🌟🔴🌟🎲🌟🔴🌟MONEYPOT.com -- Crypto Casino Suite and Web Wallet 🌟🔴🌟🎲🌟🔴🌟 - page 15.

CrazyCraig

sr. member

Activity: 501

Merit: 340

Bye Felisha!

Quote from: ?? on ??

Quote from: Pseudo-Random on March 09, 2017, 03:51:00 PM

My question is regarding this "Relay time : Wednesday, March 8th 2017, 21:29:39 -05:00" - Noticed that appearing there last night, is 'relay' meaning it's after 3 days that a miner / node first saw this tx as potentially mineable, or is it the time at which it was 'relayed' back to sender (and would now be sitting in some MP wallet other than my account balance, as the sending user) ..?

Just curious what that entry really means

The "first seen" is self-descriptive, it's the time that blocktrail first saw the transaction. This was went it was first made. The "relay" time is the *last* time it saw it. Wallets like bitcoin core will continually rebroadcast transactions (every 24h?) so that the network doesn't forget about it (most nodes expire mempool transactions after 48-72h). In your case, this is the opposite of what you want (I guess they forgot to `abandontransaction` on it).

The simplest solution would be to use CPFP to push the transaction through. Either you could do that (using the output they sent to) or moneypot could do it (using the change address) and it would confirm with ~30 minutes or so

I personally have attempted to abandon the transaction. The RPC states that the transaction is not eligible for abandonment.

The user above's account has been credited. I believe his question was more out of curiosity.

CrazyCraig

sr. member

Activity: 501

Merit: 340

Bye Felisha!

Quote from: Pseudo-Random on March 09, 2017, 03:51:00 PM

Ah

I'm gonna look a bit like a fool here since I don't know after all this time the details of how tx failure is handled on the network ; But I thought transactions were dropped from mempool / returned to sender after 3 days - it's been 90 hours now and still trying to confirm this mistake..?

My question is regarding this "Relay time : Wednesday, March 8th 2017, 21:29:39 -05:00" - Noticed that appearing there last night, is 'relay' meaning it's after 3 days that a miner / node first saw this tx as potentially mineable, or is it the time at which it was 'relayed' back to sender (and would now be sitting in some MP wallet other than my account balance, as the sending user) ..?

Just curious what that entry really means

Eventually, miners will drop it from their unconfirmed backlog and the transaction will remain spent in our wallet. If we wish to use these funds, we would need to create a double spend.

The fund will not be re-credited to the address owner on MP's system if that is what you are asking?

Pseudo-Random

member

Activity: 80

Merit: 27

A dream that you had, about being a person.

Ah

I'm gonna look a bit like a fool here since I don't know after all this time the details of how tx failure is handled on the network ; But I thought transactions were dropped from mempool / returned to sender after 3 days - it's been 90 hours now and still trying to confirm this mistake..?

My question is regarding this "Relay time : Wednesday, March 8th 2017, 21:29:39 -05:00" - Noticed that appearing there last night, is 'relay' meaning it's after 3 days that a miner / node first saw this tx as potentially mineable, or is it the time at which it was 'relayed' back to sender (and would now be sitting in some MP wallet other than my account balance, as the sending user) ..?

Just curious what that entry really means

CrazyCraig

sr. member

Activity: 501

Merit: 340

Bye Felisha!

Quote from: superbit on March 08, 2017, 09:15:17 PM

How long for a withdrawal to be processed?

Your request exceeded the hot wallet. We once the top off confirms we will send your transacation.

superbit

hero member

Activity: 763

Merit: 500

How long for a withdrawal to be processed?

BITDV

legendary

Activity: 1400

Merit: 1001

Undeads.com - P2E Runner Game

With the 1 BILLION'th Game on MoneyPot

BTCDragon Casino has added its own additional Promo

https://btcdragon.com

staxxd

full member

Activity: 172

Merit: 100

'Bout It.

Quote from: Dogedigital on February 28, 2017, 09:27:38 AM

Good luck guys.

Hey!! Getting pretty close to ONE BILLION! Hard to keep up but at around 992,636,382 & counting fast. Just wanted to say thanks for the wonderful promotion once again and wish everyone good luck. *Since I finished typing this we are up to bet ID 992640610 Grin

Also - Wonderful update on Slot & Go Poker for Bit-Exo, J_Roc - Lots of fun. A very creative & well-made game my friend.

Pseudo-Random

member

Activity: 80

Merit: 27

A dream that you had, about being a person.

Quote from: CrazyCraig on March 05, 2017, 11:50:21 PM

The inputs are not feasible to spend. We have had several transactions that we have paid very large transaction fees on. We have had a few that were .1 for .005 transactions.

That's such a waste of money to miners, the bitcoin protocol allows so easily. Ridiculous. No way ready for anything remotely close to business mainstream adoption. I'd be beyond enraged even as an entrepreneur if I spent $130 fees for $6 items - that's completely past 'cost of doing business'.

JackpotRacer

legendary

Activity: 1974

Merit: 1014

All Games incl Racer and Lottery game are Closed

everyone is invited to give his vote to our MP related Poll

https://bitcointalksearch.org/topic/m.18072174

CrazyCraig

sr. member

Activity: 501

Merit: 340

Bye Felisha!

Quote from: NLNico on March 05, 2017, 11:32:07 PM

Quote from: CrazyCraig on March 05, 2017, 11:07:48 PM

.. for the time being we capped fees as it was generating large unnecessary fees.

I don't really understand this. The fees are based on (kilo)bytes and bitcoind should be able to calculate this properly. Therefor it would be large necessary fees Tongue

Right now it seems players make withdrawals that will never be confirmed. That is only annoying for your real players and also for you, as you will have to manually verify/credit them again after 72h.

If the micro deposits didn't get credited and the inputs are feasible to spend (which it indeed is), you would not lose money even if you have large fees? Again, might be better to make 1 big transaction. Or like RHavar said, lock those outputs and make 1 big transaction when mempool is low.

Either way, temporarily stopping withdrawals and fixing the problem (by locking inputs either by that code commit or manually) seems better than sending withdrawals with way too low fee IMO Tongue

The inputs are not feasible to spend. We have had several transactions that we have paid very large transaction fees on. We have had a few that were .1 for .005 transactions.

NLNico

legendary

Activity: 1876

Merit: 1295

DiceSites.com owner

Quote from: CrazyCraig on March 05, 2017, 11:07:48 PM

.. for the time being we capped fees as it was generating large unnecessary fees.

I don't really understand this. The fees are based on (kilo)bytes and bitcoind should be able to calculate this properly. Therefor it would be large necessary fees Tongue

Right now it seems players make withdrawals that will never be confirmed. That is only annoying for your real players and also for you, as you will have to manually verify/credit them again after 72h.

If the micro deposits didn't get credited and the inputs are feasible to spend (which it indeed is), you would not lose money even if you have large fees? Again, might be better to make 1 big transaction. Or like RHavar said, lock those outputs and make 1 big transaction when mempool is low.

Either way, temporarily stopping withdrawals and fixing the problem (by locking inputs either by that code commit or manually) seems better than sending withdrawals with way too low fee IMO Tongue

Quickseller

copper member

Activity: 2996

Merit: 2374

It looks like some of the inputs from the transaction posted above have only one transaction ever sent to it. Did the attacker have you generate a bunch of deposit addresses and then deposit very small amounts to each of them? Or how did the attacker get so many of your addresses?

If it is fairly clear that a single customer was behind the attack, and his account was not credited anything from the spam deposits, then I would suggest that you simply make a large transaction with a market fee (according to https://bitcoinfees.21.co/ or some other site), and use the money as you wish.

If however the spam deposits went to addresses that appear to be associated with random accounts, then I would do something to make bitcoind ignore these inputs, as they were very likely intended to break your customer's privacy. If they were guessing that 100 addresses belong to MP, knew with certainty that 25 of the addresses belonged to MP, then if you sign a transaction with a private key to an address that the attacker both knew belonged to MP, and that he guessed to belong to MP, then the attacker will be able to use blockchain analysis to discover more of your addresses and will know when some of your customers have deposited into MP, a fact that some customers may have wanted to keep secret.

CrazyCraig

sr. member

Activity: 501

Merit: 340

Bye Felisha!

Quote from: RHavar on March 05, 2017, 10:52:42 PM

Quote from: CrazyCraig on March 05, 2017, 10:32:52 PM

From a technical standpoint, you are correct. Earlier this evening we were subjected to a wallet based attack where the attacker had sent thounsands of micro transactions. Because of this we had to cap the fees and implement other features to resolve the matter.

Capping withdrawal fees doesn't help the problem, it'll just impact your customers withdrawals. The easiest fix is just run `bitcoin-cli lockunspent`on the outputs, so you don't waste money spending them for the time being. Or you can patch your bitcoin with this: https://github.com/RHavar/bitcoin/commit/5ef4bdb3cb1f23bea9ff7bcaa5f7acf5b0344f36

And it'll stop spending those tiny transactions. Then during saturdays there tends to be almost no fee pressure, so you can can create a big transaction at a fee rate of 10sat/byte the cleans it up.

Thanks for the help Ryan. We will look into your suggestions further.

It is annoying correct, for the time being we capped fees as it was generating large unnecessary fees.

RHavar

legendary

Activity: 1463

Merit: 1886

Quote from: NLNico on March 05, 2017, 10:38:23 PM

But if it's still feasible to use those inputs (related to fee it costs to have 1 more input), you could just use the normal fees? Assuming those micro-transactions didn't get credited to any account.

All of those transactions seem standard, which means that the cost of spending them can be < 30% of the amount. So if they're not being credited to an account, then it's just free money. If they're being credited to account(s) it's annoying because there's no deposit fee but each of the deposits has a cost associated with it.

Looking at the outputs that Pseudo-Random posted, they seems to just be faucet payouts from MoonBit.co.in

RHavar

legendary

Activity: 1463

Merit: 1886

Quote from: CrazyCraig on March 05, 2017, 10:32:52 PM

From a technical standpoint, you are correct. Earlier this evening we were subjected to a wallet based attack where the attacker had sent thounsands of micro transactions. Because of this we had to cap the fees and implement other features to resolve the matter.

Capping withdrawal fees doesn't help the problem, it'll just impact your customers withdrawals. The easiest fix is just run `bitcoin-cli lockunspent`on the outputs, so you don't waste money spending them for the time being. Or you can patch your bitcoin with this: https://github.com/RHavar/bitcoin/commit/5ef4bdb3cb1f23bea9ff7bcaa5f7acf5b0344f36

And it'll stop spending those tiny transactions. Then during saturdays there tends to be almost no fee pressure, so you can can create a big transaction at a fee rate of 10sat/byte the cleans it up.

CrazyCraig

sr. member

Activity: 501

Merit: 340

Bye Felisha!

Quote from: NLNico on March 05, 2017, 10:38:23 PM

TBH that sounds like a strange solution (although I can imagine dealing with such attack is annoying.)

But if it's still feasible to use those inputs (related to fee it costs to have 1 more input), you could just use the normal fees? Assuming those micro-transactions didn't get credited to any account.

Then again, if it is an attack to track your site or to break privacy of your players (not sure?), then it might be best to manually disable withdrawals first and make 1 huge TX (with big fee) moving all of those inputs to 1 address?

I want to state that have not been breached.

Defaulting to the maximum fee will result in transactions encompassing huge fees such as .1 btc. We are still working a solution.

NLNico

legendary

Activity: 1876

Merit: 1295

DiceSites.com owner

TBH that sounds like a strange solution (although I can imagine dealing with such attack is annoying.)

But if it's still feasible to use those inputs (related to fee it costs to have 1 more input), you could just use the normal fees? Assuming those micro-transactions didn't get credited to any account.

Then again, if it is an attack to track your site or to break privacy of your players (not sure?), then it might be best to manually disable withdrawals first and make 1 huge TX (with big fee) moving all of those inputs to 1 address? (or even just donate to miners :X)

CrazyCraig

sr. member

Activity: 501

Merit: 340

Bye Felisha!

Quote from: Pseudo-Random on March 05, 2017, 10:26:07 PM

Hey there,

Well it seems MP's BitcoinD/whatever depositor & withdrawal system got majorly code-farted by either a Cloudbleed side effect, or from the constant slow-bets DDoS issues this weekend, or something else - because multiple players are reporting confirmed deposits not debiting to their MP wallet ; while this pure gem of a giant withdrawal monstrosity somehow got created ->

Not 2, 3 or 6 but 42 small inputs, turning a TXiD into 28 times the normal size ; thus leaving even such an expensive .001BTC fee with nevertheless no chance at all of getting this 6200 byte monster included in a block before 2018 ; Since unfortunately the mempool has been clogged on a permanent basis for several weeks - and at ~15.6 satoshi per byte, that's 5-13 times too little to get a confirmation in at least ~30-48hrs (~1-2 dozen block delays later on average)

So is it possible any of the 42 input addresses^ in this mistake would be able to perform CPFP or RBF (debit more fees from my account, np), or, if not, when this thing gets dropped from mempool and returned to all those little MP multi-wallet addresses, can this amount be credited to my account? Thanks Wink

From a technical standpoint, you are correct. Earlier this evening we were subjected to a possible wallet based attack where we received thousands of micro transactions. Because of this we had to cap the fees and implement other features to resolve the matter.

We are still addressing the issue. If any transfer fails we will investigate the matter and credit your account.

Pseudo-Random

member

Activity: 80

Merit: 27

A dream that you had, about being a person.

Hey there,

Well it seems MP's BitcoinD/whatever depositor & withdrawal system got majorly code-farted by either a Cloudbleed side effect, or from the constant slow-bets DDoS issues this weekend, or something else - because multiple players are reporting confirmed deposits not debiting to their MP wallet ; while this pure gem of a giant withdrawal monstrosity somehow got created ->

Not 2, 3 or 6 but 42 small inputs, turning a TXiD into 28 times the normal size ; thus leaving even such an expensive .001BTC fee with nevertheless no chance at all of getting this 6200 byte monster included in a block before 2018 ; Since unfortunately the mempool has been clogged on a permanent basis for several weeks - and at ~15.6 satoshi per byte, that's 5-13 times too little to get a confirmation in at least ~30-48hrs (~1-2 dozen block delays later on average)

So is it possible any of the 42 input addresses^ in this mistake would be able to perform CPFP or RBF (debit more fees from my account, np), or, if not, when this thing gets dropped from mempool and returned to all those little MP multi-wallet addresses, can this amount be credited to my account? Thanks Wink

superiorus

legendary

Activity: 1358

Merit: 1000

Bitcoiner since start, and continue to love it!

Quote from: ralle14 on March 05, 2017, 05:23:50 AM

Anyone know what happened to the moneypot app bubblesbit? My browser isn't letting me go through the site because it says my connection isn't private. You guys experience the same thing as me?

It seems just a SSL error ...
You can contact the owner (https://bitcointalksearch.org/user/yahoo62278-355846) and I'm sure they will fix it.

Topic: 🌟🔴🌟🎲🌟🔴🌟MONEYPOT.com -- Crypto Casino Suite and Web Wallet 🌟🔴🌟🎲🌟🔴🌟 - page 15. (Read 30196 times)