Bitcoin Forum

...

This latest "Stress Test", or whatever is going on, is really starting to bother me.  I am waiting on a small payment (with a close to $0.00 miner's fee) to be confirmed from 11 and 1/2 hours ago.

 >:(

I did make a small transaction a little while ago (30 mins or so), included a miner's fee of 0.0006 BTC ($0.16), and it was confirmed by the next two blocks.

So, maybe the Stress Test only affects tiny transactions that the miners (understandably here) do not want to confirm (no fees).

Nonetheless, these attackers are clogging the system with their garbage.  It is an attack upon all of us users.  It is inconvenient and abusive.  

And they should not be allowed to get away with it.

I have little technical knowledge of the Bitcoin Ecosystem, but this problem WILL come back.  Maybe ISIS would love to take down the Infidels...

So, what can be done?  Well, I am not competent to really know.  But for the purposes of this thread, why don't I suggest a remedy or two to get the conversation started:

1)  Ban / not recognize transactions under BTC0.001

2)  Ban / not recognize miner's fees under BTC0.0002

3)  I read somewhere here at bitcointalk that skilled and determined hackers can track payments by ISP location...

I understand the above would go against the idea of BTC being a system which could support "micropayments", but it looks like having micropayments invites this kind of abuse...

Comments?

I have a feeling the miners are behind this. If they can create a large backlog of transactions people who wish to use the network will be forced to pay more.
Remember in just 55 weeks they will lose 50% of their income on block_halfing day. http://bitcoinclock.com/


If the top 3 mining pools joined forces (in secret) they could control the fees.

...

Quantus

Hey, could be that the miner's might be behind the attack.  But that would require, as you wrote, three of them (the biggest, else even more).

One thing I learned when I worked at .gov is that sexy secrets are very hard to keep...

Maybe if they were Chinese miners and kept their mouths shut.

"Three can keep a secret if two of them are dead."
 - Mark Twain

What percentage of their income currently comes from block rewards versus transaction fees?

Was there really another stress test today?  I didn't hear any lead up to this one, but that may be because of the "fork of july" incident covered up any coverage of it.  I do see 11K unconfirmed tx at the time of this writing, which seems rather high.

I imagine someone has this handy: a chart over time of the number of unconfirmed transactions according to some well known nodes.  I keep popping in on blockchain.info/unconfirmed-transactions but that only shows you the situation in a single moment.  I'd write a script to build my own chart but it's something that I assume is already out there.  Who's got the right place to look on that?

for the top pools its 100%

who is going to ban / or not recognize these kinds of transactions? miners?
the thing is, they are more than happy to receive these kinds of transaction since there is a lot of fees involved and that means more income for the miners.
i don't see any reason why they should stop receiving these transactions.

From the chart on https://btc.blockr.io/charts, the average block fee is around 0.1 btc to 0.2 btc. With a 25 btc block reward, the tx fee is just 0.4% to 0.8% in average.

Maybe I'm wrong because even if they forced everyone to pay 5 cents (100,000 transactions a day) that's still a drop in the bucket compared to what they make now.

But those values could become much better for the miners if the value of bitcoin increases further and the usage increases..

$0.00 miner's fee transactions naturaly will not get confirmed quicker. If you were a miner, and you had to chose to accept a transactions with zero fees or a transaction with some fees, what would you choose?

The "Stress Tests" is not a attack on the system... It's just a opportunity for miners to make more money from increased miners fees within the system. If it's raining money, you will have to be in the right place, wiith the biggest bucket. {Do not waste mining resources on transactions with zero fees}

The result of these tests will provide benefits for the whole community.   ;)

 

#1 Definitely no, this ismore likely into banning micropayment which is not the best thing to do because it is like putting up a limitation for the small guy and definitely a discrimination. BTC supposed to be useable for any kind of transaction, micro to huge amount. BTC take its pride into providing every single payment and this is definitely a bad option

#2 Miners fee had nothing to do with this btw and it has become "standard" to use 0.0001 BTC per transaction but definitely this can change in the future. Not recoqnizing miners fee under X BTC is the same as option 1, discrimanation and limitation of BTC usage which is not healthy for its growth



There is none "announced" stress test ongoing right now. I consider this to be most likely this is more like an attack to the network and highly likely from those who are behind supporting the increasing of this blocksize. Even stress test is a joke because we have a testnet for someone to try it out without interfering with the live network. The number of the unconfirmed transaction is increasing rapidly

That is true.
If you don't put something to the test, then you do not know what it can handle or not and how it will perform.
None the less, the testers did inform the community about the upcoming test.
What if it was an actual attack by a malicious party with NO warning/announcement of what's coming?

If an attacker had a large enough short position, it could be profitable for the attacker to fund a prolonged attack.

Fees don't work well and should be eliminated, but some other solution must take place - a solution that actually works and that can't be used as an extortion instrument by miners like the fees

yeah there was apparently the network was try to handle 150 TX per second, which is obvious too much for it, and tiny amount have lower priority overall


then what's the point of doing it now?, they can always do it later, with the same fee


what is your suggestion? because it is to say "we need to remove this and remove that"; but if none can offer a better solution, we cannot move ahead

i think for the time being, i'll not bother with it, even in the case that you need to pay 100k satoshi per TX this is not even a problem, we are talking about few cents, camo on, with visa and mastercard fee are far beyond this

First eliminate the fees.

Then there are several ways to avoid spam thru' Atificial Intelligence, Rate Limiting and more. Other systems avoid spam and DoS-like behaviour successfully without charging anything for system access. I could suggest several methods but i'm sure the bitcoin developers can come up with much better and appropriate IF they want to.

BTC is a POW coin which means that it will always need miners to mine blocks for confirmation and the fees are incentive for the miners to work after the block reward goes to 0. Without fees at all, miners will have no incentive to mine anymore . This may only happen in after 100+ years however eliminating fees is not the answer for this.
As for the rate limiting are you actually meant limiting transaction that below X BTC or so? This is also not an option because BTC suppose to give the ability to send micropayment as well . Limitation to the transaction isnt healthy for BTC because it is merely discrimation to some transaction

You got it wrong.

1- Fees are NOT an incentive for miners, it is an incentive for individuals to not spam.

2 - No, i don't mean that. You are refering to a simple value limit, anyway fee is a lazy solution that does not work well and virtually blocks ALL transactions that don't pay the toll. Any solution regardless of the type or the mix of them will be somewhat complex but far less complex than the blockchain itself.

Nope, arallmuus is quite correct.  Fees might not be a huge incentive for miners just yet, but they will be in future.  Unless your solution involves altering the block reward (https://en.bitcoin.it/wiki/Controlled_supply), which in turn alters inflation and the subsequent value of each coin (so no one in their right mind would agree to changing it), then the fees will become the primary incentive for miners as the reward diminishes over time.  Or were you working under the assumption that miners are going to keep securing the network forever for free?



For 1 and 2, your only solution is to release a client that imposes those limits and see if the users securing the network choose to run that code or not.  For 3, what's your plan when you find their origin?  A sternly worded email?

Depends on what your point of view is. I see the miner's fee as the incentive to keep the miners to secure the network even after there isnt any block reward anymore. Given the fact that no one will be mining for loss, then this fee will keep the network running .

Fees has never been an issue, blocksize will be an issue in my point of view . It may be an "attack" for the network now or some people consider it as a spam however I see it from my own point of view that this will be how the network looks like once we have more people to use BTC .
Some people may just say to increase the miners fee and your problem will be gone however increasing the miners fee to push your transaction is merely eliminating micropayment which is hindering the growth of BTC I would say. There should no and never be any limitation about micropayment or so because BTC wasnt created to support large amount of transaction only


People are afraid of a small fork to the blocksize, let alone the block reward then it will be a chaos here. I think he is looking it in the shorter term and didnt put into account of 100+ years later that the block reward will be gone though. Cant blame them, each people has their own opinion anyway

I understand the point you are trying to make, but judging by current fee included in blocks, which is on average about 0,25 btc - 1% of the block reward, they either are not
behind this, or are not successful in making themselves profit.

I find two things more probable:

- The people who are promoting the increase in clock size are making sort of a statement/proof with this spam.
- Since litecoin is being obviously pumped right now beyond any logic (from 1,xx to 5,xx currently), maybe the same people pumping it are trashing bitcoin in efforts to
  convince bitcoin investors into switching/investing into it.

Whatever the case is, it is obvious that developers and community should join efforts and find and implement a solution to this spam, because it is making
bitcoin look bad in a way, is delaying transactions users make, is forcing users to increase the transaction fee, and is generally pissing people off, myself included.

cheers

it is not possible to ignore small transactions like 0.001 BTC because there is a lot of micro transactions being done with bitcoin and banning those transactions is just not possible.

I think this is the key point.  There is no legitimate need for this.  If you want to test you can do so for free on testnet, that is what it is there for.  So it seems that "stress test" is just a way to spin a network attack or some other sort of self serving behavior.

Sorry that space pilot has not gotten his script yet.

Good Luck!

I've had no problems whatsoever with my confirmation times, transactions going on normally. In any case deal with it. We better solve a way to not be affected by this if we want mainstream volume being viable.

Stress testers can test what they want on testnet. They are bloating the blockchain and interferring with real transactions. Miners can increase the transaction fee temporarily to stop this spamming. They don't want to because they want more fees.

Some defense against this sort of attack needs building into Bitcoin. It cost me money waiting for my transaction to go through. I wanted to sell some of my Bitcoins at the top, and by the time my transaction got through the price had gone down, and I had to sell for less.

One of the funny things about these threads is, that people think, when they write often enough, that a stress test is "not allowed" on the main-chain, people will stop doing them.
It's not even like this stress tester have used a bug or something. There is no "not allowed", no court to decide, that what they did is bad.
People who are whining on forums are not the judges. Everything the stress testers did/do is according to the (coded) rules of Bitcoin.
These stress tests are a reminder to the devs, that there are still unsolved problems. Problems that could really hurt Bitcoin, not just if somebody wants to do so, but also when Bitcoin goes mainstream(and nobody knows, when that will happen).

The question is, was BTC designed to become a worldwide payment system that could deal with microtransactions as well "as it"? or did satoshi predict the blocksize problem and the usage of a LN type of solution to deal with the volume?

you solve this my increasing the fee bro, or by leaving your bitcoin into the exchange
it's a good strategy tha many do when they trade daily


this is one of strengh of bitcoin, microtransaction with visa and mastercard are not really possible
and they would cost you a fortune, here microtransaction can be free if someone want

What turvarya said is completely right here.  Shouting to stop is obviously not going to stop them.  They have the right to send transactions and to waste their money on it as they see fit.


These ideas from OROBTC are, IMO, completely crazy.  If you dedice to "ban" transactions under 0.001BTC then you are effectively doing away with the Satoshi and making 0.001 the smallest unit of bitcoin.  But there's clearly a lot of use for a unit smaller than that.  Already bitcoin has hit price points where 0.001 was equal to 1 USD.  Surely you can see the utility of sending less than 1$ sometimes.

(2) is even more crazy because it doesn't recognize that the cost/benefit of mining a transaction isn't in only its fee but in the relation of its fee size to its size in bytes.  There's a cap on the number of bytes in a block, not the number of transactions.  If a miner can squeeze in 5K transactions with just a small fee each (say 1KSat) that's better than 500 transactions with fees of (2Ksat).  Also, as long as there's a coinbase block reward, fees aren't the only motivating factor in the block race.

Your litecoin theory is very interesting. It sounds a little crazy (and is) but if (big if) someone was doing that we could be witnessing the first shot of a new age of crypto wars. Instead of just trashing each others coins with FUD, this could be the opening salvo of active attacks in order to boost the price of an alternative currency.  Imagine if there were no laws protecting Visa or Mastercard.  Supporters/investors of each credit card could strategically attack the others payment system in a series of ongoing actions. If a small number of decentralized cryptocurrencies, say three or four, emerged as the primary competitors with each other I could see this kind of cryptowar emerging.

Reminds me a little of when the record companies tried to flood some of the p2p music sharing networks with fake music files.

You are wrong, do your research. In case you are correct the bitcoin developers lied to everyone about it, sou you saying they are all liars?

I do believe that no one could predict the future but atleast he had foreseen this problem and that is why he leave this all to the hands of what we call as the "BTC developer".
The main point is that BTC was designed and created to be a decentralized currency and there is no denying that it should support micropayment as well. Leaving that aside, few years back the community was so proud that with BTC you could have a micropayment which is not available for visa I believe and now some people are ditching this block issue by saying "just increase your miners fee" which is killing the micropayment


Ever thought of the possiblity that both are correct? Both of the option have a valid point about this I believe. After there is no block reward , then this miners fee will be the incentive for miners to keep on mining otherwise if there is no incentives than no one will sacrifice their electricity cost just for the sake of finding block I believe

Yes, both are valid points, it is an incentive to miners but it is a side effect it is not the intention.  Original implementation was not supposed to have any fees, then fees were  lazily created to avoid spam. And that is what fees are for, to prevent spam,  if they are not serving the purpose fees should be eliminated and an actual solution should be created.

Erm... no.  Fees always have been and always will be part of the plan.  Quoting from the Bitcoin whitepaper (https://bitcoin.org/bitcoin.pdf):


So again, once the block reward diminishes, fees will play a vital role in securing the network.



No one lied about anything, there just seems to have been a misunderstanding here.  I don't know where you got this idea from that fees aren't an integral part of the design, but I'm pretty sure it wasn't one of the developers.

Yes fees are vital to bitcoin miners as incentive and vital to the network to stop spam attacks.
Because of this Micro transactions will have to be forced off the block chain. As of right now 99.4% of the incentive to mine bitcoin
comes from block rewards and only 0.6% from transaction fees.

People need to understand that no change in the protocol can force miners to include more transactions into each block. They are the gate keepers and they hold all the keys.

If we want a secured network, at its current size, without block rewards, we would have to pay 5 USD per transaction (right now) that's the reality of the situation.
It costs 5 USD per transaction to secure the network right now. In the future that cost will have to be paid in full with transaction fees.


Your options are this. Higher fees and a secure network  OR  Micro transactions/low fees and virtually no security.

Yep it's very bad. Yesterday it took around 20 hours to confirm a tx sent to me.
But if someone has paid the miners' fee, he/she has the right to get the ride.
Though there's no convenient way to solve it, some improvement is urgently required.

On the one hand, a raised block size is desirable. We simply cannot just rely on waiting for sidechain or lightning to come up with a solution on this.
On the other hand is the floating tx fee. It should be set to default in every wallet software, especially the SPV clients. At least a prompt to alert the users who want to send any coins when such attacks are going on. The latest Bitcoin core has already got this nice little feature.

But people want a very secure network and zero fees!

(as well as 1GB blocks and instant confirmations)

Stop the damn stress tests I say. I'd like to move money around like it used to be, with low fees and fast confirmations.

End the madness!!

6. Incentive
By convention, the first transaction in a block is a special transaction that starts a new coin owned
by the creator of the block. This adds an incentive for nodes to support the network, and provides
a way to initially distribute coins into circulation, since there is no central authority to issue them.
The steady addition of a constant of amount of new coins is analogous to gold miners expending
resources to add gold to circulation. In our case, it is CPU time and electricity that is expended.

The fees as incentive were a suggestion . not the convention and were not originally part of Bitcoin. If you search this very board you will find very old posts about it, when fees were added. It was all about spam. I will try to find them if i have the time.

bananas you are simply wrong in every way. I will not waste time trying to argue with you tho because your clearly mentally challenged.

I also want to see some evidence that fee's was not part of the original design. Please feel free to post it here.  ;)

According to my research, it was part of the original design to substitute the block rewards. So according to you...What will secure the network when the block reward runs out?

In the end, it does not matter... The fee's lubricate the Bitcoin gears in my opinion... and if it helps to stop spam, it's beneficial to the whole network.  ;D

Stress test still going?  ::) took 10 hours yesterday to clear a simple tx for me. I think it's time to add a higher tx fee for the moment

Can someone give me a headsup?
Is it still going on?
Looking at https://blockchain.info/unconfirmed-transactions there is still a backlog of 16 MB.
at https://tradeblock.com/blockchain it is even 49 MB.

Its not a stress test its an attack. And its only costing a few hundred dollars a day to carry out.
Within 24 hours there will be over 400k unconfirmed transactions in the mempool within a week it will be over 1 million and anyone who tries to run a full node won't have enough RAM and the whole network will fall. Any computer that tries to load all those unconfirmed transactions (because the mempool has no overflow) will crash. Even if we could get consensus for a 2mb block it would only delay the inevitable collapse of the network.  Bitcoin will be dead in 30 days. Bitcoin transactions will slow down far worse then it is now.

Edit: you guys know I"m fucking with you right?

yeah, but it's not that grim in my opinion. micro transaction services will move to dogecoin or other altcoins to prevent days of waiting time, if this continues. correct me if I'm wrong, but we will always be able to send decent value with a little higher fee in a timely fashion?

Yeah the real numbers are much lower around 20k in the mempool but the number is growing. This is not a stress test.

.


agreed

An easier remedy using two options, mainly for miners:
Reserve 1/5th of each block for high priority/low fee transactions.  The spam has very low priority, and won't be helped by it.  Transactions using older inputs and having decent sized outputs will get high priority:

-blockprioritysize=200000  # Default is 50000

Set minimum fee for relaying a transaction to 0.0003 BTC.  This is more than the current spam.  Transactions with lower fees will still confirm as long as the priority (see above) is sufficient:

-minrelaytxfee=0.0003   # Default is 0.00001

The last option will reduce mempool for everyone running full nodes, and will make it harder for the spam to propagate through the network.  If you mine, please consider pushing your pool to do this, or switch to a more bitcoin friendly pool.

This last one is new to me.  Is the idea that you dno't even relay transactions that are othewise valid unless they include a fee?  I didn't realize that anyone had already implemented such a policy.

I think banning any low fee Tx is a reasonable countermeasure. Remember the so called "test" costs money to launch. I do not support asking them to stop, I support bleeding them dry if they want to continue DDOSing the blockchain.

Perhaps we tie the fees to unconfirmed tx.. The more there are the higher the cost.  Want to spam the network when it's got 100k uncofirmed, $10 bucks a transaction, 200k, $50, 500k, $100.

Break the bastards by making miners rich.

Yes I think it's still going on. I checked the amount of unconfirmed transactions on blockchain.info hours ago and it was over 15000. Now it's almost 20000. Someone has to be spamming the network for the unconfirmed transactions to increase that quickly. I've been waiting for a transaction to be confirmed for 20 hours now, and I've not had one confirmation yet.

Stress tests my ass.

Gavin's Bitcoin-XT lovers are f*cking the network, I've been waiting for confirmation since yesterday and now there're over 20MB tx which is not confirmed.

Hey Gavin, is that you? F*cking the network to force everyone for bigger blocks?

I wish that telling them to stop is easy. This is not a stress test, people. This is an attack. If this is a stress testing, some people will claim that they are behind this, seeing whether the network could sustain this spams until it wouldn't work anymore to come up with a possible solution. Turns out, it isn't. People who posted that we've reached the peak of transactions don't understand what really is happening.

I think it's looking like an attack the longer it goes on. KingAfurah, the person who started the stress testing, said he wanted to study what effect increasing the mempool size to >150MB was in one of his last posts. He's been silent for over a week, but i think it's him doing the spamming.

Hmm maybe he is still busy studying the effects of it. Well, it's already exhibiting in the network. I made a transaction 5 hours ago with sufficient fees and until now, 0 confirmations. ??? This is quite annoying, because I need those bitcoins in an hour. If he is up for study, then he must come up also with a plausible and possible solution to end this madness if ever this happened again.

+1

If you must move low amounts of BTC exchange to exchange just use altcoins. There are many of them with high enough liquidity that converting BTC to altcoin, sending altcoin to destination exchange and converting altcoin back to BTC should not be an issue in terms of losing much money (you might actualy make money, e.g. exchange arbitrage). Depending on altcoin, the whole deal can be completed in under few minutes which is mostly not the case with sending BTC.

20MB blocks would solve this  ;)

Even if it is not the original implementation of BTC , the fees will simply be a solution after the block reward run out. It may be a niche thing to consider for now since the block reward is still there however once there is no more block reward then this fees will serve its purpose. Eliminating the fees could be a solution for the current problem however it will make another problem which is why this could not be use as a solution for this problem


Banning the low fee Transaction could not be a countermeasure at all because the attacker will simply only need to increase his "attacks" fee to continue this and everyone will be forced to put in an even higher fee and that is killing micropayment. Killing micropayment is hindering BTC growth I'd say


Not a stress test, more like an attack to prove that we actually need a bigger block sizes


I thought most people here support BTC rather than altcoin? So know people are suggesting to ignore micropayment and ask them to shift into altcoin instead? lol


Yes it is but it creates a whole lot more problem so it is not the solution

better to know how the network behaves when stressed, than be surprised later, once some rich stupids will start with real stress never ending testing scenarios. what we have see now is really "nothing", compare to fact, how much it cost and what is overall impact to transactions..

just think about it.

Yes, not relaying transactions if the fee per kB is too low has been the default policy for years.  Unless the priority is high enough.

Of course, and so far nothing new has been discovered.  If this was done by a responsible person instead of a malicious attacker, it would have been tested on the testnet.  This is the reason why we have a testnet.

any one knows when is the next stress test going to be ? i would like to be ready before sending out any transactions with manual fees last time i was stuck for like 6 hours with no confirmations.

sure, but still...actually doesn't matter who is behind. he must pay fees and he is stressing bitcoin network basically for free. even, it is not comfortable, we all can see now, what you can do with literally few bitcoin and how many users and services you may affect only if you want.

I hope that core developers closely watching what is happening and it will simply help them to find out some feasible solution.

btw, sorry for conspiracy theory, but if if these stress tests are done by miners in order to increase their income, because this is pushing people to pay higher fees for comfortable waiting time?:))

A stress test on a live network is a good way of making people realize about what will "happen" in the future . By simply doing this in live network, he could easily prove that a bigger block sizes is needed because most people will affected by this rather than doing it in a testnet and later presented the effect of it in form of data and charts.

However this is not a test at all because it is not announced before hand and no one knows who did this at all. It is simply an attack to the network. Completely wild west way of showing things to people, you killed a person, throw their dead body in the middle of the city and put up an unintentional message that everyone will suffer if you dont follow up the rules

That is beside my point but in any case if you support just Bitcoin you are not really smart, like people supporting just gold or silver or Facebook stock or whatever other single-point-of-failure stuff.

Probably not, and certainly not without generating other problems.  We only know for sure that larger blocks will have room for more malicious spam and slow everything else down.

why not implement fee rules? like if a tx is under a certain threshhold it MUST offer a fee

Bigger block sizes?  So far this has only demonstrated the fact that the blocks are too big already.  Perhaps this will be an adequate size in the future, but right now 1 MB blocks are too large.  The prospect of even larger blocks with room for even more malicious spam is not a good one.

Announcing a malicious spam attack does not make it a "test".  It is still a malicious spam attack.  Move over to the testnet if you want to test something, e.g. how to stop transaction spam.

This is already a soft rule.  The priority is calculated from the sum of (input size multiplied by age) and divided by transaction size.  Spam usually use young inputs as well, making the priority very low.  If the priority of a transaction is too low, Bitcoin Core won't relay it at all if it doesn't offer a fee.  Certainly not mine it.

Only 6 hours? Hahaha dude, i am waiting for more than 48 hours now and still not even one confirmation, WTF  :o

I see. but it won't stop the mempool from being filled. shouldn't this affect nodes at some point? what's the critical mass to shut a node down in terms of mempool size?

If BTC were to fall to dirt cheap some day, it means people are losing faith with cryptocurrency and there is no denying that the other cryptocurrency will follow the same path as well. The altcoins are merely nothing without BTC anyway ( however there is no denying as well there are some good project of altcoins )


I merely look at this "stress test" from two sides point of view. First, this is an "attack" to the live network and there is no denying to that and second, this "stress test" is actually another way to show people about the effect of the current blocksize rather than showing them data and charts after you do your testing on testnet. People will "learn" much easier from "experiences" .

As for the larger blocks issue, I would say that the potential for larger spam will always be there however limiting the blocksize to the current one or even make it smaller is not the solution for it as well. This may be an "attack" for now however I am seeing this from the brighter side that is how the network will be in the upcoming years after there are alot more people that jump into BTC

This attacker had shown their intention I guess, they want "bigger block size" . Could be from Gavin and his minnions however atleast thanks to this I have a projection of what the future network will be (atleast a glimpse of it )

...

OK, do recall that my proposed ideas to not recognize transactions of under BTC0.001 and/or fees under BTC0.0002 were just "talking points" by a non-expert.  I chose those amounts as "reasonable looking".

Nonetheless, I saw in another thread this idea:

You can have lots of micropayments and very low / no fees

OR

You can have a Bitcoin system that is secure and works well for important payments.

*   *   *

Can we only pick one?  

Serious question.  

Great dialogue, these questions need to be answered IMO.

*   *   *

Also, IMO, this is a MALICIOUS ATTACK.  Anyone or group doing this is inconveniencing others, and in some cases causing real problems, so I do not accept this as a "test".  Should it be miners (or whoever) doing this, I imagine we would find out in due course.

uhh, did you pay some TX fee? can you please post here TX of transaction? this seems very unlikely to me tbh. during last test, I tried by myself to sent 0.1BTC with 0 tx fee and it was done in 15 hours..

https://blockchain.info/address/1BnvHarta94JvXrY7oT272qTBEvgMdRk2S    Here it is, 49 hours and counting,the fee was 1114 satoshis. I dont know should i laugh or cry hahaha

your tx fee is very very low... this could take some time. even without an attack

...

Following up on Xialla's experience, I sent a small amount as a test (roughly 0.005 BTC) with a fairly large fee (BTC 0.0006) to see how long it would take at about 10:30 PM US ET on July 6.

Confirmed by the next block.  And second confirmation by the next one after that.

For me, getting a "real payment" (say, over $1.00) confirmed, even at a higher amount of fee (and I would think that a BTC 0.0002 would be reasonable), is much important than being able to send out $0.20 for free......


But, hey, that's just me.



(Edited for correct amount sent and date, doh!)

Just made another transaction, and so far it's getting better and better on my end, at least. Only took 30 minutes to have 3 confirmations. :)
https://blockchain.info/tx/8807f0e2d48116daa661fa76b55fa852ff4277df6979d9dfd51fb6841f8e54f5

I know and i regret now. Is it possible to undo things? to send somehow a bigger fee for that transaction? sorry if it is a stupid question but I am desperate.

Nope. Bitcoin transactions are irreversible just so you know. You have to wait for days or possibly weeks for that to confirm. In some cases, it takes a couple of weeks before a transaction is confirmed, especially if the sender put no fees for it.

Wrong. There is not a single older tech that has not been replaced by advanced and better tech. For example, Netscape Navigator to MS IE to Firefox to now Chrome, with Internet web-browsers.

Right now you can start using altcoins as a way to move BTC from exchange to exchange faster than using Bitcoin network or you can continue bitching and whinning here all you want, your choice.

Nothing more to add to this debate.

Yes.  Two ways:
1. You can either create a new transaction using the same inputs and a higher fee, hoping for this double spend to confirm.
2. Spend the change output with a high fee, and hope for a pool practicing child-pays-for-parent to pick up both and confirm them.  Eligius, EclipseMC and possibly others do this.

Method 2 has a higher probability of success, as the double spend will have problems getting relayed to the miners.

Never knew that this is possible until I read it today. However I have a question, what if I have created two transactions, one with fees and one without fees? I send 1 btc to the same address two times but with only different fees. Is it possible that the two tx be confirmed at the same time if I only have 1btc?

No, only one can be confirmed, and the other will become invalid at the same time.

Even if they get confirmed at the exact same time? Which transaction will be flagged as invalid by the network? The one with the fees or the one without fees?

It is not possible.  If they confirm in the same block, the block is invalid and will be discarded by all full nodes.

Oh now I get it. It would only be accepted by the nodes if I have sufficient balance to cover for the two transactions.

Ninja edit*

you are right. in other hand, if projected value per btc is in thousands of dollars per coin, it is not so low, how it may looks like. another point is, that far small TX is higher fee just killing. believe me, that average Joe hates waiting and hates fees. and once he will decide to buy a coffee with his fancy smartphone, will be faster and comfortable and cheaper (for free) to use fiat (and some XYZPay implementation) over fiat..

maybe is possible to scale fees depends on how much you are sending or something..

in theory, he could just double spend it with a higher fee. if it goes through the other one will be invalid

You can't double spend it, since a node that has the "old" transaction already in their mempool doesn't accept the "new" transaction.
That is to prevent double spends.

F2Pool has recently enabled 'replace-by-fee' feature (http://lists.linuxfoundation.org/pipermail/bitcoin-dev/2015-June/008843.html), which allows you to 'double-spend' unconfirmed txs by adding higher fee.
Though most nodes would refuse to accept to mempool and relay a replacement tx until they have this feature enabled (which may happen in the future (https://github.com/petertodd/bips/blob/bip-full-rbf-deadline/bip-full-rbf-deadline.mediawiki)).

they have changed to First-seen-safe replace-by-fee [FSS RBF]

https://www.reddit.com/r/Bitcoin/comments/3aejmu/f2pool_we_recognize_the_problem_we_will_switch_to/

there is a lot of confusion as bitcoin had been known famously for free transactions

I agree, I think bitcoin will be surpassed by another cryptocurrency, the only question is when.  If there can be consensus on certain decisions and changes can be implemented the time until that happens can be extended, and might be a long time away.  If not, it will happen sooner.  Either way, network effect or not, there are some things that will be just easier to implement in a new system rather than trying to come to a consensus in order to change an already existing decentralized system.

Wrong.  A typical 1 input - 1 output tx is 192 bytes.  A more typical 1-2 or 2-1 is about 226 bytes.  (Compressed keys, not multisig.)  0.0001 BTC fee is fine for both.  High priority transactions will still confirm with no fees.

indeed. it's better to set a higher tx fee for as long as this attack goes. even $0.20 in fees is nothing compared to other payment processors

It won't take that long.  Some pools, e.g. Eligius, filters the spam.  Some other pools, e.g. BitMinter, has set blockprioritysize to 500KB, half the block, to let high priority non-spammy transactions through regardless of fee.  If only people could stop mining on the China pools, and move over to the more bitcoin friendly pools.

assuming I send with a .001 (25 cents) transaction fee, my transaction should go through in one of the first blocks no?

...

traderbit

Since the attacks started, I have sent two small BTC transactions (one on July 6, one this afternoon), both with a generous BTC0.0006 (= $0.13, 13 US cents).  Both transactions were confirmed by the next block.

It does look like the way to finesse these attacks (at least for now) is just to be generous to the miners.  They have a big role in our Bitcoin Ecosystem.

Agree with above sturle's comments about BTC-friendly miners.  But, I am no expert.

Or if some nefarious party wanted to do it for kicks. It's insanely inexpensive.


Works for me^

If your transaction is a normal one (eg. 1 input, 2 outputs, and a size of ~230 byte), then yes it will very likely get confirmed in the next block. But if your transaction has lots of inputs and outputs, then the answer would be a no.

FYI: http://bitcoinexchangerate.org/fees

I've been deciding not to do anything different and I'm paying the consequences.  Small ones, but annoyances for sure.  I just realized that a transaction I sent yesterday to a cold storage address hasn't confirmed yet.  It's not really a big deal because, as I said, it's going to a savings address and it will get confirmed eventually, but it's still quite an annoyance.  27K unconfirmed is a lot.  Does anyone know yet who's behind this latest attack?

Here:

Thanks, I was trying to find that thread.   Cheers.

I did a transaction on the 9th of July with only 0.0001 btc and my transaction went through and was confirmed within a span of 30 minutes. Has there been any stress testing occurring in the blockchain as of this moment?

There's still between 13MB - 14MB of transactions queued, so unless we've seen a sudden and unexpected spike in legitimate usage (which seems an unlikely coincidence), it's fair to assume that either there's still some testing going on, or it recently finished and we haven't cleared the backlog yet.  According to statoshi (http://statoshi.info/), we peaked at 152 transactions per second earlier today.

the current mempool size is around 70mb, so it's still a way to go before it's safe to use the usual fees https://tradeblock.com/blockchain

Hmm the transaction peak seemed to point out that there might be some testing still going on, though we cannot differentiate thoroughly what is made legitimately or what is made as a part of the testing. There are still 10000+ unconfirmed transactions as of this moment, still a 14MB transaction queue.

EDIT: 11000+ unconfirmed transactions as of this moment.

Yes, safe, but the consequence is you'll need to wait for longer times before you see your transaction confirmed, which is annoying and irritating. In order to get your transaction confirmed faster, you need to set a higher fee than the usual. I don't like that.

Mmmm Hmmm. Which is why, of course, the only browser used is Chrome, having summarily eradicated all other browsers, including IE and Firefox from the face of the earth.

protip - you may want to check the following: https://www.netmarketshare.com/browser-market-share.aspx?qprid=2&qpcustomd=0 (https://www.netmarketshare.com/browser-market-share.aspx?qprid=2&qpcustomd=0)


Third alternative: enclose a sufficient transaction fee, and get your trasactions confirmed almost immediately. As reports of experiments directly upthread report.


<naah... I'll refrain>

Not quite. The network knows nothing of your balance. It only knows of your unspent output. Which can only be unspent until it is spent.

I have been sending transactions with 0.3 mBTC and they always get included in the first block. You might need more if your transaction size is larger.