[FACT] There is NO way to cheat pools with fake shares

[Spoetnik]

I had a guy try and convince me off past pool cheating on here recently and he is convinced
he has proof (non technical proof) that its happened lots.. this is just not true.
and i really think i need to say this and make it loud and clear and dispell the myth.

what pushed me over the edge was the Dwarf Pool guy advertising his pool with all kinds of clams in a bullet point format..
and one of the points he made was that his new JPC pool was Fake Share proof.
This is dishonest bullshit !

I went through this last year on the Secure coin topic (no not the SRC Poloniex added a different one LOL)
And various miner coders coin devs and pool coders ALL unified backed me up on this..
It just isn't possible..

and hey if it's possible then i expect a source code sample with proof of concept.. that is it.. that is the only proof i will accept.

and i will take it a step further and for security purposes will any Pool operators out there
be willing to pitch in on a bounty if someone can proof a legit fake share miner hack exists ?
this seems to me like something pool operators should be concerned about.

[1714987993]

1714987993

[1714987993]

1714987993

[1714987993]

1714987993

[Cloudpost]

Normally, pool checkes shares (are they valud or not). When some "new" algos were released, some guys modded pool code so it will support these "new" algos. Pool code wasn't modded properly and didn't check the shares (are they valid or not), so fake shares attacks happened to early MYR pools, if I'm not mistaken. I also remember HIRO x11 early pool that was exploited with fake shares... I guess it was a poor coding problem... Correct me if I'm wrong.

[Spoetnik]

miner source code wanted..

if you wanted to create a pool that accepts Oreo cookies, you could.. that does not mean it's possible to cheat pools lol

that is like saying he it's easy to break into any banks vault all you gotta do is convince the manger to leave the door open when he goes home lol

[r0ach]

I can confirm Myriad did in fact have a pandemic problem of fake shares for the Skein algorithm.

[Spoetnik]

plenty of claims but.....

[Spoetnik]

Let me put this into perspective a bit for my view on this..

Last year i spent a lot of months looking into it on Securecoin and Quark and i also was accused of making a cheating miner too.
Long story short, after all the vast amount of code experiments & recompiling the miner(s) and testing on pool(s) i thought i found something. *at first.
But it turned out that i didn't find any kind of an exploit but a possible slight advantage at best and that is..

A miner coded to mine for easier than normal shares combined with a pool setup in a non standard way to accept easier shares is all that can be done.
This requires the pool owner / coder to set this up or it don't work period .
*with the right kind of pool too.. the exploit would not help finding blocks but would be for getting more shares submitted over other miners..

so if any of you are claiming this happened i highly suggest avoiding the pool operator who is making these claims
because he's the one cheating you NOT the miners..

Let me pose this scenario.. say i started a pool and i set the difficulty far below normal and then i coded my own custom miners
so that they accept lower difficulty shares this would give users of my miner an advantage over regular miners mining on the proper difficulty.
I found this out by accident basically by stumbling on to a pool that was doing this before.

and i stand by what i said earlier 100% !

OBVIOUSLY if you created / modified a pool to cheat then the pool is going to get cheated
and THAT is not what i am discussing.

Bottom line is i have never seen one real coder say it happens.. because it's not possible code wise period.
the only people saying this are NOT coders and sorry but you don't know what your talking about.
and if a pool owner says it true ? stay away from his pools

edit:
It's worth mentioning i have been coding and cracking for almost 15 years too.
For example i regularly crack or keygen a bunch of apps and i have done so on many platforms from Motorola Modems to Linux to various MIPS to Windows etc.
I reverse engineered the aglo (SHA256+custom) and i coded a keygen for my Firewall and i re-keygen it every time the dev tries to break my keygen too.
Same with some other apps.. no matter what the devs do over the years it takes me a few minutes to re-crack their app's LOL
So i know reverse engineering and i have the experience and mind set of a cracker.. looking for exploits is my hobby
and.... i looked at the code a LOT on all these miners

and hey, maybe it is possible but i don't think there is a chance in hell anyone will come forward with a miner code sample..
because i personally think it's virtually impossible.

[tylerderden]

i read something about this on github, it sounds like it was fixed already but did go on with one of the darkcoin pools for a bit. I'm not a coder but if you really want to know there is some stuff out there about it.

[memecoin]

A miner coded to mine for easier than normal shares combined with a pool setup in a non standard way to accept easier shares is all that can be done.
This requires the pool owner / coder to set this up or it don't work period .
*with the right kind of pool too.. the exploit would not help finding blocks but would be for getting more shares submitted over other miners..

so if any of you are claiming this happened i highly suggest avoiding the pool operator who is making these claims
because he's the one cheating you NOT the miners..

That last part is not necessarily true.  There was a vulnerability in Stratum sometime last year that allowed that very exploit, and you can't tell me that every single pool owner that ran that version of Stratum was cheating the miners.  Here is a link: https://bitcointalk.org/index.php?topic=220641.0.  A simple code oversight allowed some miners an unfair advantage.  Although technically you are right in that the extremely low difficulty shares were still valid shares, the likelihood that each submitted share would lead to a block being solved was drastically reduced.  I know you said this sort of exploit is not what you were discussing, but I think it depends on your definition of valid, honestly. 

Why isn't this in the Mining sub-forum?