Topic: We're not going anywhere, until the 51% question is answered (Read 6400 times)

51% attack vulnerability is just one of bitcoins core flaws that won't be fixed. hybrid or pure PoS mining is adressing this issue.

Even if the people who run the pool, stay nice, then there can always be security vulnerabilities that will give access to people who don't have plans to stay nice.
And 51% attack isn't only a threat if a single pool reaches it. If already 2 pools make up 51%, and both of their security is compromised, then BTC is still screwed.

you guys figured it out yet ?
report back when figured... if problem.. fix... if not... report back amusing story

thanks
so many threads about this...

I have no particulator solution i.e. I'm providing requirements but not analysis/design.

What changes would you propose to implement to make solo mining easier?

The Bitcoin difficulty is based on the network hashrate and the network has a hashrate that makes it so most small to medium sized miners would not likely find any blocks during any two week period. Pool mining is the solution to this problem.

I see 40%
https://blockchain.info/pools?timespan=24hrs

I guess the market forgets after a while short term attention span and or it gave the community time to make a solution and seeing none is just going back to its old routes.

GHash is currently 44% ... What the fuck people

These guys seem to think we are doomed:

http://hackingdistributed.com/2014/06/13/time-for-a-hard-bitcoin-fork/

http://hackingdistributed.com/2014/06/16/how-a-mining-monopoly-can-attack-bitcoin/

While I was somewhat comforted by the fact that a 51% attacker can't take my coins, the fact that they can basically destroy the other pools is a bit alarming.  They also do a good job talking about what Death and Taxes was referring to as the human problem.

I just don't see what keeps other pools from being as attractive as Ghash.io, that would help with the issue.

Also, even if they did it and all PoW coins crashed to 0$. Peercoin would still remain working as if nothing happened. Thus, they cannot destroy crypto currencies as a concept by performing a 51% attack. All sane bitcoin investors have diversified between alts long time ago, so even their wealth cannot be damaged by such an attack. All in all, 51% attack is not a realistic threat. At least not for the (s)perma-bulls.

no, you're completely right. they wouldn't do it. it would be pointless, unless your point was to destroy bitcoin and the wealth of all the miners involved. the retards like failing seem to think some 'organization' is going to 'take control' of ghash and essentially perform a very expensive terrorist attack on bitcoin. its not to make money, even though the only reason they've cited has been 'greed'. but those guys are VERY dumb so i wouldn't take much of what they say into consideration.

It isn't a protocol issue, it is more a human behavior issue and that is a lot harder to "fix".   Logically a miners largest cost is the electrical bill and electrical bills are only due once a month.  Even a pool with 2% of the hashrate will have a 95% confidence to be +/- 10% of the expected revenue after 30 days.   However most miners want a reward now, right now.  That probably isn't going to change. p2pool is a good example.  Right now p2pool shares are ~1000x easier than solo mining however most miners do use it.  Logically even if you radically changed the protocol such that solo mining was 1000x easier most miners would probably not use it. 

Still the risk of 51% attack from pools is overstated.  Lets say right now ghash had 70% of the hashrate.  How would they produce an attack chain?  Well they would need to stop mining the main chain.  ghash on average would go from producing say 8 blocks in a 2 hour period to producing zero.  They can't mine both the main chain and attack chain at the same time.

The risk is small but it could be made even smaller by making it easy for miners to verify they are extending the longest chain.  For example the current block hash of the last block in the longest chain is 0000000000000000302b107dcfdaf35d3b09366638e19aa24ffa91dd7f91d57c and thus all miners should currently be working on a block that extends that block.  Right now if all pool miners are extending block 0000000000000000302b107dcfdaf35d3b09366638e19aa24ffa91dd7f91d57c then you know that no 51% attack is ongoing.  Even if all pools were infiltrated the only thing an attacker could do is steal the mining revenue (by changing the coinbase tx). 

It would be possible to modify the blockheader so that miners (and third party observers) can verify that they are mining the longest legit chain.  This could be done in a backwards compatible manner (won't break existing ASIC hardware) by including a partial block hash of the prior block in the last segment of the block header.  There is limited space in the final segment of the blockheader (the one that the miner adds the nonce and thus must always be provided in full by the pool) so the full blockheader is far too long.  By extending the blockheader and putting the rightmost 8 bytes of the prior blockhash (i.e. 4ffa91dd7f91d57c for the block above) in the last segment it would be easy to spot attempted misuse of miner computing power.  The mining software (using SPV protocol) would maintain an independent record of the longest chain (and the last block extending it) and would verify the work provided by the pool, extends that chain.  If the software detects a scenario where they are not mining the longest chain they would stop mining (or switch pools to a pool that is extending the longest block) and warn the user.   

This type of "verify I am on the longest chain" can be done without a change in the blockheader but it would require pools to send the full blockheader to miners.  For efficiency reasons they don't and that is unlikely to change.  Also if a pool doesn't it can't be forced to do so and most miners wouldn't see a lack of the verifying information as sufficient reason to not use the pool.  Changing the blockheader would ensure the necessary information is always sent to the miner.

Even though ghash is now at 30% I don't think this question is answered at all and this still very much an issue. This is because an attacker could could potentially gain control over more than one pool or more than one pool could collaborate to execute an attack. I would really like to see the pools such that not even any given 5 pools total to 51%. Ideally we get rid of centralized pools alltogether or some serious protocol change is made to make it easier to solo mine, or so that 51% attack doesn't work any more alltogether.

I just find the topic odd simply because it would be a bad investment the 51% attack took place, so why would people put their investment at risk by not simply mining elsewhere?  Probably something I've overlooked but hey I don't mine BTC in the first place.

There are probably two reasons why cex/ghash is so popular

One - they are very good at marketing and making their website look good. If you look at their stats pages, their worker pages, the trading platform, the balance history it is very organized looks neat, is very detailed and is appealing to the eye. Now look at the stats page for eligius, the stats are nowhere near as detailed, you don't have as much information and someone who does not have a lot of knowledge about bitcoin may not be able to interpret what everything means. I think eligius is a very good pool and that absolutely nothing is wrong with it, and you generally will make exactly the same with both pools (they use different payment methods PPLNS vs CPPSRB [capped paid per share with recent backpay]).

Two - They offer merged mining with several altcoins. This will increase profitability by a little bit (the small amounts matter in mining) and offer a platform to sell two of the altcoins that are merged mined.

For cex site, people pay to get hashing power. Cex uses a whole bunch of miners for the ghash pool. The fees are high but I guess people like trading the hashing power to the next person (for a quick profit).

Thanks for this and for the links that explain how 51% isn't automagically a disaster.

Does anyone have a link or explanation for how Ghash.io became so popular?  What is so great about them?  I've been to their site and don't see anything super special or am too ignorant to recognize their awesomeness.  Their rise, regardless of what they do with it, seems pretty remarkable.

Thanks!

Does anyone have data on double spends that go back further than this?: https://blockchain.info/double-spends

Not going to disagree with that sentiment
The market will probably react immediately once someone abuses that power
Anyways been hearing of ways to fix this problem but this is a puzzle for developers and the networks to find a solution to

Looks like the community is answering this 51% attack rather quickly and violently. 24-hour block discovery is down to 36% for GHash.
Things that have happened since the attack:

+ GHash was DDOS for a moment last night.
+ BitFury Pulls 1PH/s of Mining Power from Ghash.io
+ P2P reaches 1% network hashrate. Discovers blocks (see 4-day timechart)

In spite of the temporary relief, I suspect the GHash reduction is superficial, and will guarantee to surface again, like herpes--or China.

+1
i dont think they would. makes no sense to mess up all for a few double spends b4 we catch on

Ghash is very good at marketing and has likely attracted a lot of "hands off" type investors that want to look at the pretty numbers. Ghash also offers merged mining of other SHA coins, increasing the EV of mining with them a little bit.

Even if ghash were to control 51%+ of the network does not necessarily mean they will carry out an attack.