What just happened?

[wtf]

Ok, so my balance just jumped down.
I look in my bitcoin client and see this:
http://imageshack.us/m/199/8594/unledxk.jpg

Yes, those transactions (received by me) went from confirmed to unconfirmed but they're still sorted as if they were confirmed. They are now back in the queue.
Happened right after 2 new blocks within a minute appeared, the 2nd of those containing over 27000 BTC. This looks bad..

[kiba]

That must be a block race?

[ArsenShnurkov]

wait for some more confirmations (might be the client chooses the longest chain)

[Tyran]

I just had a transaction roll back too. I guess it just happens occasionally when 2 blocks are found at almost the same time, and your client initially accepted the "wrong" one.
The massive amount of BTC in that block kind of worries me though, I would like to see the data for the block that got invalidated.

[bitlotto]

How many confirmations did it get before it changed?

[Cryptoman]

Block chain split?

[Tyran]

How many confirmations did it get before it changed?

I wasn't watching when it happened.
Maybe someone else can tell:

SetBestChain: height=123555
AddToWallet xxxxxxxxxx new
AddToWallet xxxxxxxxxx update
SetBestChain: height=123556
REORGANIZE
SetBestChain: height=123557

Then another AddToWallet update right before the block that confirmed it the 2nd time.
The log has no timestamps so I really can't tell if it was 1 or 2 confirmations. Probably just one?

[mjsbuddha]

deepbit just had its first invalid block. don't know if its the same block or not.

[FooDSt4mP]

I feel like I should be selling right now because of this thread...

Please do.

[FreeMoney]

Yes, those transactions (received by me) went from confirmed to unconfirmed but they're still sorted as if they were confirmed. They are now back in the queue.
Happened right after 2 new blocks within a minute appeared, the 2nd of those containing over 27000 BTC. This looks bad..

Did they go from 1/unconfirmed to 0 or from like 5 and 6?

[Mike Hearn]

It'd help to quote block hashes rather than numbers here, for obvious reasons.

I assume you guys are talking about these two blocks:

123561 (1,159940732) 0000000000005ce5229e  05/13/11 03:15:05  tx 93
123562 (1,159983756) 0000000000006680ec6b  05/13/11 03:16:15  tx 9

My node hasn't seen any splits longer than a single block for a long time. You can use the bitcoind -printblocktree command to find out if your node did. I also don't see any references to transactions that failed to connect.

[Tyran]

It'd help to quote block hashes rather than numbers here, for obvious reasons.

I assume you guys are talking about these two blocks:

123561 (1,159940732) 0000000000005ce5229e  05/13/11 03:15:05  tx 93
123562 (1,159983756) 0000000000006680ec6b  05/13/11 03:16:15  tx 9

My node hasn't seen any splits longer than a single block for a long time. You can use the bitcoind -printblocktree command to find out if your node did. I also don't see any references to transactions that failed to connect.

Code:

123555 (1,159628576) 00000000000030e507bd  05/13/11 01:09:48  tx 24
|\
| 123556 (1,159637580) 0000000000000202ab51  05/13/11 01:24:22  tx 115
|
123556 (1,159675761) 0000000000006a1a6624  05/13/11 01:24:28  tx 19
123557 (1,159683842) 00000000000038a59f89  05/13/11 02:24:26  tx 73

Nothing to see here

[njloof]

So to recap, if I may:

- most likely no malice or funny business afoot
- two miners found alternate solutions within a short period of time, too short for the entire network to reach consensus
- consensus is restored in the blocks that follow

Agreed?

[mewantsbitcoins]

Balance in the universe is restored 

[sniper_sniperson]

Block chain split?

I think someone from bitcoin network tried to divide by zero ... 

[error]

Yes, this happened last night. Ironically it was the Deepbit pool that got its block thrown out.