Title: Namecoin appears to be under a 51% attack (80% of the hashpower) Post by: biolizard89 on August 15, 2013, 04:30:38 PM Here's a chatlog from the #namecoin channel... couldn't find anything about this on the forums already, sorry if I missed something:
Quote [2013-08-15 10:19:28] <Jeremy_Rand> okay, so the last 6 blocks have had no transactions in them... anyone know why that might be happening? [2013-08-15 10:20:11] <c0rw1n> no idea, maybe try to message DrHaribo, he should be on freenode and operates Bitminter [2013-08-15 10:20:29] <shader> are there no transactions waiting? [2013-08-15 10:21:11] <Jeremy_Rand> it actually seems to be fairly common... looking at the explorer, it's common for there to be 6 or so blocks with nothing, and then like 20-30 transactions in the next block [2013-08-15 10:21:27] <Jeremy_Rand> I'm pretty sure that's statistically not by random chance [2013-08-15 10:21:29] <shader> so, some miners aren't doing anything? [2013-08-15 10:21:47] <Jeremy_Rand> that's what it looks like to me [2013-08-15 10:21:51] <Jeremy_Rand> but I'm not an expert [2013-08-15 10:22:38] <c0rw1n> well i know i'm mining namecoin, but i have no idea what my pool operator's doing [2013-08-15 10:22:51] <Jeremy_Rand> ok, my transaction just got into a block about 10 seconds ago [2013-08-15 10:22:57] <Jeremy_Rand> even though it was sent an hour ago [2013-08-15 10:23:11] <Jeremy_Rand> and there were 158 tx's in this new block [2013-08-15 10:23:17] <Jeremy_Rand> so a lot of people were waiting [2013-08-15 10:23:32] <shader> hmm [2013-08-15 10:23:36] <shader> that doesn't sound good [2013-08-15 10:23:47] <Jeremy_Rand> agreed [2013-08-15 10:24:00] <shader> do you think it's possible to track down which miners were responsible for the last few blocks? [2013-08-15 10:24:14] <Jeremy_Rand> no idea [2013-08-15 10:25:11] <Jeremy_Rand> I can see their 50NMC tx's to themselves... so we could probably track down the miners via blockchain analysis that way [2013-08-15 10:25:15] <Jeremy_Rand> but that sounds like a lot of work [2013-08-15 10:26:46] <shader> that one miner seems to get a ton of the blocks, and doesn't do any transactions... [2013-08-15 10:26:55] <shader> N4BbUHfarRXx8GBNj9DHAqthhuvxsKpnZE [2013-08-15 10:27:04] <c0rw1n> wohoha okay, i wanted to see if my pool had minted that 158 tx block, and nope : we're at 100M PoW and 99.25% CDF, mining for 7 hours [2013-08-15 10:27:24] <shader> lol [2013-08-15 10:31:55] |<-- knotwork has left freenode (Read error: Connection timed out) [2013-08-15 10:32:25] -->| knotwork (~markm@unaffiliated/knotwork) has joined #namecoin [2013-08-15 10:47:51] -->| SwiftDK (redacted) has joined #namecoin [2013-08-15 10:53:16] <Jeremy_Rand> of the last 15 blocks, only 3 have any transactions [2013-08-15 10:54:25] |<-- X-Rob has left freenode (Read error: Connection reset by peer) [2013-08-15 10:54:35] -->| X-Rob (~Rob7@unaffiliated/x-rob) has joined #namecoin [2013-08-15 10:55:30] -->| domob (redacted) has joined #namecoin [2013-08-15 10:55:50] <Jeremy_Rand> and all 12 of the empty blocks are by the same miner [2013-08-15 10:56:01] <Jeremy_Rand> I believe that would make this a 51% attack? [2013-08-15 10:58:43] <Jeremy_Rand> is there any way to track down the miner's identity from his payment address? I'm wondering if it's a pool or a freelancer [2013-08-15 10:59:14] <Jeremy_Rand> whoever it is probably has the ability to reverse transactions / steal names if I understand the implications of a 51% attack [2013-08-15 11:05:51] <c0rw1n> ah but [2013-08-15 11:06:00] <c0rw1n> only the current transactions [2013-08-15 11:06:17] <c0rw1n> if he wants to screw old transactions, he has to rehash each of the blocks [2013-08-15 11:06:30] <c0rw1n> AND keep up [2013-08-15 11:06:44] <Jeremy_Rand> he has 80% of the hashpower, that sounds quite doable for names that were registered recently [2013-08-15 11:06:57] <c0rw1n> he has waitwut ? [2013-08-15 11:07:18] <Jeremy_Rand> 12 of the last 15 blocks have their mining payout to the same Namecoin address, i.e. the same miner [2013-08-15 11:08:53] <Jeremy_Rand> I think this is a Bad Thing (TM) [2013-08-15 11:09:28] <c0rw1n> i hope it's a bug in a pool and will get fixed soon [2013-08-15 11:09:29] <c0rw1n> oh yes it is [2013-08-15 11:10:01] <c0rw1n> namecoin is a solution to so many problems [2013-08-15 11:10:46] <Jeremy_Rand> I wonder how long this attack has been going on... anyone want to look through the block explorer and see how long he's had this much of the hashpower? [2013-08-15 11:12:33] |<-- c0rw1n has left freenode (Remote host closed the connection) [2013-08-15 11:15:32] <mhanne> received 6609 payouts since 12/06/13 02:26:56 (didn't check if all are generations but the vast majority seems to be 50nmc exactly) [2013-08-15 11:16:15] |<-- domob has left freenode (Ping timeout: 276 seconds) [2013-08-15 11:18:14] -->| domob (redacted) has joined #namecoin [2013-08-15 11:19:19] <shader> makes you wonder what they'd want to do it for... I mean, it's not really a very logical attack [2013-08-15 11:19:45] <shader> most likely they set up merged mining incorrectly or something [2013-08-15 11:19:48] <Jeremy_Rand> crap, so over the past 9 weeks he's averaged 71% of the hashpower [2013-08-15 11:19:59] <shader> ... [2013-08-15 11:20:30] <shader> and the difficulty isn't low either, so they must be one of the mining pools [2013-08-15 11:21:06] <Jeremy_Rand> if it's a pool, then one of the pool's users should be able to trace his 50NMC generation to their payout, right? [2013-08-15 11:22:00] <shader> true [2013-08-15 11:22:17] <shader> are there any tools for tracing transactions? Title: Re: Namecoin appears to be under a 51% attack (80% of the hashpower) Post by: gigawatt on August 15, 2013, 05:10:15 PM Don't a lot of pools co-mine BTC and NMC? It seems like you'd need an ungodly amount of hashing power to pull a 51% attack. You'd have to outgun entire pools.
Title: Re: Namecoin appears to be under a 51% attack (80% of the hashpower) Post by: biolizard89 on August 15, 2013, 05:40:18 PM Don't a lot of pools co-mine BTC and NMC? It seems like you'd need an ungodly amount of hashing power to pull a 51% attack. You'd have to outgun entire pools. Yes, that's why it seems likely that a pool is doing it.Title: Re: Namecoin appears to be under a 51% attack (80% of the hashpower) Post by: xorxor on August 15, 2013, 06:04:49 PM That is the main problem of mergemining. It is so easy for pool operators to 51% it that it has to happen.
Title: Re: Namecoin appears to be under a 51% attack (80% of the hashpower) Post by: Vivisector999 on August 15, 2013, 08:22:52 PM My guess is that tremor in the force is the BTC-Guild pool, I think it's been about 2 months or so since they started merge mining NMC.
Title: Re: Namecoin appears to be under a 51% attack (80% of the hashpower) Post by: phelix on August 16, 2013, 10:11:30 AM This is indeed BTC-Guild. Their transaction selection is not working as it should, often dropping transactions. This has been confirmed by Eleuthria. I will ask him to fix it once we have released the next, hopefully more stable version of Namecoin.
Snailbrain & co are currently working on some smaller updates and fixes for the new Namecoin-Qt. After that we will do some more testing, probably come up with a release candidate. So the next official release can be expected within a month or so. |