Bitcoin Forum
May 02, 2024, 09:50:13 PM *
News: Latest Bitcoin Core release: 27.0 [Torrent]
 
   Home   Help Search Login Register More  
Pages: « 1 2 3 4 5 6 7 8 9 10 11 12 13 [14] 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 »
  Print  
Author Topic: Are we stress testing again?  (Read 33162 times)
JorgeStolfi
Hero Member
*****
Offline Offline

Activity: 910
Merit: 1003



View Profile
July 12, 2015, 02:22:23 AM
 #261

The core client wallet lets you see estimated fees to get your transaction confirmed within the next Nth block, and adjust the fees above or below the threshold as desired.  The devs of those wallet apps need to adjust their coding, not the core devs, imo.

I tried to explain to the devs that such an algorithm cannot exist, for the same reason that there is no algorithm that will choose the right price to bid at an auction: the desired answer depends entirely on data that does not exist yet, and must be higher than the value that others will choose, using the same or better algorithms -- in other words, the answer must be X such that X = X + 1.  

I suppose that I should have added the "Grin" when I wrote:

clients who have read the source code of the core implementation (and are aware of the modifications and parameter choices made by the major miners and relay nodes) can easily compute the fee that will let their transaction go through in the next N blocks, provided only that they correctly guess what fees will be paid by the 20'000 x N transactions that will be issued by the testers in the next 10 x N minutes, and also by the 800 x N transactions  that will be issued by ordinary clients who want to get their transactions in front of the queue.

That scenario is valid now.  If things go according to the "new devs"' plans, by the end of the year the network would be *naturally* close to saturation.  These are estimated parameters:

  Max block size limit: 1 MB/block
  Average transaction size: 400 bytes/tx
  Average block size with full queues:  500 kB/block = 1250 tx/block
  Network capacity: 2.08 tx/s = 1250 tx/block = 7500 tx/h = 180'000 tx/day
  Daily average traffic: 1.74 tx/s = 1041 tx/block = 6250 tx/h = 150'000 tx/day (83% of capacity)

Now suppose that, when you are about to pay for your airline ticket, there is a "traffic jam" caused by a temporary surge of demand

  Short-term traffic: 3.12 tx/s = 1873 tx/block = 11200 tx/h = 270'000 tx/day  (50% above capacity, 80% above daily average)
  Queue backlog: 30'000 tx = 24 blockfuls

Let's say that you are happy to wait 2 hours = 12 blocks for your first confirmation.  The fee that you will have to pay to get nto the next N = 12 blocks (that will carry only 1250 x N = 21'600 tx) will depend on the fees of (a) those 30'000 tx in the queue, (b) the 1873 x N = 22'476 transactions that are expected to arrive in the next 2 hours, and (c) those transactions, among those 30'000 + 22'476, that will be reissued by their clients in the next 2 hours with fees adjusted (by various algorithms) so as to try to get ahead of your transaction.  

Quote
The results would be a) the core client would keep working just like it always does [ ... ]  Is it good for the immediate health of bitcoin to suffer these types of "attacks"?  Not particularly, because application devs were not prepared in some cases, and can't react fast enough to satisfy their users, which might hurt some levels of bitcoin adoption.  For the most part, though, I think it's a great thing, and hope someone keeps spending their money on this kind of thing.  It drives advancement, thought, and prices.

The only thing that is more resilient than the bitcoin network is the faith of the faithful bitcoiner.  If a bug in the core implementation were to expose the private keys of half the UTXOs generated since 2012, and hackers stole all those coins, the faithful bitcoiner would point at that with pride and say that, once more, the network worked just as intended -- and that was good for bitcoin.  Grin

Note, again, that this was not an hostile attack; it was a test, even if rather harsh and inconsiderate.  A malicious attacker would have tried to jam a substantial part of the legitmate traffic by adjusting his fees as needed to keep those transactions out of the next block, even as they updated their fees.

By the way, here is the interim report of the previous (prematurely aborted) stress test, a couple of weeks ago:
CoinWallet.eu Stress Test Analysis

Academic interest in bitcoin only. Not owner, not trader, very skeptical of its longterm success.
Advertised sites are not endorsed by the Bitcoin Forum. They may be unsafe, untrustworthy, or illegal in your jurisdiction.
1714686613
Hero Member
*
Offline Offline

Posts: 1714686613

View Profile Personal Message (Offline)

Ignore
1714686613
Reply with quote  #2

1714686613
Report to moderator
1714686613
Hero Member
*
Offline Offline

Posts: 1714686613

View Profile Personal Message (Offline)

Ignore
1714686613
Reply with quote  #2

1714686613
Report to moderator
Quickseller
Copper Member
Legendary
*
Offline Offline

Activity: 2870
Merit: 2298


View Profile
July 12, 2015, 03:46:41 AM
Last edit: July 12, 2015, 04:07:03 AM by Quickseller
 #262

It appears that node is a p2pool node, I am not 100% sure how p2pool chooses which transactions to include in their found blocks, however I would think that it would have to do with something one of the nodes decides. If you do not want to include low fee transactions in your found blocks then the above would probably be more optimal settings (it would probably also somewhat increase your mining revenue).


As a former p2pool miner, I can tell you that one of the motivations behind those settings was to reduce orphaned p2pool shares by reducing the size of the mined blocks. For a low-powered node the it was a difference between getting all your shares orphaned and the resulting 0 income vs. getting at least some income. Since p2pool shares are actually valid low-difficulty Bitcoin blocks that are that are solved once every 30 seconds (instead of 10 minutes), any negative impact of increasing the blocksize limit would be likely felt first by the p2pool network before the actual Bitcoin network.
The protocol that P2Pool runs on should either be fixed/changed so this is not an issue or people will stop mining on p2pool nodes.

P2Pool does sound great on paper, however I do not think it is *that* important to keep the network decentralized to that extent. As long as the people who own the miners keep their equipment mining on a reputable pool that does not control an outsized share of the network, acts generally ethically (e.g. does not actively allow people to double spend their unconfirmed transactions), confirms transactions while only taking profit (and ethics) into consideration (not personal beliefs) and that pays out mining revenue quickly then the network should remain sufficiently decentralized. When a particular pool starts to get too high a percentage of the network hashrate, people will start to worry and take their mining power elsewhere. When a pool scams/gets hacked, it's miners will only be marginally affected because only a small amount of mining revenue would be unpaid. When pools start to do shady things, it's miners will take their equipment elsewhere.

Most pools are for-profit enterprises, so they have financial incentives to act in ways that allow them to continue to operate over the long term. If they act in ways that is not good for it's miners then the miners will take their equipment elsewhere because they have a financial incentive for bitcoin to be successful because if it is not (at least over the short to medium term) then their investment in their mining equipment will have diminished (or possibly severely negative) returns.

Even satoshi had predicted that, over time the network would become somewhat centralized, some might even say that he blessed the somewhat centralization of mining (however this would probably be more speculation then fact):

--snip--

I anticipate there will never be more than 100K nodes, probably less.  It will reach an equilibrium where it's not worth it for more nodes to join in.  The rest will be lightweight clients, which could be millions.

At equilibrium size, many nodes will be server farms with one or two network nodes that feed the rest of the farm over a LAN.

-snip-

I do think the obvious solution to the spam attack is to accelerate the date when the protocol is hard-forked to allow 8 (or 20) MB blocks, which would make it much more expensive to conduct such an attack in the future, and would make clearing the backlog a much quicker process

and what would that achieve? People would just go over from spamming tx to delay them to carrying out denial of service attacks by bloating the blockchain to a point where bitcoind quits because drive space is up - I mean seriously, we're at 45g currently, its getting ridiculous to run a full node(and unfortunately at least for me bitcoin core is currently the only option to run off) by now, especially on mobile computers which traditionally have limited space. I submitted a patch to bitcoin git over 2 years ago that would create an option for bitcoin core to truncate the blockchain by simply discarding all transactions that had all outputs spent and confirmed by 120 blocks - it got rejected, but especially these days it would save about 95% of the disk space that is currently used by bitcoins block chain
According to blockchain.info, the blockchain size is currently ~37 GB, over a year, 20 MB blocks would take up roughly 1 TB of disk space, however it is unlikely that blocks would be full right away. I can say that if the block size is not increased then Bitcoin will become something similar to Western Union or the ACH system when you need to spend a lot of money to be able to move your money, which is not what Bitcoin is about.

As of now, we are essentially at the limit as to how many transactions the network can handle based on the max block size. The number of transactions per time might be able to increase a little bit before no additional transactions per time can be processed by the network. To put this another way, the maximum amount of growth in user adoption with a 1 MB block size limit that Bitcoin can achieve is very limited. Total fees per transaction will be able to be increased in order for existing users to get their transactions to confirm quickly, however few additional users will be able to send transactions. If there is a limited number of additional people who can send their bitcoin, then there is a limited number of people who will be interested in buying bitcoin, this means the upside to the price is limited.

The 1 MB block limit can handle roughly 510 1-input, 1-output transactions per block, and assuming a $30 transaction fee (which would be slightly
scarsbergholden
Hero Member
*****
Offline Offline

Activity: 686
Merit: 500



View Profile
July 12, 2015, 05:03:07 AM
 #263

I find it great that the transactions volume is at a new level it shows growth even if is just spamming of stress testing, all network most be tested and taken to its limits to improve and innovate to a better one.

JorgeStolfi
Hero Member
*****
Offline Offline

Activity: 910
Merit: 1003



View Profile
July 12, 2015, 05:51:41 AM
 #264

As long as the people who own the miners keep their equipment mining on a reputable pool that does not control an outsized share of the network, acts generally ethically (e.g. does not actively allow people to double spend their unconfirmed transactions), confirms transactions while only taking profit (and ethics) into consideration (not personal beliefs) and that pays out mining revenue quickly then the network should remain sufficiently decentralized. When a particular pool starts to get too high a percentage of the network hashrate, people will start to worry and take their mining power elsewhere. When a pool scams/gets hacked, it's miners will only be marginally affected because only a small amount of mining revenue would be unpaid. When pools start to do shady things, it's miners will take their equipment elsewhere.

Most pools are for-profit enterprises, so they have financial incentives to act in ways that allow them to continue to operate over the long term. If they act in ways that is not good for it's miners then the miners will take their equipment elsewhere because they have a financial incentive for bitcoin to be successful because if it is not (at least over the short to medium term) then their investment in their mining equipment will have diminished (or possibly severely negative) returns.

You cannot assume that incentives and motivations of miners are different from those of pool operators.  

It seems that many pool members are themselves small or medium-size industrial installations (see BitFury vs GHash).  While those "whale" members may not dominate in numbers, they may dominate in terms of hashpower.  It is possible for someone to follow idealistic principles when that means a monthly loss of 10 $; but that is very unlikely if the loss is 10'000 or 10 million $.  

Miners know that their equipment will be unprofitable in 2-3 years, as chip efficiency (H/J) improves.  So, their goal is to maximize their revenue in that time frame, even if it harms the coin's price in the longer run.

If a pool can get a higher hit rate with the same hashpower, by using shortcuts that imply centralization or degradation of service (like mining empty blocks, serving only the header template to its members, "stealing" block hashes from other pools, etc.), it will pay more to its members; and members will then be reluctant to leave it, even if it reaches 51%.

What happened to GHash when they went over 50% was a charade that only fooled those who wished to be fooled. It is pointless to prevent a single pool to exceed 51%, if no one knows who owns the pools and miners and what contracts they have made among themselves.  

Right now the top 4 Chinese pools have 60% of the hashrate.  Are their members Chinese too? If the Chinese government decided to force those pools to do something "special", like block all transactions out of some address indefinitely, would they comply?  Would their members leave and join non-Chinese pools?  

The top 5 miners -- those 4 plus the Ukranian BitFury -- have 73% of the total.  Wouldn't they form a cartel to do something special that increases their share of the rewards?  Would their members notice it?  Would they really care?

Quote
Even satoshi had predicted that, over time the network would become somewhat centralized, some might even say that he blessed the somewhat centralization of mining (however this would probably be more speculation then fact):

--snip--

I anticipate there will never be more than 100K nodes, probably less.  It will reach an equilibrium where it's not worth it for more nodes to join in.  The rest will be lightweight clients, which could be millions.

At equilibrium size, many nodes will be server farms with one or two network nodes that feed the rest of the farm over a LAN.

Satoshi was very unhappy about the prospects of GPU mining and called for a moratorium on such programs until bitcoin was better established.

His vision seems clear:  he expected mining to remain decentralized, done by the clients themselves, off and on at random times; while the relay nodes (that must be running continuously, and would have a heavier load) would be relatively centralized, as above.  

Instead, mining became an industrial activity and inevitably centralized because of economies of scale and geographic advantages.  Whereas now we are told that it is the relay nodes that must remain decentralized and run by individual volunteers...

By the way, Adam and the other small-blockians have been insisting all along that raising the block size to 8 MB would make full nodes too expensive to run, and that would lead to more centralization.  That is already a dishonest argument because they always replace size limit by size, even after that "misunderstanding" is pointed out to them.

People have asked Adam & co many times for an explicit analysis, with numbers, of the impact of the 8 MB limit, but they never gave one.  Well, yesterday I tried to do such an analysis myself. I don't know whether it is correct, but it showed that the impact would be small, and 8 MB limit might even be better for the full nodes in case of a spam attack.  

But then I learned that what they call "full node" is actually a "selfish node" -- that just downloads and checks the blockchain, but does not relay transactions or blocks to anyone else.  

Bummer.  Yes, those nodes will be more costly if the average block size (not just the limit!) increases, because their network, procesing, and disk loads are proportional to the blockchain size only.  But they don't contribute anything to the security of the network, only to their own.  If 99% of the clients are already SPV mode, what does it matter if those selfish nodes drop from 6000 to 3000? (Or if they are forced to do only partial validation, by random sampling?)

If one consider the full nodes that do transaction propagation and serve queue data to miners and other applications, the impact of an increase in average block size (not just the limit!) is much less, because it will not affect the larger cost of propagating and validating transactions.  Bummer.


Academic interest in bitcoin only. Not owner, not trader, very skeptical of its longterm success.
Quickseller
Copper Member
Legendary
*
Offline Offline

Activity: 2870
Merit: 2298


View Profile
July 12, 2015, 08:31:05 AM
 #265

As long as the people who own the miners keep their equipment mining on a reputable pool that does not control an outsized share of the network, acts generally ethically (e.g. does not actively allow people to double spend their unconfirmed transactions), confirms transactions while only taking profit (and ethics) into consideration (not personal beliefs) and that pays out mining revenue quickly then the network should remain sufficiently decentralized. When a particular pool starts to get too high a percentage of the network hashrate, people will start to worry and take their mining power elsewhere. When a pool scams/gets hacked, it's miners will only be marginally affected because only a small amount of mining revenue would be unpaid. When pools start to do shady things, it's miners will take their equipment elsewhere.

Most pools are for-profit enterprises, so they have financial incentives to act in ways that allow them to continue to operate over the long term. If they act in ways that is not good for it's miners then the miners will take their equipment elsewhere because they have a financial incentive for bitcoin to be successful because if it is not (at least over the short to medium term) then their investment in their mining equipment will have diminished (or possibly severely negative) returns.

You cannot assume that incentives and motivations of miners are different from those of pool operators. 
They will generally be aligned, although if pool operators start to act in ways that is not in the best interest of the miners, then the miners will likely remove their equipment from their pool. 
Miners know that their equipment will be unprofitable in 2-3 years, as chip efficiency (H/J) improves.  So, their goal is to maximize their revenue in that time frame, even if it harms the coin's price in the longer run.
If someone were to actively attack the network with 51% of the hashrate, or otherwise attempt to attack the network, then the short term effect of bitcoin would likely be severe to the negative. If someone were to execute a successful 51% attack, then why would someone want to buy bitcoin? Why would someone not want to sell it until such attack is over? IMO any actual attack on the network that is successful will have very serious short term effects on the price of bitcoin. I don't think simply having 51% of the network alone would have any serious impact other then that miners will likely migrate away from that pool.
If a pool can get a higher hit rate with the same hashpower, by using shortcuts that imply centralization or degradation of service (like mining empty blocks, serving only the header template to its members, "stealing" block hashes from other pools, etc.), it will pay more to its members; and members will then be reluctant to leave it, even if it reaches 51%.
These types of things are not without risks. From the looks of it, the recent issues with the soft fork and f2pool mining invalid blocks cost them >100 BTC, although they claim it was less. The fact that they have not stopped SPV mining would imply that they made in excess of what they lost in additional mining revenue as a result of their SPV mining, probably from winning orphan races and from being able to mine additional blocks they would not otherwise be able to mine.

Pool operators are going to want to maximize the long term revenue of their pool. If they take actions that result in losses to them or their miners then their long term profit potential is not going to be great.
What happened to GHash when they went over 50% was a charade that only fooled those who wished to be fooled. It is pointless to prevent a single pool to exceed 51%, if no one knows who owns the pools and miners and what contracts they have made among themselves.   
Owning 51% of the network hashrate in itself is not a major issue, the problem lies when a single entity starts to actually attack the network after they reach such a percentage. Even if they were to disguise their found blocks as belonging to other entities, it should still be fairly obvious when they are executing some kind of 51% attack.
Right now the top 4 Chinese pools have 60% of the hashrate.  Are their members Chinese too? If the Chinese government decided to force those pools to do something "special", like block all transactions out of some address indefinitely, would they comply?  Would their members leave and join non-Chinese pools? 
Are all the miners who mine on chinese pools chinese? Probably not. If they were then I don't think it would be nontrivial for them to move to a non-chines pool. Could the Chinese government force the Chinese pools to do something like block all transactions to/from the address "1censorship...." however this would likely eventually (more likely quickly) leak and it will quickly be apparent it would be in the miners best interests to stop mining on those pools. This risk is similar regardless of where the pools are located politically.
The top 5 miners -- those 4 plus the Ukranian BitFury -- have 73% of the total.  Wouldn't they form a cartel to do something special that increases their share of the rewards?  Would their members notice it?  Would they really care?
I don't think they would form a cartel. Forming such a cartel would have similar negative effects that a 51% attack would have. It would also be difficult for any of the parties to have any substantial leverage in negotiations because of how easily various miners can leave the various pools.

The various pools also have incentives to get as much of an advantage over other pools as possible. Forming cartels would take away from those advantages.
Quote
Even satoshi had predicted that, over time the network would become somewhat centralized, some might even say that he blessed the somewhat centralization of mining (however this would probably be more speculation then fact):

--snip--

I anticipate there will never be more than 100K nodes, probably less.  It will reach an equilibrium where it's not worth it for more nodes to join in.  The rest will be lightweight clients, which could be millions.

At equilibrium size, many nodes will be server farms with one or two network nodes that feed the rest of the farm over a LAN.

Satoshi was very unhappy about the prospects of GPU mining and called for a moratorium on such programs until bitcoin was better established.
My understanding is that satoshi thought that GPU mining was starting too soon and wanted others to have a chance to mine bitcoin at a lower difficulty before people started GPU mining in mass. He obviously knew it was going to come, and some even speculated that he was GPU mining at first.
His vision seems clear:  he expected mining to remain decentralized, done by the clients themselves, off and on at random times; while the relay nodes (that must be running continuously, and would have a heavier load) would be relatively centralized, as above.
I am not sure about this. Back in 2010, the terms nodes and miners were synonyms. He stated that he predicted that one or two nodes would feed a server farm via  LAN, and I cannot think of any reason why any kind of server farm would be doing anything to do with bitcoin other then it is mining. 

By the way, Adam and the other small-blockians have been insisting all along that raising the block size to 8 MB would make full nodes too expensive to run, and that would lead to more centralization.  That is already a dishonest argument because they always replace size limit by size, even after that "misunderstanding" is pointed out to them.
Yup.
People have asked Adam & co many times for an explicit analysis, with numbers, of the impact of the 8 MB limit, but they never gave one.  Well, yesterday I tried to do such an analysis myself. I don't know whether it is correct, but it showed that the impact would be small, and 8 MB limit might even be better for the full nodes in case of a spam attack. 
The bottom line is that increasing the block size limit would make similar attacks that we are encountering today to be uneconomical and once an attack has spent the money it can spend, it will be much easier for the miners to "clean up" the mess left behind.
But then I learned that what they call "full node" is actually a "selfish node" -- that just downloads and checks the blockchain, but does not relay transactions or blocks to anyone else. 
I would not agree with that definition personally. That may be that person's definition of a full node, but it is not mine. I would argue that someone needs to accept incoming connections (e.g. relay the blockchain and transactions to others).

If one consider the full nodes that do transaction propagation and serve queue data to miners and other applications, the impact of an increase in average block size (not just the limit!) is much less, because it will not affect the larger cost of propagating and validating transactions.  Bummer.
I suspect that over the long term, entities that have a vested interest in ensuring that transactions are properly propagated will be running a large number of full nodes. This would include entities similar to bitpay/coinbase who process transaction on behalf of merchants and have a vested interest in having all transactions (not just those being spent to them) in being properly propagated throughout the network.
Meuh6879
Legendary
*
Offline Offline

Activity: 1512
Merit: 1011



View Profile
July 12, 2015, 12:38:39 PM
 #266

Bitcoin network work better ... ?

favdesu
Legendary
*
Offline Offline

Activity: 1764
Merit: 1000



View Profile WWW
July 12, 2015, 12:57:46 PM
 #267

Bitcoin network work better ... ?


yes, transactions are way down. however, the mempool size is still pretty big see https://tradeblock.com/blockchain

Herbert2020
Legendary
*
Offline Offline

Activity: 1946
Merit: 1137


View Profile
July 12, 2015, 01:17:39 PM
 #268

I find it great that the transactions volume is at a new level it shows growth even if is just spamming of stress testing, all network most be tested and taken to its limits to improve and innovate to a better one.

ok, you test and report the results so that the subject can be improved.
first of all there is a reason why a testnet exists.
second, these so called tests are getting annoying to everyone by delaying their bitcoin transactions, which means it is hurting the users more than it can prove any flaw in bitcoin system.

Weak hands have been complaining about missing out ever since bitcoin was $1 and never buy the dip.
Whales are those who keep buying the dip.
canth
Legendary
*
Offline Offline

Activity: 1442
Merit: 1001



View Profile
July 12, 2015, 01:34:03 PM
 #269

I find it great that the transactions volume is at a new level it shows growth even if is just spamming of stress testing, all network most be tested and taken to its limits to improve and innovate to a better one.

ok, you test and report the results so that the subject can be improved.
first of all there is a reason why a testnet exists.
second, these so called tests are getting annoying to everyone by delaying their bitcoin transactions, which means it is hurting the users more than it can prove any flaw in bitcoin system.

Bitcoin isn't about "doing the right thing" - if there exists an opportunity to annoy others but which allows a motivated individuals can effect change, it will happen. Aka, it doesn't matter that you're annoyed. It matters that Bitcoin allows for low cost spam attacks and that this in itself should be resolved.

TLDR; if I were the one running this spam/trx attack and people kept on complaining about how annoying it was, I wouldn't stop - I'd keep it up until something changed.

JorgeStolfi
Hero Member
*****
Offline Offline

Activity: 910
Merit: 1003



View Profile
July 12, 2015, 02:17:50 PM
 #270

If someone were to actively attack the network with 51% of the hashrate, or otherwise attempt to attack the network, then the short term effect of bitcoin would likely be severe to the negative. If someone were to execute a successful 51% attack, then why would someone want to buy bitcoin? Why would someone not want to sell it until such attack is over? IMO any actual attack on the network that is successful will have very serious short term effects on the price of bitcoin. I don't think simply having 51% of the network alone would have any serious impact other then that miners will likely migrate away from that pool.

Those who dismiss the risk of a 51% attack by that argument usually assume that (a) the attack will be a double-spend or some other banal fraud, and (b) most bitcoiners will scream 'bitcoin is dead" and dump all their holdings when that happens.

As for (a): A cartel or monopoly is usually formed with the goal of increasing the members' revenue over the long term.  In a free market (which is not "a market without government regulation"), the price of the product or service ends up being  a "fair" price, that makes its suppliers roughly as profitable as any other business. In a monopoly or oligopoly market, where new suppliers are excluded by non-market means, the price is such that maximizes the total net revenue of the suppliers.   The banks use their cartel power by charging high fees, not by stealing money from client accounts.

It should be easy to see how a mining cartel with (say) 67% of the hashpower could make smaller miners or pools unprofitable, by selectively orphaning some of their blocks.  If the cartel wished, they could starve all other miners that way, and earn 100% of the block reward instead of only 67%.  That would be a 50% increase in their total revenue, probably a 100% increase in their profits.  With current prices, that extra revenue would be almost 360'000 USD/day, or 130 million USD/year.  The temptation is clearly there -- especially if a price drop to, say 200 USD/BTC, or the next halving of the reward, would make them unprofitable.

A large miner or mining cartel, even with less than 50%, could force an increase in the transaction fees, if there are enough clients who would pay a higher fee for a faster service.  At present, that is not worthwhile, since even if they were 5 or 10 times higher, the transaction fees would still be peanuts compared to the block reward.  However, a majority cartel could force arbitrarily high fees, by orphaning any block that contains transactions with lower fees.  

In fact, I have described in detail, several times, how a majority cartel could impose a change in the protocol --such as a postponment of the next halving -- to all the other players.  Most bitcoiners simply refuse to understand, and claim that the "economic majority" has the power, not the miners.  The few who do understand have ready a "defense" against that: all faithful bitcoiners would abandon bitcoin, and create an altcoin that can be mined only with CPUs -- and declare it as being "the" true bitcoin.

As for (b), that is not the normal reaction of people in such cases.  Americans did not burn their dollar bills when the US government removed their gold/silver backing.  People do not stop using banks or credit cards in protest when these use their cartel power to raise fees or cut services.  People do not stop buying medicines when the pharma industry uss patent laws to charge exorbitant prices.  In general, a cartel or monopoly can force its users to accept any change (such as higher prices) that would be less harmful to them than doing without the service.

For example, suppose that a small miner complained of having a higher orphan rate than the rest, even when he won the block race by tens of seconds.  I bet that 99% of the bitcoin users and holders would not even get to know about his complaint; and the other 1% would blame it on network delays, or some fault of the miner himself.  

Even if it became obvious that a cartel of miners was targeting him, the few bitcoiners who cared would rationalize that "it is the free market at work" (obviously not!), "it has little effect on the hashrate", or "it is good for bitcoin". (Like: "Remember the 6-block reorg after the BIP66 fork? It was started by BTCNuggets, a small miner that had not yet upgraded to v3 when 95% of them already had.  If there were no small miners, the reorg would not have happened.  Small miners increase the risk of a persistent coin split at a fork.")

As for the miners in the cartel pools, why would they switch to a non-cartel pool?  By staying with the cartel, they would increase their revenue as the small competitors are driven out of the field.  By switching, they would risk becoming the next targets of the cartel.  On the contrary, miners in the non-cartel pools would be tempted to join the cartel.

Even if the cartel used its majority power to impose a change in the halving schedule, I bet that most holders would hide their fears and speak out in support of the change, "because it strengthens the security of the network", "because it avoids a fee hike that would be fatal for bitcoin's adoption", etc..  They would do that precisely to preserve the value of their investment.
 
Quote
From the looks of it, the recent issues with the soft fork and f2pool mining invalid blocks cost [ the hash-stealing miners ] >100 BTC, although they claim it was less. The fact that they have not stopped SPV mining would imply that they made in excess of what they lost in additional mining revenue as a result of their SPV mining, probably from winning orphan races and from being able to mine additional blocks they would not otherwise be able to mine.

Indeed. I read somewhere a quote from F2Pool's reply to the critics.  They said that they had disabled parent-block validation (in parallel with next-block mining) after losing a orphan race a while ago.  They would re-enable that, but had no intention of stopping the hash-stealing practice (that implies mining an empty block, and possibly winning it, well before they get the whole parent block).

Quote
Owning 51% of the network hashrate in itself is not a major issue, the problem lies when a single entity starts to actually attack the network after they reach such a percentage. Even if they were to disguise their found blocks as belonging to other entities, it should still be fairly obvious when they are executing some kind of 51% attack.

IIRC, no one cared when GHash.io was accused of attempting double-spends against some gambling site.  The "community" (actually, the 1% or so who follows such news) reacted only when they passed 51% -- because, until then, it was common knowledge that the security of the network would not be guaranteed if a miner had more than 51%.  

However, the "51%" risk is not much less if >60% of the power is in the hands of 4 Chinese miners (who, in spite of being competitors, have already a history of cooperation on issues of common interest).  Only that, since then, the "community" has developed a new "defense" against such risks: they convinced themselves that it is the "economic majority" that matters, not the mining majority.

Quote
I am not sure about this. Back in 2010, the terms nodes and miners were synonyms. He stated that he predicted that one or two nodes would feed a server farm via  LAN, and I cannot think of any reason why any kind of server farm would be doing anything to do with bitcoin other then it is mining.  


You've completely misunderstood what satoshi said.

Well, you can believe what you want.  Considering how the protocol is supposed to work, decentralization of mining is much more critical than decentralization of relay nodes.  As long as a client has access to one honest relay node, he will get and recognize the true blockchain, no matter what all the other nodes do; and he can drop any nodes that he suspects are trying to fool him.  Whereas every client is forced by the protocol to use the blockchain that is being maintained by the majority of the miners, and cannot selectively "drop" suspect miners from it...

I suspect that over the long term, entities that have a vested interest in ensuring that transactions are properly propagated will be running a large number of full nodes. This would include entities similar to bitpay/coinbase who process transaction on behalf of merchants and have a vested interest in having all transactions (not just those being spent to them) in being properly propagated throughout the network.

That is exactly my understanding of what Satoshi meant by that quote.  And it makes sense, too. Wink

Academic interest in bitcoin only. Not owner, not trader, very skeptical of its longterm success.
scarsbergholden
Hero Member
*****
Offline Offline

Activity: 686
Merit: 500



View Profile
July 12, 2015, 02:58:10 PM
 #271

A lot of good points and pretty long debate of what can transform from this type of stress testing i do find that the network transactions are sticking around 170k to 200k in the last week is that about a 30 to 70% increase with the stress testing or spamming transactions going on, i have noticed the panic in people and some regular users being stress about the confirmation time period but we are in a good path to evolution.

unamis76
Legendary
*
Offline Offline

Activity: 1512
Merit: 1005


View Profile
July 12, 2015, 03:03:54 PM
 #272

I don't know if this has been discussed, but here it goes: what is the best way to spend the outputs that spammers are leaving on well known private keys? We could try and drain their funds, a bit like f2pool did.
JorgeStolfi
Hero Member
*****
Offline Offline

Activity: 910
Merit: 1003



View Profile
July 12, 2015, 03:13:44 PM
 #273

I don't know if this has been discussed, but here it goes: what is the best way to spend the outputs that spammers are leaving on well known private keys? We could try and drain their funds, a bit like f2pool did.

I read that those are being swept away as fast as the backlog will allow.

Academic interest in bitcoin only. Not owner, not trader, very skeptical of its longterm success.
unamis76
Legendary
*
Offline Offline

Activity: 1512
Merit: 1005


View Profile
July 12, 2015, 03:15:40 PM
 #274

I don't know if this has been discussed, but here it goes: what is the best way to spend the outputs that spammers are leaving on well known private keys? We could try and drain their funds, a bit like f2pool did.

I read that those are being swept away as fast as the backlog will allow.

By pools/miners or regular users?
achow101
Staff
Legendary
*
Offline Offline

Activity: 3388
Merit: 6577


Just writing some code


View Profile WWW
July 12, 2015, 03:23:02 PM
 #275

I don't know if this has been discussed, but here it goes: what is the best way to spend the outputs that spammers are leaving on well known private keys? We could try and drain their funds, a bit like f2pool did.
I tried that already with 2 of the addresses that have been spamming. It would cost 2.4 BTC fee and the transaction would be so large it wouldn't even fit in a block.

ACCTseller
Hero Member
*****
Offline Offline

Activity: 532
Merit: 500

no longer selling accounts


View Profile
July 12, 2015, 03:41:37 PM
 #276

I don't know if this has been discussed, but here it goes: what is the best way to spend the outputs that spammers are leaving on well known private keys? We could try and drain their funds, a bit like f2pool did.
I tried that already with 2 of the addresses that have been spamming. It would cost 2.4 BTC fee and the transaction would be so large it wouldn't even fit in a block.
You would obviously need to split the transaction up into a number of other transactions. Right now there is roughly 1.41 BTC unspent in the addresses whose private keys have been disclosed, you would want to have 200 or so inputs per transaction making a roughly 30kb transaction and paying below market fees. The fact that there is such a large backlog in the mempool means that it would be unlikely that these kinds of transactions will get confirmed anytime soon. It also means that it will probably be more profitable for the miners to confirm other transactions (with higher tx fees) then to mop up the unspent outputs from the spam
RoadTrain
Legendary
*
Offline Offline

Activity: 1386
Merit: 1009


View Profile
July 12, 2015, 04:44:04 PM
 #277

Those who dismiss the risk of a 51% attack by that argument usually assume that (a) the attack will be a double-spend or some other banal fraud, and (b) most bitcoiners will scream 'bitcoin is dead" and dump all their holdings when that happens.

/snip

These are actually very good points you've made, though they have been discussed before; still, it's a good summary.

I think these issues are more pressing, and the community must be aware of them. The developing effort must be directed towards these.
yeponlyone
Hero Member
*****
Offline Offline

Activity: 504
Merit: 500



View Profile
July 12, 2015, 05:57:12 PM
 #278

New Bitcoin-core v0.11 released here.

It will take people time to update, but it will interesting to see if we will see gradual improvements quite early. Also comfortable that they are working on a more robust solution.
Meuh6879
Legendary
*
Offline Offline

Activity: 1512
Merit: 1011



View Profile
July 12, 2015, 06:25:28 PM
 #279

Did you guys not increase your minrelaytxfee? I'm guessing some bitcoin clients are going to crash if it continues to grow ... (if it didn't happen already)

not crash ... but you must have an high bandwidth capacity to stop ... spamming "imputs already spends".

Terabot
Newbie
*
Offline Offline

Activity: 9
Merit: 0


View Profile
July 12, 2015, 08:02:50 PM
 #280

No i do not think we are doing that.
Pages: « 1 2 3 4 5 6 7 8 9 10 11 12 13 [14] 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 »
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.19 | SMF © 2006-2009, Simple Machines Valid XHTML 1.0! Valid CSS!