Bitcoin Forum

How do you know it's not just relaying someone's blocks ?

they wrote a paper about Bitcoin some days ago

Bitcoin is a decentralized payment system that is
based on Proof-of-Work. Bitcoin is currently gaining
popularity as a digital currency; several businesses
are starting to accept Bitcoin transactions.
An example case of the growing use of Bitcoin was recently
reported in the media; here, Bitcoins were used as a
form of fast payment in a local fast-food restaurant.

In this paper, we analyze the security of using
Bitcoin for fast payments, where the time between
the exchange of currency and goods is short (i.e.,
in the order of few seconds). We focus on doublespending attacks on fast payments and demonstrate
that these attacks can be mounted at low cost on
currently deployed versions of Bitcoin. We further
show that the measures recommended by Bitcoin developers for the use of Bitcoin in fast transactions are
not always effective in resisting double-spending; we
show that if those recommendations are integrated
in future Bitcoin implementations, double-spending
attacks on Bitcoin will still be possible. Finally, we
leverage on our findings and propose a lightweight
countermeasure that enables the detection of doublespending attacks in fast transactions.

Where do i find this paper?

Are the bitcoin core developers in communication with ETH Zürich? (Capkun, Androulaki, Karami.)

http://www.syssec.ethz.ch/research/Bitcoin

This is not a bad thing. That is a legitimate organization, they are not doing it with hostile intent. Very interesting though!

it doesn't matter if the party is "good" or "bad". every party has to be rejected if they have more than 50% of the network if it was the case.

Swiss... they have something to defend :(

I don't understand what in the translated paper. Do they have great hashing power or discovered a flaw in the bitcoin protocol?

... Until now, double-spending attacks on fast pay-
ments in Bitcoin or mechanisms for their prevention
have not been studied. In this work, we analyze
double spending attacks in detail and we demon-
strate that double-spending attacks can be mounted
on currently deployed version of Bitcoin, when used
in fast payments. We further show that the measures
recommended by Bitcoin developers for fast trans-
actions are not always effective in resisting double-
spending; we argue that if those recommendations
are followed, double-spending attacks on Bitcoin
are still possible. Finally, we propose a lightweight
countermeasure to detect double-spending attacks in
fast transactions.

More specifically, our contributions in this paper
can be summarized as follows:

We measure and analyze the time required to con-
firm transactions in Bitcoin. Our analysis shows
that transaction confirmation in Bitcoin can be
modeled with a shifted geometric distribution and
that, although the average time to confirm transac-
tions is almost 10 minutes, its standard deviation is
approximately 15 minutes. We argue that this hin-
ders the reliance of transaction confirmation when
dealing with fast payment scenarios.

We thoroughly analyze the conditions for perform-
ing successful double-spending attacks against fast
payments in Bitcoin. We then present the first
comprehensive double-spending measurements in
Bitcoin. Our experiments were conducted us-
ing modified Bitcoin clients running on a hand-
ful of hosts located around the globe. Our results
demonstrate the feasibility and easy realization of
double-spending attacks in current Bitcoin client
implementations.

We explore and evaluate empirically a number of
solutions for preventing double-spending attacks
against fast payments in Bitcoin. We show that the
recommendations of Bitcoin developers on how to
counter double-spending are not always effective.
Leveraging on our results, we propose a lightweight
countermeasure that enables the secure verification
of fast payments. ...

I dont know whether they mine with full power yet, but its not 51%, its 8.547% actually.


Number of Blocks relayed by 82.130.102.160: 20
First block relayed at Blockheight: 200691
Current Blockheight: 200925
dBlockheight=200925-200691=234 -> they have competed for 234 blocks yet

20/234*100%= 8.547%

They have currently relayed 8.547% of all blocks they have competed for.

Swiss... they have something to defend

This is not a bad thing. That is a legitimate organization, they are not doing it with hostile intent. Very interesting though!

what is the timeframe for an 51% attack. 6 blocks?

what is the timeframe for an 51% attack. 6 blocks?

IIRC (http://bitcoin.org/bitcoin.pdf), maintaining a supermajority attack gets exponentially hard over n-blocks.

Hence: "Double-Spending Attacks on Fast Payments in Bitcoin"

Unless you have 51% of the whole network's hashing power right?

Jeesus......that is a lot of power to just "flip the switch" on.

I would love to know what they are running. 

Jeesus......that is a lot of power to just "flip the switch" on.

I would love to know what they are running. 

Maybe they have tons of fpga unused? They talked about cheap double spending. I don't know how they accounted for 10% of the network power in a "cheap" way.

Maybe they have access to supercomputers we can only dream off.

They are 500Km away, I should go there and investigate :D

give em a call, fake a Swiss accent and they won't even know you are from the internet ;)

, fake a Swiss accent

did you ever tried to fake a swiss accent? very impossible! they have different accent in every canton and every village. bad idea!  ;D

did you ever tried to fake a swiss accent? very impossible! they have different accent in every canton and every village. bad idea!  ;D

In that case just choke up as many croutons from the back of your throat as possible and say you come from the most backward village of them all!

I think it's quite bad:

ok, lol but this is Zurich, north of Swiss, so they speak german...

What's this: Peer 82.130.xx.xx connected to me 3 times, sending constant pings?  (Read 512 times) (https://bitcointalk.org/index.php?topic=110390)

bad idea too! I dont want to get a bullet in my head!  ::) they get their army gun to home if they leave the regular army service but without cartridges. some years ago they had to give back the cartridges from home. but I do not trust in that!  ;)

Well hell, look at the files on port 80, bittheif???.  :(

http://82.130.102.160/

No way..wtf ?!?

oh, btw... don't execute the files... 'BitThief' is not something you want on your computer ;)

It seems like Bitthief is not related to Bitcoins:

http://bitthief.ethz.ch/

Reading their paper, it seems that their goal is to send coins to one address, but after that confirm a spending of the same coins to their own address by their miners.

One should note that the paper has already been submitted to, accepted by,
and is due to be presented on the ACM CCS 2012 conference in about two weeks.

So: Whatever they are doing now, it is not related to that specific paper.

So can I defend myself? Can the LOIC weapon be turned on their machines?

So let me get this straight? They're not trying a sustained 51% attack...that would take too much power. Instead they're running at about 8-9% of the network, and hoping to get 6 blocks in a row to reverse a transaction (double spending)?

So let me get this straight? They're not trying a sustained 51% attack...that would take too much power. Instead they're running at about 8-9% of the network, and hoping to get 6 blocks in a row to reverse a transaction (double spending)?

http://bgp.he.net/ip/82.130.102.160

ETH Zürich have been experimenting with fast double spend attacks, and is currently (as of today) bringing massive mining power to bear on the main network.

Their public http server has a file named BitThief.exe on it, and they're spamming ping messages. (Denial of Service?) Did i miss anything?

CSCS' main function is a so-called National User Lab. It is open to all Swiss researchers and their collaborators, who can get free access to CSCS' supercomputers in a competitive scientific evaluation process. In addition, the centre operates dedicated computing facilities for specific research projects and national mandates, e.g. weather forecasting. It is the national competence centre for high-performance computing and serves as a technology platform for Swiss research in computational science. [3]. CSCS is an autonomous unit of the Swiss Federal Institute of Technology in Zurich (ETH Zurich) and closely collaborates with the local University of Lugano (USI).

Bitcoin "FLOPS" computation on bitcoinwatch

bitcoinwatch.com/ calculates PFLOPS of bitcoin network as: take number of Hashes/second (Terahashes/s of SHA256) and multiply by 12700 to get a "Single-precision FLOPS estimate". One hash calculation is considered as 6350 32-bit integer operations, and each integer operation is considered equal to two single-precision flops. Source of constants is: http://forum.bitcoin.org/index.php?topic=4689.0 (with reference to bincoinwatch's admin). Actual bitcoin mining contains no (or almost no) floating-point calculations.

So they have around 9% network hashing power...

How massive of a supercomputer would they have to use to get 2 TH/s?  And does anyone know if they HAVE a supercomputer at that university?

https://bitcointalk.org/index.php?topic=112889

The new Mira Supercomputer does 8 petaflops. I did some quick napkin calculations, and I estimate that gets ~600GH/s.

It seems like Bitthief is not related to Bitcoins:

http://www.disco.ethz.ch/members/cdecker.html

"Tampering with Distributed Hash Tables"

I agree.  But do they have access to said supercomputer?  And even if they did, it's only 1/4 of the estimated hashing power it is implied that they somehow have access to.

this is why we need asic mining rigs, cheap asic mining rigs.

actually, this is really annoying now, since there are 9 of them

206.12.16.155
129.74.74.20
128.6.192.156
129.130.252.140
82.130.102.160 (on about 30 or 40 different ports)
137.99.11.86
130.253.21.123
147.102.3.117
129.130.252.141

whenever one connects, i get flooded with the ping fail

http://bgp.he.net/ip/206.12.16.155     -->BCnet
http://bgp.he.net/ip/129.74.74.20       -->University of Notre Dame
http://bgp.he.net/ip/128.6.192.156     -->Rutgers University
http://bgp.he.net/ip/129.130.252.140 -->Kansas State University
http://bgp.he.net/ip/82.130.102.160   -->Swiss Federal Institute of Technology Zurich
http://bgp.he.net/ip/137.99.11.86       -->University of Connecticut
http://bgp.he.net/ip/130.253.21.123   -->University of Denver
http://bgp.he.net/ip/147.102.3.117     -->National Technical University of Athens
http://bgp.he.net/ip/129.130.252.141 -->Kansas State University

http://bgp.he.net/ip/206.12.16.155     -->BCnet
http://bgp.he.net/ip/129.74.74.20       -->University of Notre Dame
http://bgp.he.net/ip/128.6.192.156     -->Rutgers University
http://bgp.he.net/ip/129.130.252.140 -->Kansas State University
http://bgp.he.net/ip/82.130.102.160   -->Swiss Federal Institute of Technology Zurich
http://bgp.he.net/ip/137.99.11.86       -->University of Connecticut
http://bgp.he.net/ip/130.253.21.123   -->University of Denver
http://bgp.he.net/ip/147.102.3.117     -->National Technical University of Athens
http://bgp.he.net/ip/129.130.252.141 -->Kansas State University
...

All universities. Hmmm. Even the BCnet addres is registered to "Simon Fraser University".
Could these entities be working together to attain high hash rates?

So they have around 9% network hashing power...

How massive of a supercomputer would they have to use to get 2 TH/s?  And does anyone know if they HAVE a supercomputer at that university?

Instead they're running at about 8-9% of the network, and hoping to get 6 blocks in a row to reverse a transaction (double spending)?

I don't know what I'm talking about, but: could there be a connection between BitThief and a massive mining botnet?

So they have around 9% network hashing power...

How massive of a supercomputer would they have to use to get 2 TH/s?  And does anyone know if they HAVE a supercomputer at that university?

The BitThief exploit allows someone to 'leech' from the BitTorrent network without actually contributing resources yourself.

Is it possible that they are using a similar exploit to leach block discoveries from the bitcoin network without actually doing any mining on their own?

That could explain how they are getting so much hashpower (i.e. they don't really have it).

The BitThief exploit allows someone to 'leech' from the BitTorrent network without actually contributing resources yourself.

Is it possible that they are using a similar exploit to leach block discoveries from the bitcoin network without actually doing any mining on their own?

That could explain how they are getting so much hashpower (i.e. they don't really have it).

How do you know it's not just mining some blocks ?

The BitThief exploit allows someone to 'leech' from the BitTorrent network without actually contributing resources yourself.

Is it possible that they are using a similar exploit to leach block discoveries from the bitcoin network without actually doing any mining on their own?

That could explain how they are getting so much hashpower (i.e. they don't really have it).

Well,
1) I suspect that you still don't believe us about how the block relaying and blockchaininfo's IP detection works.
2) Some of those blocks were mined by Deepbit and other pools.

So I think that this is just a well-connected node.

On the question of is it possible, yes it is theoretically possible to see a block submitted to the network and then publish that block quickly from a highly connected node and it will look like it came from you. So maybe that's what they are doing here.

To be clear, though, the coins still go to the address in the generation transaction, so they can't steal your coins. It just makes it look like the block came from a different IP.

I still don't think that applies here.

On the question of is it possible, yes it is theoretically possible to see a block submitted to the network and then publish that block quickly from a highly connected node and it will look like it came from you. So maybe that's what they are doing here.

To be clear, though, the coins still go to the address in the generation transaction, so they can't steal your coins. It just makes it look like the block came from a different IP.

I still don't think that applies here.

On the question of is it possible, yes it is theoretically possible to see a block submitted to the network and then publish that block quickly from a highly connected node and it will look like it came from you. So maybe that's what they are doing here.

Almost certainly. I've decompiled the Python code in snoopy.tar.bz2 from that web server and that appears to be exactly what it's doing.

You'd think they would make more bitcoin simply by mining.

It's an interesting experiment, but IMHO they must be stopped at all costs.

Now, if a person orders twenty of the 1TH/s BFL rigs, could they not do the same thing?  

So out of the list of blocks relayed by that IP, how many were generated by known mining pools?