CoinWallet says Bitcoin stress test in September will create 30-day backlog

[ArticMine]

...
The fees only rise temporarily so their isn't an economic incentive for miners to attack the blockchain with spam

Cryddit has already identified a scenario where there is an economic incentive.  I suspect that if proper simulations are run one can show this scenario is just the tip of the iceberg.

[1715685992]

1715685992

[1715685992]

1715685992

[gentlemand]

If you need to resort to fear mongering like political hacks to prove your point then you already lost the debate.

If it can be done, chances are some misanthrope is going to do it. I agree that it's shit but BTC isn't going to go anywhere if it can be torpedoed in such a piffling manner.

[Panthers52]

Does anyone know if having larger maximum block sizes would make these kinds of stress tests/attacks less effective?

[ArticMine]

Does anyone know if having larger maximum block sizes would make these kinds of stress tests/attacks less effective?

Yes because the attacks need a certain minimum percentage of the block to be filled with "real" transactions. Take Cryddit's example: In order for the attack to work the percentage of "real" transactions in the block has to be 50% or higher. Now, if one increases the size of the block by say a factor of eight while keeping the real transactions approximately constant then the economics of the attack will fall apart, since 50% has now become 6.25%.  

The way to answer this question is to run numerical simulations of this attack, for various "real" transaction block fill ratios and percentage hashrate controlled by the attacker.

[Panthers52]

Does anyone know if having larger maximum block sizes would make these kinds of stress tests/attacks less effective?

Yes because the attacks need a certain minimum percentage of the block to be filled with "real" transactions. Take Cryddit's example: In order for the attack to work the percentage of "real" transactions in the block has to be 50% or higher. Now, if one increases the size of the block by say a factor of eight while keeping the real transactions approximately constant then the economics of the attack will fall apart, since 50% has now become 6.25%.  

The way to answer this question is to run numerical simulations of this attack, for various "real" transaction block fill ratios and percentage hashrate controlled by the attacker.

Hmmm, it is almost as if it might be a good idea to raise the maximum block size.

Does anyone know why the core developers who work for blockstream are so opposed to raising the maximum block size? They don't want to profit from smaller blocks do they?

[Jeremycoin]

Is this a serious problem, 30 days is a long time tho.
But however we have to keep running

[Carlton Banks]

Does anyone know if having larger maximum block sizes would make these kinds of stress tests/attacks less effective?

Yes because the attacks need a certain minimum percentage of the block to be filled with "real" transactions. Take Cryddit's example: In order for the attack to work the percentage of "real" transactions in the block has to be 50% or higher. Now, if one increases the size of the block by say a factor of eight while keeping the real transactions approximately constant then the economics of the attack will fall apart, since 50% has now become 6.25%.  

The way to answer this question is to run numerical simulations of this attack, for various "real" transaction block fill ratios and percentage hashrate controlled by the attacker.

Hmmm, it is almost as if it might be a good idea to raise the maximum block size.

Does anyone know why the core developers who work for blockstream are so opposed to raising the maximum block size? They don't want to profit from smaller blocks do they?

They don't. Side chains require bigger blocks, hence the "chain" part. You can't store the information in thin air.

Does anyone know why people keep repeating this untrue statement?

[Xialla]

Lets call this "stress test" what it is, an attack on Bitcoin to promote an agenda.

really? I would say, it is testing beyond borders, which is somebody doing for free and without charging anybody. literally dozens times bigger tests can be done by any gov. or bitcoin competitors and without single announcement, anytime and indefinitely.

I don't want XT and I want dev. core team to react..sadly this seems like only way how to speed up decisions and change priorities.

[frenulum]

How will we know when it's happening? Will smaller transactions slow to a halt?

[ArticMine]

Does anyone know if having larger maximum block sizes would make these kinds of stress tests/attacks less effective?

Yes because the attacks need a certain minimum percentage of the block to be filled with "real" transactions. Take Cryddit's example: In order for the attack to work the percentage of "real" transactions in the block has to be 50% or higher. Now, if one increases the size of the block by say a factor of eight while keeping the real transactions approximately constant then the economics of the attack will fall apart, since 50% has now become 6.25%.  

The way to answer this question is to run numerical simulations of this attack, for various "real" transaction block fill ratios and percentage hashrate controlled by the attacker.

Hmmm, it is almost as if it might be a good idea to raise the maximum block size.

Does anyone know why the core developers who work for blockstream are so opposed to raising the maximum block size? They don't want to profit from smaller blocks do they?

They don't. Side chains require bigger blocks, hence the "chain" part. You can't store the information in thin air.

Does anyone know why people keep repeating this untrue statement?

The choice of the 20th percentile over the median in BIP 100 because of a Blockstream recommendation is a perfect example. https://bitcointalk.org/index.php?topic=1164429.0 This turns what could be a workable proposal into one that is way more vulnerable to the attack mentioned above.

Edit: It also creates a situation that is much worse in that the loosing side could have well over 50% of the hashrate creating the perfect storm for a hostile fork.

[CohibAA]

How will we know when it's happening? Will smaller transactions slow to a halt?

You have to watch node stats.  During the last "test" the unconfirmed transactions count was somewhere around 30,000, iirc.  This can vary widely from one node to another, though, as node operators can decide which transactions are included in their mempool, to a degree.

[CoinBateman]

Should also expect some price volatility along with this.

[ph.amracyshop]

This is interesting but it feels like a sabotage. 

[Q7]

I don't understand what are they trying to prove by conducting the test? To convince people to accept the fact that the larger block size is indeed needed? I thought that was shown clearly the last time. Apart from that, it doesn't serve any purpose and will just end up a plain act of sabotage that brings nothing but inconveniences to everyone. 

[Envrin]

If CoinWallet goes ahead with this, then I'm starting a "boycott CoinWallet" petition.

What cunts.  We have a testnet for a reason.  If there's one thing in life you don't do, it's screw with other people's money.  They may not realize this, but there's many people's livelihoods that currently rely on the Bitcoin financial network being online, including my own.

[johnyj]

I guess no stress test will happen this time because it did not cause the panic it wanted last time, and this time every nodes is more prepared to filter out the spam transactions

[Carlton Banks]

If CoinWallet goes ahead with this, then I'm starting a "boycott CoinWallet" petition.

What cunts.  We have a testnet for a reason.  If there's one thing in life you don't do, it's screw with other people's money.  They may not realize this, but there's many people's livelihoods that currently rely on the Bitcoin financial network being online, including my own.

It's fairly certain that CoinWallet is simply a front for what this really is: an attack. If you go to CoinWallet.eu, the website looks functional, but is not in reality. A few credible bitcointalkers signed up to investigate whether they are a real service serving customers or not, and it appears not.

[biggus dickus]

If CoinWallet goes ahead with this, then I'm starting a "boycott CoinWallet" petition.

What cunts.  We have a testnet for a reason.  If there's one thing in life you don't do, it's screw with other people's money.  They may not realize this, but there's many people's livelihoods that currently rely on the Bitcoin financial network being online, including my own.

It's fairly certain that CoinWallet is simply a front for what this really is: an attack. If you go to CoinWallet.eu, the website looks functional, but is not in reality. A few credible bitcointalkers signed up to investigate whether they are a real service serving customers or not, and it appears not.

I looked at their site yesterday and there was nothing about their stress tests on it. They have a FAQ and I expected to find something about the stress testing there because its had such a big effect on the transaction speed. If they had a legit reason for their stress tests I think it would be published on their site. As it stands I consider it an attack rather than a test.

[BornBlazed]

BTC will shit the bed  , buy it  for super cheap after the weak hands sell.

[mallard]

CoinWallet are crypto terrorists.
Have they said what transaction fee they will be using for the next spam attack?