Topic: How to attack Bitcoin mining? (Read 2738 times)

You can't "keep the good hashes". Either you send them to the pool, or you throw them away. If you throw them away, you get nothing. Yes, you can do that, but it's a waste for you.

More generally, what if I participate in a pool and just keep the good hashes for myself? I guess this would be strictly more profitable for me than just participating in a pool. But if every one does the same, pools should just never find a block. Right? Then: Why do pools exist in the first place?

Yeah well I alone could think of ways to close this hole so if there is a hole it is just the lazy devs or the stubborn bitcoiners that havent implemented it yet.

Maybe what you thought was incorrect not the information presented.  Google "block withholding attack".  It has been know and discussed pretty much as long as there has been pooled mining.


This is incorrect.  A miner instantly knows if he solved a block or not.  However that solution is tied to that specific attempted block (with predetermined coinbase or "reward" address).  A miner can't profit from this information.  He can either submit it or not submit it.  Not submitting it is the block withholding attack.  This will reduce the gross revenue produced by the pool and will equally affect all miners in the pool (including the attacker) and the pool operator.

There are ways to make the block withholding attack impossible however they require a hard fork in the protocol (a change in the format of the block header).

So in a nutshell you're suggesting that you cannot chain pools together.  If this is accurate than the OP claim of vulnerability has been resolved as False.

Because of the importance of mining is exactly why we should talk about the vulnerabilities associated with it.  It is only then that we can mitigate them and strengthen Bitcoin.

Michael, CEH & OSCP
Security Pen Tester

There is so much Wrong information here that goes against everything I thought. Im pretty sure it works this way:

1. Pool has "complicated problem" to find a new block
2. Pool takes part of this "complicated problem" and sends to client
3. Client solves the new "problem" (Important; Client dont know if their solution works on "complicated problem" as client dont have that)
4. Client sends in the solution
5. Pool checks to see solution works on the given problem
6. If solution works GOOD POOL FOUND NEW BLOCK! if not it still had a chance to find the solution. Either way the client gets payed
7. Pool sees how often you submit a solution to decide what your hashrate approx is.

This also means that it takes some time for the pool to calc your computer speed fx. your computer submits 100 solutions an hour and the pool
calc's this into 10G/Hash (JUST EXAMPLE).


Today I learned: If I took a lot of computers to mine in pool and started to withhold possible solutions, the pool would see it as the hashrate was dropped as you are normally paid per successful solutions send to pool.

Or maybe im just stupid and bitcoin is very easy to kill.

When you sit with a solution you have no idea if pool can use it or not (Neither does pool yet), you know that there is a much bigger chance of this being the solution the pool can use than if you took a random answer. You only get paid when you submit it to the pool.

For children:


Imagine that your parents had this rare football.
They just didnt know where this rare football was.
So they paid you to bring them round objects.
Everytime you bought them a round object you get paid a little.
The parents could not tell you that it is the football they are looking for,
if they told you and you found it, you would keep it too your self.
so they tell you to bring them round objects, now you dont know if your parents actually profit when you bring them a round object.

You bring them a tennisball
you bring them a tire
you bring them a melon

It might take long but in time and with enough people someone will bring the football, they will give it too the parents because it has no value to them, they dont know it is rare, all they know is that it is a round object.

The parents sell the football and use the money the pay the children for each round object they found.

Really hard to put it more simple sorry.

Now there are different systems.

Named PPS, PPLNS and so on.

In some systems the children are payed per thing they submit that qualify.

In others no one is payed until the football is found and then they are payed equal to effort.

sometimes the children might get lucky and find the football after just 4 things have been submitted and the 4 people all share the reward. (Thus getting payed much to be lucky)

Other times there will go along time without anyone finding what item is actually worth anything, in this case the people will get payed litle when you consider the number of possible things they have submitted.



ITS all about who takes Variance that the pool or the miners.

Overtime thou it shouldn't matter.

There are some modified versions of mining software (I believe it was cgminer) that can be set so that found blocks are not submitted to the pool. There was a miner in China earlier this year that was mining on the eligius pool, withheld what should have been several blocks (they would have been expected to find several blocks verses what they actually found); it ended up costing the pool several hundred BTC. There was also likely a similar attack on BTC guild that lasted several months before that as their luck was way below what it should have been.

It is hard to say. The community would be closely watching such a pool and would likely take countermeasures the moment that pool does any underhanded stuff. This may include custom clients that would completely prevent blocks solved by that pool from propagating through the network, if such blocks could be identified easily.

Your absolutely right of coarse, the miner could 'guess' what shares may solve a block without too much trouble.

Neil

P.S. I blame a head cold, I took some pseudoephedrine about an hour or two ago and I just realized I am not thinking all that straight.... I think I will stop doing test restores of VM's as well before I do some damage .

Maybe it's also possible to mine for two pools at the same time?

Would like to have a wiki article on attack vectors on mining operators where these questions are answered.


I thought so, but how does that actually work?

You may be right about that, I'm not that firm in the theory. But in order to mine at other pools or pools at the same time, they'd need you to find a solution for the same Bitcoin address as the coinbase (not the company) address.

interesting stuff, waiting to read more!

Hi Neil, thanks for the input. This is the latest response I've got from BM on the forum  https://bitsharestalk.org/index.php?topic=7003.msg94085#msg94085

in short no..

2) if your pools is charging different miners negative fee, then your not making income
3) if your attempts to on the other pool your equipment was on, did actually succeed in preventing them making blocks.. your simply stopping yourself getting paid too...
4) every miner is not submitting winning hashes thousands of times a second.. all hashes are losers until (average tenth minute) one single hash is the winner. thus your not doing any damage,
5) as described in (3)(4) you cant subsidize YOUR pool if you are not making any funds from the other pools..

6) a 3% hashrate farm wont make a blind bit of difference.. all your doing is wasting your own funds and costing yourself alot of wasted time, for nothing.

but goodluck trying it.

Wrong.
https://bitcointalk.org/?topic=441465.msg7282674#msg7282674

When mining you don't know if you have the winning hash until after you submit it (sometimes not even then).

Neil

First post didn't make sense to me, this post makes way more sense as to what you're saying.
Interesting idea, scorchingsun says you could find those withholding finds, but would this be efficient?

I COULD be done easier than that^ just still not really.

1) and what next?
2) ghash already uses merged mining



To win in a modern competition in economics you should not drop bombs and have the largest military forces.
The only thing you should do is reducing the revenue of competitors.
This can be done by printing your currency and spread it to a pool miners native citiziens (negative fee, budget deficit)
Analogy: bitcoin=gold, namecoin=dollar, community=world, your miners=us citizens, 1492...2014 and so on

No, I think the result is either to do a 51% attack on Bitcoin or build a big mining pool like Ghash, for profit, from a small start using this tactic. 

Previously I'd been lead to believe someone would have to get control of 51% of the hashing power which is very expensive.
The above example makes it seem you could achieve the same result with 3/4% of the hashing power in a fairly short period of time.
Which is why I'd like to debunk it.

(I don't get the country analogy, sounds interesting, but it went over my head I'm afraid.)

Yes. It is possible.
But what it the long-time expected result?
Do you want to gain all 21m bitcoins?

OK, I suggest another idea:
1) Lets create a fork with 100% premine.
2) Profit!
Voila! You have all hash power and all coins without any initial spendings.
----
Update: what country are you from? compare your steps 1...5 with the history of your country

I asked about the 'smart heuristics' on the BitShares forum scorchingsun,



If someone like BCX, who seems to know his stuff, could give me a simple answer why this attack wouldn't work I would greatly appreciate it. Thanks!

You are so nice.

I don't want to attack it! Half my alt-coin investment is in Bitcoin! However if someone says it may be possible to attack Bitcoin with 3/4% of the hashing power, I'd like to understand why it's wrong.

To me it seemed like, (To keep the numbers simple)...

----
- If there were a $1 Billion in new coins made a year then a pool with 30% hash power could expect to earn $300 million in revenue.

- However Bitcoin margins are tight. How tight? I don't know but I doubt they're making more than $15-30 million profit.

- If 3% secret hash power (or 10% of the pool) was not submitting winning hashes, it would take $30 million of revenue from them, making them unprofitable.

- As you're part of their pool though you would still receive a 97% payout,  $27 million. So you may be at a loss too but only a $1 million or two and for that you get to make the main mining pool unprofitable.

- Hashers would leave the main unprofitable pool for the one a bad intentioned person controls, which could then easily accumulate 51% hashing power.
----


That sounds like it may be an answer, why this can't work thanks!

That strategy isn't that bad, actually! No wait, it is! Your whole plan fails due to the fact that you only mine looking for a solution for the pool's address! Withholding the winning shares will hurt the pool, but you can't subsidize anything with that.
Oh and withholding shares can be be detected with some smart heuristics.

Do you even have the slightest idea about how important Bitcoin Mining!!! and you want to attack it!!! Why!!!

That's something like: how to get rich
1) Use time machine to go back in time
2) Buy BTC
3) sell in 2020.
In theory it works, but in reality not really.
How about the person who posted it in the forum tries it and see what happens.

I highly suggest you talk to someone with a clue, seriously.


~BCX~

I'm not that technical. But I've seen this approach being discussed on the BitShares forum as perhaps a weakness of POW.
Would this work, is this a threat at all?