2014/08/04 Attack: Update 1
Here's what we've ascertained thus far: the "bad block" occurred at 202612. Immediately thereafter there was 202613, which also has oddities, and then at 202614 the network forked. The "A" part of the network has "c29e3dc37d8da3e72e506e31a213a58771b24450144305bcba9e70fa4d6ea6fb" as the block hash on 202614, the "B" part of the network has "ed4eea6109a1b662cf4a3bb372ed4bdee588160b0ac371c2ad78c5e603b8f2ac" as the block hash on 202614. Ultimately the "A" part won, and is the current fork. The "B" part is dead and has not proceeded beyond block 202647.
Thus the good news is that the consensus code works, and only one fork has survived. The bad news is that those on the bad fork are not reorganising back to the fork, and that is something we will need to look at fixing in future - the consensus code is so aggressive that it refuses to believe that the peers on the "good" fork are correct.
To determine if you are on the good or the bad (dead) fork, you can check "diff" in your daemon window. If you are stuck on 202647 or thereabouts, you will need to delete your blockchain and sync from scratch. I strongly suggest using the blockchain bootstrap downloads from the OP, as those are on the correct fork. Please note that the Windows blockchain bootstrap download has not been checked and updated as yet, I will be doing so in the next few hours, so we have rolled back to a Windows blockchain from a few weeks back. The Linux and OS X blockchain bootstraps are fine.
The next step is to determine how, exactly, 202612 & 202613 managed to cause this, which is what we've been doing concurrently and will continue to do. You can use Monero with caution if you are on the main chain, we will continue to update as we proceed.
NB: no funds have been lost, and there will be no blockchain rollback. If you transacted on the dead fork it was only live for about 35 minutes, so chances are little happened of relevance. Once you are on the good fork you can re-sync your wallets and any funds you sent on the dead fork will reappear.
Here's what we've ascertained thus far: the "bad block" occurred at 202612. Immediately thereafter there was 202613, which also has oddities, and then at 202614 the network forked. The "A" part of the network has "c29e3dc37d8da3e72e506e31a213a58771b24450144305bcba9e70fa4d6ea6fb" as the block hash on 202614, the "B" part of the network has "ed4eea6109a1b662cf4a3bb372ed4bdee588160b0ac371c2ad78c5e603b8f2ac" as the block hash on 202614. Ultimately the "A" part won, and is the current fork. The "B" part is dead and has not proceeded beyond block 202647.
Thus the good news is that the consensus code works, and only one fork has survived. The bad news is that those on the bad fork are not reorganising back to the fork, and that is something we will need to look at fixing in future - the consensus code is so aggressive that it refuses to believe that the peers on the "good" fork are correct.
To determine if you are on the good or the bad (dead) fork, you can check "diff" in your daemon window. If you are stuck on 202647 or thereabouts, you will need to delete your blockchain and sync from scratch. I strongly suggest using the blockchain bootstrap downloads from the OP, as those are on the correct fork. Please note that the Windows blockchain bootstrap download has not been checked and updated as yet, I will be doing so in the next few hours, so we have rolled back to a Windows blockchain from a few weeks back. The Linux and OS X blockchain bootstraps are fine.
The next step is to determine how, exactly, 202612 & 202613 managed to cause this, which is what we've been doing concurrently and will continue to do. You can use Monero with caution if you are on the main chain, we will continue to update as we proceed.
NB: no funds have been lost, and there will be no blockchain rollback. If you transacted on the dead fork it was only live for about 35 minutes, so chances are little happened of relevance. Once you are on the good fork you can re-sync your wallets and any funds you sent on the dead fork will reappear.
Thanks for the update. Keep up the good work. Hopefully you will be able to fix the issue in short time.
should be XMR ...in hurry??