|
|
|
|
|
|
|
The Bitcoin network protocol was designed to be extremely flexible. It can be used to create timed transactions, escrow transactions, multi-signature transactions, etc. The current features of the client only hint at what will be possible in the future.
|
|
|
Advertised sites are not endorsed by the Bitcoin Forum. They may be unsafe, untrustworthy, or illegal in your jurisdiction.
|
|
|
|
|
|
PrintCoins
|
 |
February 10, 2012, 12:36:30 AM |
|
Doesn't that mean you gave it to the miner?
|
|
|
|
koin
Legendary
 Offline
Activity: 873
Merit: 1000
|
|
February 10, 2012, 12:41:27 AM |
|
before anyone misunderstands ... this is on testnet and not the bitcoin blockchain
|
|
|
|
|
grue
Legendary
Offline
Activity: 2058
Merit: 1431
|
|
February 10, 2012, 12:46:39 AM |
|
lol  |
|
|
|
Night (OP)
Newbie
Offline
Activity: 43
Merit: 0
|
|
February 10, 2012, 12:47:24 AM |
|
What is the difference of the Bitcoin and Bitcoin TestNet? Technically, nothing! So it can happen in Bitcoin
|
|
|
|
|
Littleshop
Legendary
Offline
Activity: 1386
Merit: 1003
|
|
February 10, 2012, 01:08:29 AM |
|
|
|
|
|
|
bittenbob
|
|
February 10, 2012, 01:28:51 AM |
|
Are they testing out a new BIP? This could be due to a bug in a BIP someone is trying to implement and why I am an advocate of an excessive level of testing on testnet before it ever makes it to the real bitcoind.
|
|
|
|
|
|
ineededausername
|
|
February 10, 2012, 01:31:59 AM |
|
negative balance.. oh dear
|
(BFL)^2 < 0
|
|
|
koin
Legendary
Offline
Activity: 873
Merit: 1000
|
|
February 10, 2012, 01:42:13 AM
Last edit: February 10, 2012, 01:55:25 AM by koin |
|
bip -1
to compete with the banking system, bitcoin now supports overdrafts. any overdraft will cost you in fees enough to pad a new bonus system to overly compensate the bitcoin executives at the top.
seriously though ..., this is a problem somewhere, right? (but is isolated to whatever code blockexplorer is using for testnet only, right?)
|
|
|
|
|
Gavin Andresen
Legendary
Offline
Activity: 1652
Merit: 2216
Chief Scientist
|
|
February 10, 2012, 02:27:33 AM |
|
Don't panic.
roconnor has been experimenting with spending "duplicate coinbase transactions" on the testnet. Block explorer is confused, and isn't seeing the 50 bitcoins generated to that address in testnet block 45,442 (because the generation transaction has the same ID as the generation transaction in block 45,333).
Expect code changes before the 0.6 release is final to discourage and eventually prohibit accepting blocks with duplicate coinbase transactions; although we can't see a way to exploit this weirdness to cheat anybody (it is easy to cheat yourself out of bitcoins using duplicate coinbase transactions that cannot be spent), it's definitely not a good thing.
|
How often do you get the chance to work on a potentially world-changing project?
|
|
|
|
finway
|
|
February 10, 2012, 02:59:03 AM |
|
No worries.
|
|
|
|
theymos
Administrator
Legendary
Offline
Activity: 5180
Merit: 12880
|
|
February 10, 2012, 02:59:47 AM |
|
Notice that tx 0018417e23... and fae8bfd6a2... on that address ledger spend the exact same output. It looks like a double-spend, but it's actually a less severe bug in the protocol that probably doesn't allow for any easy attacks. (Because the bug causes very strange behavior, though, it might allow for attacks that have not been thought of yet.)
These transactions actually broke Bitcoin Block Explorer when they first appeared because there was a database constraint preventing double-spends from being inserted into the database. I "fixed" it by removing the constraint, but this causes BBE to display these transactions strangely. A lot of code would need to be changed, but I might fix it fully some day.
|
1NXYoJ5xU91Jp83XfVMHwwTUyZFK64BoAD
|
|
|
ThiagoCMC
Legendary
Offline
Activity: 1204
Merit: 1000
฿itcoin: Currency of Resistance!
|
|
February 10, 2012, 03:04:12 AM |
|
Good job Gavin!! Testnet is... FOR TEST!!!  |
|
|
|
|
|
ineededausername
|
|
February 10, 2012, 03:56:49 AM |
|
Notice that tx 0018417e23... and fae8bfd6a2... on that address ledger spend the exact same output. It looks like a double-spend, but it's actually a less severe bug in the protocol that probably doesn't allow for any easy attacks. (Because the bug causes very strange behavior, though, it might allow for attacks that have not been thought of yet.)
That sounds pretty terrible.  |
(BFL)^2 < 0
|
|
|
theymos
Administrator
Legendary
Offline
Activity: 5180
Merit: 12880
|
|
February 10, 2012, 04:49:02 AM |
|
Technical explanation:
When you create a block, the coinbase transaction which generates 50 BTC can easily be made to exactly equal a past coinbase transaction: it just needs to pay out to the same address and have the same (arbitrary) scriptSig. Unfortunately, this transaction will then have the exact same hash as another transaction in the chain. Satoshi did not anticipate this, so it's allowed by the protocol and it confuses the Bitcoin code.
In this case, roconnor:
- Generated a block containing a normal coinbase transaction.
- Spend the BTC generated in the coinbase transaction. Bitcoin nodes mark the coinbase transaction's output as spent.
- Created a block containing a duplicate of the first coinbase transaction. Bitcoin nodes will now overwrite the first coinbase transaction in the transaction database. The output of the coinbase transaction is marked as unspent. The transaction spending the original coinbase output is not affected and is still considered valid.
- Spent the BTC generated in the second coinbase. There are now two transactions spending the "same" output.
|
1NXYoJ5xU91Jp83XfVMHwwTUyZFK64BoAD
|
|
|
FreeMoney
Legendary
Offline
Activity: 1246
Merit: 1014
Strength in numbers
|
|
February 10, 2012, 05:05:48 AM |
|
Technical explanation:
When you create a block, the coinbase transaction which generates 50 BTC can easily be made to exactly equal a past coinbase transaction: it just needs to pay out to the same address and have the same (arbitrary) scriptSig. Unfortunately, this transaction will then have the exact same hash as another transaction in the chain. Satoshi did not anticipate this, so it's allowed by the protocol and it confuses the Bitcoin code.
In this case, roconnor:
- Generated a block containing a normal coinbase transaction.
- Spend the BTC generated in the coinbase transaction. Bitcoin nodes mark the coinbase transaction's output as spent.
- Created a block containing a duplicate of the first coinbase transaction. Bitcoin nodes will now overwrite the first coinbase transaction in the transaction database. The output of the coinbase transaction is marked as unspent. The transaction spending the original coinbase output is not affected and is still considered valid.
- Spent the BTC generated in the second coinbase. There are now two transactions spending the "same" output.
Do you lose the original if it isn't spent when you make the new identical one? Because when you spend either now it'll be marked as spent? |
Play Bitcoin Poker at sealswithclubs.eu. We're active and open to everyone. |
|
|
theymos
Administrator
Legendary
Offline
Activity: 5180
Merit: 12880
|
|
February 10, 2012, 05:12:11 AM |
|
Do you lose the original if it isn't spent when you make the new identical one? Because when you spend either now it'll be marked as spent?
Right. This has happened a few times on the main network, actually. Not sure if anyone realized until recently that it was possible to spend both, though. |
1NXYoJ5xU91Jp83XfVMHwwTUyZFK64BoAD
|
|
|
ThiagoCMC
Legendary
Offline
Activity: 1204
Merit: 1000
฿itcoin: Currency of Resistance!
|
|
February 10, 2012, 05:20:43 AM |
|
Do you lose the original if it isn't spent when you make the new identical one? Because when you spend either now it'll be marked as spent?
Right. This has happened a few times on the main network, actually. Not sure if anyone realized until recently that it was possible to spend both, though. Anyway, now this guys will solve a bug! |
|
|
|
|
|
payb.tc
|
|
February 10, 2012, 05:35:32 AM |
|
so to clarify, the main consequence of the bug was that a block reward could become lost forever? still... turning 100 BTC into 50 is less serious of a bug than being able to turn 50 BTC into 100 |
|
|
|
|
ThiagoCMC
Legendary
Offline
Activity: 1204
Merit: 1000
฿itcoin: Currency of Resistance!
|
|
February 10, 2012, 05:41:34 AM |
|
so to clarify, the main consequence of the bug was that a block reward could become lost forever? still... turning 100 BTC into 50 is less serious of a bug than being able to turn 50 BTC into 100 Well, I think that ANY overwrite is serious... But you got a point! |
|
|
|
|
|
