Bitcoin Forum

Block:
https://blockchain.info/block/000000000000000003dd2fdbb484d6d9c349d644d8bbb3cbfa5e67f639a465fe

This block crashes many web-based blockchain browsers, including the linked one.

So the relevant information is:

1) mined by Chinese pool f2pool
2) contains only 2 transactions: coinbase and single regular one
3) the regular one has no fees, but 5569 inputs and 1 one output. It appears to mop up dust outputs for a total of BTC0.05569000
4) it probably should get a medal for UTxO minification
5) the total block size is 999,923 bytes

Edit: Nothing to add, just quoting the reply from Danny in case he changes it:

The BTC0.05569000 total is paid to the same address as the coinbase transaction.  As such, it would appear that the transaction was built by the pool, and that the entire transaction is fees.

captain obvious detected :)

Don't make jokes out of Danny, it's not obvious for everyone for example for people in this thread https://bitcointalk.org/index.php?topic=1112968.msg11816136#msg11816136 (https://bitcointalk.org/index.php?topic=1112968.msg11816136#msg11816136).

Here it all started: https://www.reddit.com/r/Bitcoin/comments/3cf6qg/need_help_moving_coins_to_new_address/

By the way, bitcoin has not yet died even after three months of your death prediction. Good luck in reading tarrot cards ;)

So just getting all of someones satoshis together takes up an entire block and crashes block explorers?
Why is it being marked as a double spend by blockchain.info?

Seems weird to me, with stress tests and forks recently there have been a few mishaps in the last few days, this seems like another.

This one is not crashing...

https://www.blocktrail.com/BTC/tx/bb41a757f405890fb0f5856228e23b715702d714d59bf2b1feb70d8b2b4e3e08

Now that you posted a link it got jinxed and crashes/hangs too.

Kinda cool way to show the weakness of the "web-based services" paradigm: they can't tell the sudden spike of popularity from a garden-variety DDoS. And all this added JavaScript sauce, it could really serve as an example how not to design UI widget architectures.

It is still not crashing at my end. In case, it is crashing at your end, here is a list of alternatives...

OK, there are 330k such outputs in blockchain now
One block can reduce this number by 5569
So we need 60 blocks in a row filled by 1 tx just to eliminate these spam outputs from a blockchain  ;D

I was wrong. It will die soon.

Soon does not mean anything. Give us a specific period like you did last time and we'll meet again ;)

Before halving block reward.

I dont think, you have read your tarrot cards properly this time as well. Here is something I am copying from elsewhere...

Previous Scenario :-
2012 end: Block reward halving
2013 beginning: Hard Fork
2013 mid: Cyprus crisis
2013 end: Bitcoin reached ATH

Current Scenario :-
2015 mid: Greece crisis
2016 beginning: Hard Fork
2016 mid: Block reward halving
2016 end: ? ? ?

You might like to place a bet on your prediction on bitbet.us as well ;)

What is a reason to bet bitcoins if they would cost nothing when (and if) I win?  ;D

The crash/hang is actually a symptom of the site activating its DDoS defenses.

If you are on an ISP with few bitcoiners the DDoS defenders are unlikely to activate the throttle/reduction and other defenses. You may be also accessing that site during the time when most of the people are asleep in your area.

Thanks for the alternate links, but please be aware of the above issues when posting easily clickable links. I have a proxy/vpn on the opposite side of the Globe, that does succeed loading that page.

Wouldn't these spamtransactions be lost after some time when no miner includes them in a block?

I know that this would mean that the miner had to keep all transactions in the RAM but i think it's better to adjus the RAM instead giving in to spam.

Miners love these txs because the fee is 0.0002 per kb today  ;D ;D ;D

Just got another 1 MB transaction from Discus Fish!

edit: I think TradeBlock and Blockchain.info got taken down at around the same time as this

364423   0000000000000000058319fa2282779aada74b249369ecf9baa7ffaed3e11fb0   2015-07-08 17:49   2   976 KB   25.05569000
364422   000000000000000000283e9409575c2426cd72fcb722065431c52b5083e5bb58   2015-07-08 17:45   2   976 KB   25.05569000

This is very useful mining work  ;D
Sending 5k spam outputs takes ~34 * 5k = ~170 kilobytes to spamer
Redeeming them takes 1 megabyte :)
I have bad news for you.

Many of the addresses are compressed. 5569 is the number for uncompressed.

Not getting what's up with these 1mb transactions mined by F2Pool... At least they are achievements and someone failing in trolling :D

Why are these transactions so big? With 5569 inputs they should have ~ 805 kb instead of 976 kb (5569*143/1024).
Seems like the IsDust() calculation is broken. Size is 20% bigger than expected.

Update: we have ~520k these spam/dust UTxO in blockchain as of block #364447

How do you get these numbers? Are all spam tx send to the adresses with the know private key?

I suspect that many of these inputs are using uncompressed public keys.  As such the input size for those inputs will be 180 bytes.

((5569*180) + (44 bytes of transaction overhead and outputs))/1024 = 978 kb.

Exactly.
https://www.reddit.com/r/Bitcoin/comments/3chere/stress_test_giveaway/
https://www.reddit.com/r/Bitcoin/comments/3cf6qg/need_help_moving_coins_to_new_address/

There definitly needs to be something done. Mining all these tx isn't the solution.

Breakdown of the costs:
BTC spend by the spammer:
- Outputs: 5.2 BTC (520000*0.00001, could have been done cheaper, the minium allowed by bitcoin core is 0.00000560)
- Tx Fees: 3.51 BTC (520000/40*0.00027, could have also been done cheaper)
-> 8.72 BTC (2310 $)

"Costs" for the network:
If no one would pick them up:
- UXTO bloat: 19.83 MB (520000*40/1024*1024)
- Additional disk bloat: ~ 20 MB
- Confirmation delay: 40,000 tx got delayed 1 block (2000 tx * 20 blocks, since the spam filled 20 blocks)
- Network traffic: 468.75 GB (probably much more, only calculated with full nodes, 40 MB * 6000 nodes * 2  / 1024)

With pickup:
- UXTO bloat: 0
- Additional disk bloat: 110 MB (20 + 90 MB, 90 MB created by picking up the tx)
- Confirmation delay: 220000 tx got delayed 1 block (2000 tx * 110 blocks, since the spam filled 110 blocks)
- Network traffic: 1289 GB (probably much more, only calculated with full nodes, 110 MB * 6000 nodes * 2  / 1024)

If this guy starts investing more money or if more people start this we have a serious problem.

This problem is well known. Tragedy of the commons (https://en.wikipedia.org/wiki/Tragedy_of_the_commons)

block #364470  dust/spam utxo=558159 ( 5.58159000 btc )

block #364505  dust/spam utxo=584343 ( 5.84343000 btc )

Doesn't sound like they could get rich from it since these are huge transactions isn't it? And the total fees wouldn't be worth it i think.

I wonder what i would do as a miner. There are some that don't include any transaction because they then have an advantage propagating their block with the nodes. They don't care about the fees at all.

block #364583  dust/spam utxo=683383 ( 6.83383000 btc )

block #364649  dust/spam utxo=790472 ( 7.90472000 btc )

block #364739  dust/spam utxo=824463 ( 8.24463000 btc )

Looks like the more these spam takes place, the more price will go up to make the spam costlier. We just tested 295 on preev. Price has touched 357 USD on Huobi (https://www.reddit.com/r/Bitcoin/comments/3cs4tv/wtf_is_going_on_at_huobi_1_btc_2222cny_357usd/). Keep spamming :D

This part of forum is for development & technical discussion.
Not about trading.

Not for 'Speculation' either...


We have a dedicated sub-section for that => https://bitcointalk.org/index.php?board=57.0

I don't think that the spammers are affected by the bitcoin price. They most probably have their stash of coins and they don't buy them at an exchange on the run. The costs are the same regardless of the bitcoin price.

I wonder who does the spam. Does Hearn and Gavin have big pockets? They want their Bitcoin XT. Maybe they funded this.