactions that exploited the bug in 2010)

Your right, only the one having the private key to certain address can sign the transaction so it become valid. If you try include transaction moving the coins from any address without valid signature, your block would fail validation on all Bitcoin full nodes - "surprisingly" miner can not steal somebody else coins this way.

Theoretically it is possible. In every single Bitcoin full node there would need to be an exception coded for the block and txid of the transaction which returns the funds. Under every other circumstance, it would be considered invalid, and for anyone that is writing Bitcoin transaction validation code, it would be an annoyance to have to keep that exception there.

found the button and moved it.

you got the idea. i was thinking about the possibility that the mines spend the coins from the hackers address to a bitfinex address. they usually write transactions from a to b. so i thought they could also write a transaction where they force the attacker to send it back. i figuerd it would work fine if the majority was for it and i did not saw my mistake. I think i do now.
the miners do not have the private key of the hackers address to sign it. so this whole idea of mine would not result in a valid block. Am i right?

they are 2 differant animals and actually the problem wasn't with bitcoin it was with those using it, bitcoin functioned perfectly..... 

are you looking for a way to "reverse the transaction"?
because what you are explaining here sounds like you want that miner to "spend" the coins from the hackers bitcoin address and sends it back to bitfinex address. (not possible)

p.s. move your topic to "Development & Technical Discussion" board. there is a button on bottom left.

You can rebuild the blockchain using over 50% hash power using current rules. Instead of the transaction from bitfinex to hacker you replace it to other valid (signed) transaction: from bitfinex to new bitfinex address, basically a double spend. The problem is you need to rebuild the blockchain to get longest chain again and this means many days (since the hack) of invalidated blocks (and their rewards) while miners already used electricity to mine these blocks. Thus its not in miners interest, an even if bitfinex compensated all the invalidated block rewards to miners, the lost of confidence is lost when you can do 51% attack anytime to double spend any transaction even few days old one - would you use Bitcoin if after many days a transaction with 300+ confirmations could be still double spended ?

The second approach is hard fork to add special status just for this transaction similarly like the ethereum did, even much worse imo.

i never said i want a fork and clearly this is not happening anyway. but i also do not want this discussion in this thread.
if you want to discuss this please do so in an other thread as i am more in the technical side interested and not the moral or whatever.   

to those users I have one question: was the problem with bitcoin or the people using bitcoin?

if you think about this question for a moment then you realize how ridiculous suggesting a fork for bitcoin because bitfinex was hacked sounds like.

if we are going to have a fork for this hack then we should have a fork for every other hack.
the next day someone would ask for a fork because he has lost his private keys (forgot or lost to a keylogger) and has some funds in his bitcoin wallet and with your logic we should have a fork so he can get his funds back.

Most recently we had the DAO "Hack" and a following hard fork which resulted in ETH and ETC.
After the recent hack of Bitfinex some of the bitcoin users where thinking about a possibility to fork bitcoin with the intention of recovering the stolen funds.

Since Bitcoin and Ether use different technologies it is not that easy to compare those two cases and i do not want to do that. and it is probably to late anyway but i am interested in the theory behind it.

What i heard about how to make the hack undone is this:
we, or to be more precise over 51% of the miners, would need to go back to the transaction in the blockchain and ignore it and then restart the blockchain again.
At this point i am not sure if we had also to rewrite all other transactions manually or if it would be sufficient to let the mines start working again and write the transaction as if there where just in line to process. How would this be controlled?
Also the miners would have to run their equipment for some time for loss since the days since would be lost.

But my main question is about an alternative:
If we know the address of the stolen bitcoins and agree on an correction of the hack.
would it be possible to agree on a future blocknummber and one miner, who mines just this one block with just this one transaction from the hackers address back to bitfinex.
after this one block every thing goes back to normal.

this sounds very simple and i am sure there are some reasons why this is not going to work, but i do not see them and so i am asking you to explain it to me. thank you!