Carlton Banks
Legendary
Offline
Activity: 3430
Merit: 3080
|
|
March 04, 2017, 10:52:30 PM |
|
Even if it is spam, the fact that they can so easily cripple the network means capacity needs to grow.
It is almost trivial to spam at any block size limit that the network could possibly operate with. Proof positive that jonald and the BUcoiners stand behind a dev team coup that wish to encourage the blockchain to grow at the same pace as the spam, and become too large to download for a typical user. They know full well that no blockchain in 2017 (or any time soon) will be able to handle the transaction demand they are positing, so as to "demonstrate" that decentralised cryptocurrency is flawed. Bitcoin will only fail due to centralisation if the balloon-chain bad actors get their hands on the keys to the codebase.
|
Vires in numeris
|
|
|
jonald_fyookball
Legendary
Offline
Activity: 1302
Merit: 1008
Core dev leaves me neg feedback #abuse #political
|
|
March 04, 2017, 11:02:25 PM |
|
Side-effect: You can mine absurdly large blocks and DOS everyone. Absurdly high orphan rates are just the cherry on top of the cake.
That is your theory. I would say that miners have no incentive to allow absurdly large blocks or to engage in behavior that promotes high orphan rates. That's the whole point of POW. It makes attacking the network exponentially more difficult and expensive than sending a few thousand transactions through.
|
|
|
|
rizzlarolla
|
|
March 04, 2017, 11:20:22 PM Last edit: March 04, 2017, 11:47:32 PM by rizzlarolla |
|
Interesting. All tx paying 0.0001btc fee, All are straight into a block - no waiting, (are they even in the mempool numbers?) All blocks found by bitfury. Is that correct, already stated or relevant. (i did not check them all, but it seems to be so!) --- (thought i saw a link to another thread about this, did i miss the link?) --- Edit. from https://bitcoincore.org/en/segwit_adoption/#20 Bitfury (segwit) ready mining pool bitcoin core
|
|
|
|
-ck
Legendary
Offline
Activity: 4270
Merit: 1645
Ruu \o/
|
|
March 04, 2017, 11:43:05 PM |
|
All tx paying 0.0001btc fee, All are straight into a block - no waiting, (are they even in the mempool numbers?) All blocks found by bitfury.
There's a good chance they were lingering for a very long time before being mined by bitfury. Just because they're not seen elsewhere doesn't mean they were just sent that instant. Many nodes will simply not have had the transaction in their mempool but they could have even been sent a week ago. While most nodes tends to have the same high fee transactions in their mempool, when the network is busy the lower fee transactions visible on each node are wildly different.
|
Developer/maintainer for cgminer, ckpool/ckproxy, and the -ck kernel 2% Fee Solo mining at solo.ckpool.org -ck
|
|
|
rizzlarolla
|
|
March 05, 2017, 12:04:34 AM |
|
All tx paying 0.0001btc fee, All are straight into a block - no waiting, (are they even in the mempool numbers?) All blocks found by bitfury.
There's a good chance they were lingering for a very long time before being mined by bitfury. Just because they're not seen elsewhere doesn't mean they were just sent that instant. Many nodes will simply not have had the transaction in their mempool but they could have even been sent a week ago. While most nodes tends to have the same high fee transactions in their mempool, when the network is busy the lower fee transactions visible on each node are wildly different. There must be at least an equally good chance they were not lingering in the mempool. This clearly needs further investigation. Funny if they are all shown to be in bitfury blocks, unseen by other nodes. (queue jumping) I haven't got time tonight, maybe you, or someone else will look properly? No worries, it's all on the blockchain.
|
|
|
|
-ck
Legendary
Offline
Activity: 4270
Merit: 1645
Ruu \o/
|
|
March 05, 2017, 02:17:53 AM |
|
All tx paying 0.0001btc fee, All are straight into a block - no waiting, (are they even in the mempool numbers?) All blocks found by bitfury.
There's a good chance they were lingering for a very long time before being mined by bitfury. Just because they're not seen elsewhere doesn't mean they were just sent that instant. Many nodes will simply not have had the transaction in their mempool but they could have even been sent a week ago. While most nodes tends to have the same high fee transactions in their mempool, when the network is busy the lower fee transactions visible on each node are wildly different. There must be at least an equally good chance they were not lingering in the mempool. This clearly needs further investigation. Funny if they are all shown to be in bitfury blocks, unseen by other nodes. (queue jumping) I haven't got time tonight, maybe you, or someone else will look properly? No worries, it's all on the blockchain. Definitely looking into it... I see other pools have mined transactions to the same address; only the ones that were mined on the same block coincided with the same pool (obviously.)
|
Developer/maintainer for cgminer, ckpool/ckproxy, and the -ck kernel 2% Fee Solo mining at solo.ckpool.org -ck
|
|
|
Lauda
Legendary
Offline
Activity: 2674
Merit: 2965
Terminated.
|
|
March 05, 2017, 07:33:57 AM |
|
Side-effect: You can mine absurdly large blocks and DOS everyone. Absurdly high orphan rates are just the cherry on top of the cake.
That is your theory. I would say that miners have no incentive to allow absurdly large blocks or to engage in behavior that promotes high orphan rates. That's the whole point of POW. It makes attacking the network exponentially more difficult and expensive than sending a few thousand transactions through. There is no theory. These are facts. You don't even understand why Bitcoin has the 'low' orphan rates today. You are also viewing this completely wrong. Let's assume all the miners are being honest (which is a very bad assumption). They have two options: 1) Produce much bigger, yet non-absurdly large blocks -> this solves nothing; spamming is a bit less trivial; mempool is still flooded with spam. 2) Produce absurdly large blocks to minimize mempool size -> high orphan risk, DoS. A side effect of this is mining centralization as well. You are left with practically an 'ultimatum'; solve nothing & bloat the chain with bigger blocks or solve the problem but practically DoS everyone and risk with higher orphan rates. Interesting. All tx paying 0.0001btc fee, All are straight into a block - no waiting, (are they even in the mempool numbers?) All blocks found by bitfury. Bitfury was recently testing out how to include the most number of transactions per block. Considering that these transactions aren't large in size, it is no surprise that they've been mining them.
|
"The Times 03/Jan/2009 Chancellor on brink of second bailout for banks" 😼 Bitcoin Core ( onion)
|
|
|
rizzlarolla
|
|
March 05, 2017, 09:09:50 AM |
|
Sending tens of TXs in the same second, and a total of several thousands over a small period of time from yourself - to yourself.
I see. ok, then i guess it is spam. Network is still congested though fundamentally as you can see from the avg blocksize chart. There is no denying this one. The case that I've presented is very clear and I've also listed potential reasons for it. Your case did seem very clear - BU was spaming the network for their own gains. In the other thread you explain why this spam attack might be happening, Plenty of reasons: 1) Complain about congestion to promote the contentious fork called BU. 2) You just want to harm Bitcoin. 3) You want to earn more (in case a miner is behind this). and here, The reason for such spam could be: 1) To promote BU (politically motivated). 2) To raise fees (miners which are motivated by greed). 3) Other. But now that i identified it is bitfury that is spaming, a core supporting pool not a BU pool, you say. Bitfury was recently testing out how to include the most number of transactions per block. Considering that these transactions aren't large in size, it is no surprise that they've been mining them. Like now all is fine and dandy, you just shrug this "spam" off as "testing" now? I don't think bitfury needed to include 700 of these spam tx's (over 25% of block) in a single block to test how to fill a block. https://blockchain.info/block/000000000000000000242e1d51b4143acc9058eba5fc9a325457577a639a7e12These spam tx's are larger than median, around 380 bytes. It is clearly a bitfury address. Tx's all created and included instantly, all in bitfury blocks. This is a core attack, not BUDefinitely looking into it... I see other pools have mined transactions to the same address; only the ones that were mined on the same block coincided with the same pool (obviously.)
Oh really? i don't see any other pools mining tx's from this address. But ok, if you have found those blocks i await your forthcoming evidence.
|
|
|
|
Lauda
Legendary
Offline
Activity: 2674
Merit: 2965
Terminated.
|
|
March 05, 2017, 09:11:55 AM |
|
But now that i identified it is bitfury that is spaming, a core supporting pool not a BU pool, you say.
No, you have not done that. It is one thing to produce a transaction and it is a completely different thing to mine it. Bitfury was recently testing out how to include the most number of transactions per block. Considering that these transactions aren't large in size, it is no surprise that they've been mining them. Like now all is fine and dandy, you just shrug this "spam" off as "testing" now? I don't think bitfury need to include 700 of these spam tx's (over 25% of block) in a single block to test how to fill a block. https://blockchain.info/block/000000000000000000242e1d51b4143acc9058eba5fc9a325457577a639a7e12It is still spam. You are obviously very misinformed about who is doing what in recent times. These spam tx's are larger than median, around 380 bytes. It is clearly a bitfury address. Tx's all created and included instantly, all in bitfury blocks.
This is a core attack, not BU
Doubtful at best until I see proof of this. The average transaction size is well above the median FYI.
You've linked an interesting block though. It may be worth asking them via Twitter to see what's up.
|
"The Times 03/Jan/2009 Chancellor on brink of second bailout for banks" 😼 Bitcoin Core ( onion)
|
|
|
rizzlarolla
|
|
March 05, 2017, 09:29:00 AM |
|
When every tx is timestamped at the exact same time every block that carries those tx is mined, always by the same pool - bitfury- then we can conclude it is a bitfury address. You just said bitfury were "testing", now you think it is not their address? How am i misinformed? You thought it was a BU attack not me. I know the tx is well above median, i just told you that 'cos you previously said "these transactions aren't large in size"
|
|
|
|
Lauda
Legendary
Offline
Activity: 2674
Merit: 2965
Terminated.
|
|
March 05, 2017, 09:39:12 AM |
|
When every tx is timestamped at the exact same time every block that carries those tx is mined, always by the same pool - bitfury- then we can conclude it is a bitfury address. You just said bitfury were "testing", now you think it is not their address?
You didn't understand anything from my post; stop jumping to conclusions. Bitfury was testing how to include the most number of transactions per block from the mempool. This does not include creating their own transactions which is counter-intuitive! How am i misinformed? You thought it was a BU attack not me.
Wrong. I laid out some possible reasons behind the attack. I never said it was a "BU attack". I know the tx is well above median, i just told you that 'cos you previously said "these transactions aren't large in size"
If the transactions are below the average TX size, then they aren't large in size. Sequitur. If Bitcoin can't resolve high fee and transaction delay, will replace by another one.
Another example of nonsense from people with impaired knowledge. Any other altcoin faces the same problems if not worse (e.g. XMR, ETH scalability). P.S. Lisk is a shitcoin.
|
"The Times 03/Jan/2009 Chancellor on brink of second bailout for banks" 😼 Bitcoin Core ( onion)
|
|
|
rizzlarolla
|
|
March 05, 2017, 10:17:25 AM |
|
Ok, i should wait for ~ck to show his evidence that other pools were also including these tx's. Will you tweet bitfury to ask about that block i linked, and all 31000 tx's in general. I don't see why bitfury would have to include any of these tx's to test how to fill a block, when the mempool is at 10's of thousands. You never said "it's a BU attack" but that was top reason on both lists. You were certainly implying that was what you thought and wanted others to see. I'm not getting this response at all. Why do you keep telling me this? I know how big the tx's are!It is me that keeps telling you that these tx's are bigger than median, or bigger than average. I even stated how big they are "around 380 bytes". Just report that valueless shitposter above you and the mods may remove it. 2 posts and 2 minutes later nyanhtet rehashed his wisdom https://bitcointalk.org/index.php?topic=1550041.msg18072136#msg18072136"Bitcoin inverted to replace fiat money, if no one use, will replace by another one." Or maybe he will be the next shitposter promoted?
|
|
|
|
Lauda
Legendary
Offline
Activity: 2674
Merit: 2965
Terminated.
|
|
March 05, 2017, 10:24:29 AM |
|
Ok, i should wait for ~ck to show his evidence that other pools were also including these tx's.
Sure. Will you tweet bitfury to ask about that block i linked, and all 31000 tx's in general.
I would if I had social media. That's why I asked you to do it. Well, anyone reading this please tweet Alex. https://twitter.com/sysmannetI don't see why bitfury would have to include any of these tx's to test how to fill a block, when the mempool is at 10's of thousands.
Likewise, but this would fit their 'testing the inclusion of high number of TXs' per block. At this size they can include at least 2.5k TXs. You never said "it's a BU attack" but that was top reason on both lists. You were certainly implying that was what you thought and wanted others to see.
No. The list is not sorted based on probability, i.e. the numbering doesn't represent anything. It is just a list of potential causes without any particular order. It is me that keeps telling you that these tx's are bigger than median, or bigger than average. I even stated how big they are "around 380 bytes".
Median TX size < these TXs < average TX size (~500 bytes). See here: https://tradeblock.com/bitcoin/historical/1w-f-tsize_per_avg-11101We both know that is not likely going to happen. It is more likely that he will become the next staff member.
|
"The Times 03/Jan/2009 Chancellor on brink of second bailout for banks" 😼 Bitcoin Core ( onion)
|
|
|
|
rizzlarolla
|
|
March 05, 2017, 12:12:11 PM |
|
Ok thanks, i need to modify my statement. This addy is first used in dec 16, but not much. Out of 31000 tx's from this addy, only 500 were sent before the 19/2/17. (your linked tx was from 18/2/17) These early tx's were mined a few at a time by many different pools, but also included a good fee - usually 0.0003 (77 sat/byte) 1 block mined by BTC.com (454214) had fees as high as 0.00065 (168 sat/byte) Hard to call 0.0003 fee spam, let alone 0.00065. Of course these will be confirmed by any miner. But an attack of that nature would be very expensive for the attacker on any kind of significant scale. Then, for the other 30500 tx's from 19/2/17 till now, the fee drops to 0.0001 (25 sat/byte) leaving only bitfury mining these tx's at this fee. And confirming hundreds at a time, see this block from 25/2/17 as another example https://blockchain.info/block/0000000000000000022a0a477bb26f65793d501d7e5721715e7ad301a2e2bc5bBitfury must be confirming 2000 of these "spam" tx's per day on average. So yes, the first few hundred tx's paid good fees and were mined by many pools, a few tx's at a time, the other 30500 tx's seem to be exclusively mined by bitfury, for a much smaller fee. ------- Lauda, i don't tweet either. btw - as if by magic the shitpost was gone!
|
|
|
|
LoyceV
Legendary
Offline
Activity: 3472
Merit: 17588
Thick-Skinned Gang Leader and Golden Feather 2021
|
|
March 05, 2017, 01:27:52 PM |
|
Couldn't this be a failing piece of software stuck in a loop? 2) Some entity is heavily spamming the Bitcoin network for some malicious reason.
If that person is willing to pay for it, the network should just be able to handle it. Email spam makes up 97% of the 200 billion emails sent each day. The email system was adjusted in such a way that it can handle it, instead of making people wait longer to send an email. The reason for such spam could be: 1) To promote BU (politically motivated). 2) To raise fees (miners which are motivated by greed). 3) Other. No matter what the reason is, spam won't go away, which means the network should be able to handle it without interrupting normal usage. Side-effect: You can mine absurdly large blocks and DOS everyone. Absurdly high orphan rates are just the cherry on top of the cake.
If a miner is willing to risk 12.5 BTC by making an absurdly large block, that's an expensive gamble.
|
| | Peach BTC bitcoin | │ | Buy and Sell Bitcoin P2P | │ | . .
▄▄███████▄▄ ▄██████████████▄ ▄███████████████████▄ ▄█████████████████████▄ ▄███████████████████████▄ █████████████████████████ █████████████████████████ █████████████████████████ ▀███████████████████████▀ ▀█████████████████████▀ ▀███████████████████▀ ▀███████████████▀ ▀▀███████▀▀
▀▀▀▀███████▀▀▀▀ | | EUROPE | AFRICA LATIN AMERICA | | | ▄▀▀▀ █ █ █ █ █ █ █ █ █ █ █ ▀▄▄▄ |
███████▄█ ███████▀ ██▄▄▄▄▄░▄▄▄▄▄ █████████████▀ ▐███████████▌ ▐███████████▌ █████████████▄ ██████████████ ███▀███▀▀███▀ | . Download on the App Store | ▀▀▀▄ █ █ █ █ █ █ █ █ █ █ █ ▄▄▄▀ | ▄▀▀▀ █ █ █ █ █ █ █ █ █ █ █ ▀▄▄▄ |
▄██▄ ██████▄ █████████▄ ████████████▄ ███████████████ ████████████▀ █████████▀ ██████▀ ▀██▀ | . GET IT ON Google Play | ▀▀▀▄ █ █ █ █ █ █ █ █ █ █ █ ▄▄▄▀ |
|
|
|
Lauda
Legendary
Offline
Activity: 2674
Merit: 2965
Terminated.
|
|
March 05, 2017, 01:36:00 PM |
|
Couldn't this be a failing piece of software stuck in a loop?
That would fall under the list of reasons called 'other'. Yes that is a possibility. If that person is willing to pay for it, the network should just be able to handle it. Email spam makes up 97% of the 200 billion emails sent each day. The email system was adjusted in such a way that it can handle it, instead of making people wait longer to send an email. That statement is complete nonsense. In no way will Bitcoin be able to sustain the amount of spam that it can be thrown at it. No matter what the reason is, spam won't go away, which means the network should be able to handle it without interrupting normal usage.
DDoS can take down centralized systems which are highly efficient in comparison. How do you expect a decentralized system to resist its own DoS? If a miner is willing to risk 12.5 BTC by making an absurdly large block, that's an expensive gamble.
See my last assertion. Either you mine absurdly large blocks to process the transactions or you are left with a overflowing mempool. Big blocks solve nothing.
|
"The Times 03/Jan/2009 Chancellor on brink of second bailout for banks" 😼 Bitcoin Core ( onion)
|
|
|
jonald_fyookball
Legendary
Offline
Activity: 1302
Merit: 1008
Core dev leaves me neg feedback #abuse #political
|
|
March 05, 2017, 03:17:51 PM |
|
Couldn't this be a failing piece of software stuck in a loop?
That would fall under the list of reasons called 'other'. Yes that is a possibility. If that person is willing to pay for it, the network should just be able to handle it. Email spam makes up 97% of the 200 billion emails sent each day. The email system was adjusted in such a way that it can handle it, instead of making people wait longer to send an email. That statement is complete nonsense. In no way will Bitcoin be able to sustain the amount of spam that it can be thrown at it. No matter what the reason is, spam won't go away, which means the network should be able to handle it without interrupting normal usage.
DDoS can take down centralized systems which are highly efficient in comparison. How do you expect a decentralized system to resist its own DoS? If a miner is willing to risk 12.5 BTC by making an absurdly large block, that's an expensive gamble.
See my last assertion. Either you mine absurdly large blocks to process the transactions or you are left with a overflowing mempool. Big blocks solve nothing. As I mentioned, I don't agree with your conclusions, but please tell us: In your vision of how the network should scale, how would it be ddos/spam resistant? And if miners aren't going to be ones enforcing good policies (i.e. BU) , who will?
|
|
|
|
Lauda
Legendary
Offline
Activity: 2674
Merit: 2965
Terminated.
|
|
March 05, 2017, 04:17:27 PM |
|
As I mentioned, I don't agree with your conclusions, but please tell us:
It doesn't matter even if you don't agree that the Earth is a planet TBH. In your vision of how the network should scale, how would it be ddos/spam resistant?
If I had the perfect answer to that question, we wouldn't be in this situation, would we now? And if miners aren't going to be ones enforcing good policies (i.e. BU) , who will?
Giving more power to miners is one of the absurdest ideas that I've heard in a long time. Miners are incentivized by self-greed, not the greater good. This isn't socialism.
Here's an older article on your idea of no block size limit: https://medium.com/@bergealex4/bitcoin-is-unstable-without-the-block-size-size-limit-70db07070a54#.h8qns8clk
|
"The Times 03/Jan/2009 Chancellor on brink of second bailout for banks" 😼 Bitcoin Core ( onion)
|
|
|
jonald_fyookball
Legendary
Offline
Activity: 1302
Merit: 1008
Core dev leaves me neg feedback #abuse #political
|
|
March 05, 2017, 04:28:51 PM |
|
Giving more power to miners is one of the absurdest ideas that I've heard in a long time. Miners are incentivized by self-greed, not the greater good. This isn't socialism. 51% of mining power already has defacto control of the network. As long as honest nodes control a majority of hashing power, the network will function well, whether we are talking about avoiding empty blocks, avoiding super large blocks, or whatever. Therefore, BU doesn't actually "give miners more power", it just facilitates consensus.
|
|
|
|
|