Bitcoin Forum
March 28, 2024, 08:59:46 AM *
News: Latest Bitcoin Core release: 26.0 [Torrent]
 
   Home   Help Search Login Register More  
Pages: [1]
  Print  
Author Topic: Rollback a chain...How it works??  (Read 271 times)
criptoraily (OP)
Newbie
*
Offline Offline

Activity: 2
Merit: 0


View Profile
July 27, 2019, 08:15:16 AM
 #1

In case of a 51% attack to roll back any chain to a specific moment, the chain could be rolled back at any state in the past in  matter of minutes/blocks,  or the attack should last as much the time passed from the event that  should be reverted till the start of the attack?
1711616386
Hero Member
*
Offline Offline

Posts: 1711616386

View Profile Personal Message (Offline)

Ignore
1711616386
Reply with quote  #2

1711616386
Report to moderator
1711616386
Hero Member
*
Offline Offline

Posts: 1711616386

View Profile Personal Message (Offline)

Ignore
1711616386
Reply with quote  #2

1711616386
Report to moderator
Advertised sites are not endorsed by the Bitcoin Forum. They may be unsafe, untrustworthy, or illegal in your jurisdiction.
1711616386
Hero Member
*
Offline Offline

Posts: 1711616386

View Profile Personal Message (Offline)

Ignore
1711616386
Reply with quote  #2

1711616386
Report to moderator
keychainX
Member
**
Offline Offline

Activity: 372
Merit: 53

Telegram @keychainX


View Profile WWW
July 27, 2019, 08:32:43 AM
Merited by bones261 (2), ABCbits (1)
 #2

In case of a 51% attack to roll back any chain to a specific moment, the chain could be rolled back at any state in the past in  matter of minutes/blocks,  or the attack should last as much the time passed from the event that  should be reverted till the start of the attack?

What you refer to is called in general terms a re-org.

It happened to most of the chains and basically more than 50% of the miners need to agree to the re-org or a hard-fork.

You have several cases that happened here:
https://nulltx.com/top-4-cryptocurrency-blockchain-reorganization-events/

Or by definition by bitcoin.it
https://en.bitcoin.it/wiki/Chain_Reorganization

or more detailed description here
https://bitcoin.org/en/blockchain-guide#introduction

About bitcoin SV here: https://coinspice.io/hash-war/hash-war-sv-chain-block-reorg-attack-under-way/

or about BCH here: https://ambcrypto.com/bitcoin-cash-bch-is-unsecure-and-the-recent-bch-chain-reorg-is-a-sht-show-says-blockstream-cso/

Have fun reading!

/KX

o_e_l_e_o
In memoriam
Legendary
*
Offline Offline

Activity: 2268
Merit: 18497


View Profile
July 27, 2019, 09:09:00 AM
Merited by dothebeats (1), ABCbits (1)
 #3

In case of a 51% attack to roll back any chain to a specific moment, the chain could be rolled back at any state in the past in  matter of minutes/blocks,  or the attack should last as much the time passed from the event that  should be reverted till the start of the attack?
Neither exactly, but your second statement is closer to the truth.

When an attacker launches a 51% attack, they have to start mining an alternate chain from the block they want to reverse. Only when their alternate chain becomes longer than the current chain will it displace it and be classed as the "main chain". If they start their 51% attack from the last block, then they only have to overcome a one block deficit to become the main chain. If they start from 100 blocks ago, then they need to overcome a 100 block deficit. Bear in mind that during their attack, honest miners will still be mining on the current chain, so by the time the attacker has overcome this 100 block deficit, there will be further blocks to overcome, depending on the currency hashrate and difficulty of the network.

This is why transactions are "safer" with more confirmations, and many places require 6 confirmations before considering a transaction fully complete. A 1 block deficit would be relatively easy to overcome, whereas it gets progressively harder with each additional confirmation, as a 51% attacker has to start from a bigger deficit.
ABCbits
Legendary
*
Offline Offline

Activity: 2828
Merit: 7327



View Profile
July 27, 2019, 09:13:53 AM
Merited by bones261 (2), dothebeats (1)
 #4

That depends on the blockchain protocol & consensus. If other miner & node decide to ignore block created due to 51% attack, re-org process will happen instantly, but will cause hard-fork.
If it's ignored, blocks generated due to 51% attack will stay on longest chain.

Take note chain rollback/re-org doesn't always happen because 51% attack. It might happen when there's orphan block ,where 2 different miner create a block almost at same time and one of them isn't on longest chain.

.
.BLACKJACK ♠ FUN.
█████████
██████████████
████████████
█████████████████
████████████████▄▄
░█████████████▀░▀▀
██████████████████
░██████████████
████████████████
░██████████████
████████████
███████████████░██
██████████
CRYPTO CASINO &
SPORTS BETTING
▄▄███████▄▄
▄███████████████▄
███████████████████
█████████████████████
███████████████████████
█████████████████████████
█████████████████████████
█████████████████████████
███████████████████████
█████████████████████
███████████████████
▀███████████████▀
█████████
.
HeRetiK
Legendary
*
Offline Offline

Activity: 2884
Merit: 2061



View Profile
July 29, 2019, 01:02:45 PM
Merited by dothebeats (1)
 #5

This is why transactions are "safer" with more confirmations, and many places require 6 confirmations before considering a transaction fully complete. A 1 block deficit would be relatively easy to overcome, whereas it gets progressively harder with each additional confirmation, as a 51% attacker has to start from a bigger deficit.

Note that in case of a 51% attack no confirmation count is safe for as long as the adversary holds the majority hashing power.

6 confirmations are the recommendation to achieve reasonable security for most cases, but in case of an actual 51% attack even 6 confirmations won't suffice.

Check here for reference:
https://people.xiph.org/~greg/attack_success.html
dothebeats
Legendary
*
Offline Offline

Activity: 3598
Merit: 1352


Excel is fun


View Profile
July 29, 2019, 04:06:01 PM
 #6

This is why transactions are "safer" with more confirmations, and many places require 6 confirmations before considering a transaction fully complete. A 1 block deficit would be relatively easy to overcome, whereas it gets progressively harder with each additional confirmation, as a 51% attacker has to start from a bigger deficit.

Sadly though, some exchanges/services still impose a 1-3 confirmation rule before anything sent to them becomes valid, which I think is somewhat underestimating the possibility that something like this could happen at any point in time.

The higher the deficit/confirmations, the harder it is to nullify/remove/modify a certain transaction within the blockchain, which means the higher the costs the attacker has to spend the longer he keeps on doing the said attack, rendering it a somewhat futile effort in the long run. It only has occurred to me recently how wonderfully made the protocol is due to intricate things like this that is, for the most part, being underestimated by people. Call it as you may but I'd like to call a 51% attack--especially from the attacker's standpoint--a race against time and money.

▄▄███████▄▄
▄██████████████▄
▄██████████████████▄
▄████▀▀▀▀███▀▀▀▀█████▄
▄█████████████▄█▀████▄
███████████▄███████████
██████████▄█▀███████████
██████████▀████████████
▀█████▄█▀█████████████▀
▀████▄▄▄▄███▄▄▄▄████▀
▀██████████████████▀
▀███████████████▀
▀▀███████▀▀
.
 MΞTAWIN  THE FIRST WEB3 CASINO   
.
.. PLAY NOW ..
dothebeats
Legendary
*
Offline Offline

Activity: 3598
Merit: 1352


Excel is fun


View Profile
July 29, 2019, 08:29:41 PM
 #7

Because it's not attractive for trader and short-term investor, especially due to the fact cryptocurrency price could changed significantly in few minutes.

The exchange must realize they're trading some security for user's convenience, but AFAIK few exchange don't provide instant/automated withdraw if the amount is pretty big.

Exactly. This might have been implemented due to multiple user requests IMO especially during the days where arbitraging is still pretty much a good thing to do because KYC isn't implemented that hard on every exchanges. It's a compromise that suits the needs of traders, especially those constantly chasing actions but then again, we'll never know when would these attacks go out on the open.

As for instant withdraws though, most exchanges still follow the rule of allowing at least 10 minutes before processing the withdrawals, even on small amounts. A local exchange still does withdrawals manually, and I can understand that for their convenience and security as well.

▄▄███████▄▄
▄██████████████▄
▄██████████████████▄
▄████▀▀▀▀███▀▀▀▀█████▄
▄█████████████▄█▀████▄
███████████▄███████████
██████████▄█▀███████████
██████████▀████████████
▀█████▄█▀█████████████▀
▀████▄▄▄▄███▄▄▄▄████▀
▀██████████████████▀
▀███████████████▀
▀▀███████▀▀
.
 MΞTAWIN  THE FIRST WEB3 CASINO   
.
.. PLAY NOW ..
TimeBits
Member
**
Offline Offline

Activity: 224
Merit: 62


View Profile
July 29, 2019, 10:31:59 PM
 #8

The answer is not going back to the future, it`s not letting that future happen to begin with. You have to be out of your mind like Binance was when they wanted to do one.

Yah so all of you people who did trades on the network because of our shit security on our exchange, we gonna revert the entire network, yah right. #fundsarenotsafu
o_e_l_e_o
In memoriam
Legendary
*
Offline Offline

Activity: 2268
Merit: 18497


View Profile
July 30, 2019, 08:56:39 PM
 #9

we gonna revert the entire network, yah right. #fundsarenotsafu
It has happened to many other coins, including some major ones. Ethereum Classic exists because some miners didn't agree with the other miners who decided to hard fork Ethereum to reverse a transaction they didn't like. Bitcoin Cash was recently 51% attacked by the two biggest mining pools joining together, again because someone else made some transactions they didn't like. It should really be a huge concern for holders of these coins, knowing that their coin isn't really immutable or decentralized, when a small number of users can reverse transactions like that.
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!