Once again there was another double spend attempt on the exchange. The exchange closed block chain processing when it happened. As a result of these continued attacks I've had to close the exchange. As it says from the main page: Deposits for I0Coin and Bitcoin have been suspended, along with trading. Due to low network power of the I0Coin chain over the last couple of weeks there have been attempts to do 51% double spend attacks. The exchange would stop processing the block chain when these were detected and users will have noticed this, as deposits/withdrawals were suspended while I investigate. I'd then re-open with a higher deposit confirmation limit. Some of the double spends were successful and a total of 198 bitcoins were successfully taken from the exchange. This involved anonymous users operating over TOR depositing I0Coins, trading for bitcoins, withdrawing the bitcoins and reversing the original deposit. Due to this loss and the risk involved in continuing to operating an exchange for a chain which has, or can easily get, an attacker with >51% of the power, I've closed the exchange. Please withdraw all bitcoins and I0Coins before 31 Oct 2011 12:00 UTC. You can reach the exchange operator at admin@bitparking.com if you have any questions. It's unfortunate that the actions of a few have to stop the exchange usage for many honest users. The basic approach from the attacking user was to deposit about 120,000 I0Coins, sell them, withdraw the BTC then invalidate/remove the original deposit. All the successful attacks (totaling 198 bitcoins or so) were done by a user operating over TOR. The user used accounts with names like: * youdozeek * mooncoin * ploti * adolf * mojimcheo The bitcoin addresses that were withdraw too were: * 13kf6BR32STEuW7FSKVbM3WNUcF6nqcP3q * 1H3Zx5a7fTWnpt8pfZnwekQzQYb4497vWb * 1AKxgg8jESUvooeHtNvaSQYhbPCyv7wYZF * 1G8XACmUXoTF3FaA5bATSD3fnSR2WQXVJH * 1PoUHcuTqoSiKupYzkQUbPHE4SwTPBT9Ht If you were one of these users and don't believe you did a deliberate double spend, please get in touch. That sucks. Sorry to hear that happened. Neither of those chains had the time warp issue fixed did they? Seems proof that needs to be fixed for any chain to be viable. That and maybe also merged mining. I still think the idea of multiple chains is a good one in the long run. |
|
|
|
Is there no way to patch this problem once and for all?
Stop alternate coins from starting?  Basically I'd expect this is happen to all alternate coins. Some sooner than others. Some of them are 51% proof, like GG. GG is not 51% proof.... it is Time Warp bug fixed is all. |
|
|
|
Since that would then make the 0 coin blocks invalid blocks, it is a block forking change, so a relaunch is needed. That's not what I observed. Coined blocks built on top of the existing chain. I haven't mined every block, and have mined several orphans. I can't say for definite, but I guess that some blocks have coins & some don't. The only problem would be someone trying to spend the 0 coins - or it would be a problem if the unexpected outcome was some coins but not 25 per block. We are now at 2000 blocks, so 25% of the chain is after I started finding coins again. Won't be long before it's the majority. Why not just go with it? I'm not precious about this, it just seems that one "#" is way less problematic than a re-write, no? Did you observe that though? Or did you just make that change and keep your already downloaded blocks data file. Delete your blocks file and try again, remember they are not re-validated every time you restart the client. So unless you deleted your already existing block files you did not observe what you thought you did as far as what a fresh client or a re validation of the whole chain would see. |
|
|
|
Unfortunately this launch was a false start. There was a problem in the config causing all blocks mined after the 100th blocks to contain 0 coins. The fix would unfortunately require a restart of the coin. michaelmclees has asked me over IRC to take over for him. I will try to do my best. I will set up a new github.com repository with the source so people can do diffs. And generate a new genesis block. I will premine a few to make sure things work ok. Apparently, that was really needed. When everything is looking good, I will post the client and the source. In the meantime, please stop your clients and miners and switch back to tenebrix. What happened ? I made sure that the custom inflation (aka make magkal blocks with arbitrary coinage) is off in the config (custom_inflation=0) and set subsidy to "allways 25" (Subsidy=25), wtf did go wrong ? The code checks for the flag custom_inflation, so setting it to 0 still triggers custom inflation. Since post_Subsidy is not set, after 100 blocks, the subsidy becomes 0. doh! mmm crap. But why does inflation_trigger default to 100 ? Also, dudes, we can theoretcially fix it without restart, just with a config update. custom_inflation=1 infaltion_trigger=(current block nucmber)+ 5 blocks post_Subsidly=25 The code was changed from the multicoin defaults by someone, in the orginal multicoin the default block when not set is MAX_INT, someone made it 100 in this fork. Why I have not idea, they should have put it in the config file probably instead of putting it in the code. You still need a relaunch, or you need to use the multi step innflation trigger code that is in multicoin to add rules to keep the 0 coin blocks valid. For some reasons this code also removed that though. Anyway, give the skill sets and the fact only 100 blocks matter with coins in them anyway, a relaunch is really the best idea here. |
|
|
|
Simply removing the line in the config is enough to restore expected behaviour. No need for a re-launch. Since that would then make the 0 coin blocks invalid blocks, it is a block forking change, so a relaunch is needed. |
|
|
|
He posted FUD on the nmc exchange. He abused his IRC op privs to squash discussion on that.
By which you mean he kicked you from #namecoin when you were spreading FUD and not listening to everyone else in the channel pointing out your errors. I mean he kicked me, while I was pointing out where your message was FUD still. And your 'everyone' else in the channel was a whole two people , you and ArtForz, and I saw the last message from art that was starting to agree with me as I got kicked by you for no good reason, except that I did not agree with someone with OP privs. I also noticed you removed your FUD not very long there after so you must have even stopped agreeing with it yourself. |
|
|
|
I c where are you going with it Proposal: How about we invite, I dunno, doublec (tho I'm afraid he will refuse) I would say no. Doublec is prone to implusive moves. He shut down sc services on people with zero warning. He posted FUD on the nmc exchange. He abused his IRC op privs to squash discussion on that. |
|
|
|
|
What reason do the exchanges have for posting "It appears that the NMC blockchain has been forked." a day after the few rogue pools that patched their client to start merged mining had abandoned that effort. The fork was for around 10 blocks and was not going to be accepted by anyone running a client downloaded from dot-bit.org or built from the sources linked to from the namecoin projects web site. Why make this fork that no other clients accepted seem any more significant then if I patched namecoin and ran it on a few dozen machines myself. That could happen at any time as well. It would have as little meaning as this flawed idea of an attack defense did.
|
|
|
|
I miss hearing about it, the only good natured blockchain since namecoin (imo).
Beertoken and friends are still putting along same as ever, though sacarlson says they get little interest. |
|
|
|
How come you are the only one supporting windows users with updated clients? I've asked many times on the nmc forum to post an updated windows client,it still has the old client listed. Pathetic of the devs.I wonder if they are even interested saving this project. I can't answer the first question of course. As for the later. I think they are interested, after all we have recently seen the .60, .61 and .62 releases. So they are still working on it. They also put a lot of work into making the merged mining a reality. I just don't think they have anyone available to make the official builds for windows. I happened to already have build environments set up and knew people had to upgrade before merged mining kicked in, and so have tried to fill that gap for now. |
|
|
|
{
"version" : 32462,
"balance" : 0.00000000,
"blocks" : 18951,
"connections" : 9,
"proxy" : "",
"generate" : false,
"genproclimit" : -1,
"difficulty" : 94035.90217415,
"hashespersec" : 0,
"testnet" : false,
"keypoololdest" : 1316093191,
"paytxfee" : 0.00001000,
"errors" : "WARNING: Displayed transactions may not be correct! You may need to upgrade, or other nodes may need to upgrade."
}This is what an attacker can do to the network and they turned me into an attacker. Even deleting everything and starting over gives me this error. Yeah, I am seeing something similar when I try and start a fresh run. Luckily I do have a running instance that seems to be at the correct block count. Hopefully it will clear up soon as the network figures out what to orphan. If you are not back up later today with your namecoind, pm me and I can see what I can do to help. |
|
|
|
I talked to doublec between those posts, seems they lied to you then. He did not change.
BTW, there are more then 3 pools.
OH BOY! This is going to get UGLY. Here are the changes to my client I manually changed 3.24.62 as the code base was better. I ignored the lockins on this diff and used 3.24.62 lockins... diff -ur namecoin/src/namecoin.cpp namecoin-0.3.24.61.18950-mergedmine-REAL/src/src/namecoin.cpp
--- namecoin/src/namecoin.cpp 2011-09-13 15:14:55.000000000 +0200
+++ namecoin-0.3.24.61.18950-mergedmine-REAL/src/src/namecoin.cpp 2011-09-13 12:29:00.000000000 +0200
@@ -95,7 +95,8 @@
{
if (fTestNet)
return 0;
- return 19200;
+// return 19200;
+ return 18950;
}
string GetAlertPubkey1()
@@ -1685,11 +1686,17 @@
int CNamecoinHooks::LockinHeight()
{
- return 0;
+ if (fTestNet)
+ return 0;
+
+ return 18938;
}
bool CNamecoinHooks::Lockin(int nHeight, uint256 hash)
{
+ if (!fTestNet)
+ if ((nHeight == 18938 && hash != uint256("0x0000000000001610e3ea22652e56fe457f207ee49deb7cbb3035a885cf4d61b1")))
+ return false;
return true;
}
Only in namecoin-0.3.24.61.18950-mergedmine-REAL/src/src: namecoind
diff -ur namecoin/src/rpc.cpp namecoin-0.3.24.61.18950-mergedmine-REAL/src/src/rpc.cpp
--- namecoin/src/rpc.cpp 2011-09-13 15:14:55.000000000 +0200
+++ namecoin-0.3.24.61.18950-mergedmine-REAL/src/src/rpc.cpp 2011-09-13 12:29:00.000000000 +0200
@@ -454,6 +454,7 @@
obj.push_back(Pair("keypoololdest", (boost::int64_t)pwalletMain->GetOldestKeyPoolTime()));
obj.push_back(Pair("paytxfee", ValueFromAmount(nTransactionFee)));
obj.push_back(Pair("errors", GetWarnings("statusbar")));
+ obj.push_back(Pair("merged-mine", 18950 ));
return obj;
}
So where do we go from here? Either go back to being on the official rules, or stay rogue but realize you are now rogue. btw, what was the actual block count at when you started to run this patch? No expert in this but pretty sure I can read a patch from this bit "obj.push_back(Pair("merged-mine", 18950 ));" I'll go with 18950... that is when the code said to make the swtich to mm, I am asking what was the block count at the time they put that code live. |
|
|
|
I talked to doublec between those posts, seems they lied to you then. He did not change.
BTW, there are more then 3 pools.
OH BOY! This is going to get UGLY. Here are the changes to my client I manually changed 3.24.62 as the code base was better. I ignored the lockins on this diff and used 3.24.62 lockins... diff -ur namecoin/src/namecoin.cpp namecoin-0.3.24.61.18950-mergedmine-REAL/src/src/namecoin.cpp
--- namecoin/src/namecoin.cpp 2011-09-13 15:14:55.000000000 +0200
+++ namecoin-0.3.24.61.18950-mergedmine-REAL/src/src/namecoin.cpp 2011-09-13 12:29:00.000000000 +0200
@@ -95,7 +95,8 @@
{
if (fTestNet)
return 0;
- return 19200;
+// return 19200;
+ return 18950;
}
string GetAlertPubkey1()
@@ -1685,11 +1686,17 @@
int CNamecoinHooks::LockinHeight()
{
- return 0;
+ if (fTestNet)
+ return 0;
+
+ return 18938;
}
bool CNamecoinHooks::Lockin(int nHeight, uint256 hash)
{
+ if (!fTestNet)
+ if ((nHeight == 18938 && hash != uint256("0x0000000000001610e3ea22652e56fe457f207ee49deb7cbb3035a885cf4d61b1")))
+ return false;
return true;
}
Only in namecoin-0.3.24.61.18950-mergedmine-REAL/src/src: namecoind
diff -ur namecoin/src/rpc.cpp namecoin-0.3.24.61.18950-mergedmine-REAL/src/src/rpc.cpp
--- namecoin/src/rpc.cpp 2011-09-13 15:14:55.000000000 +0200
+++ namecoin-0.3.24.61.18950-mergedmine-REAL/src/src/rpc.cpp 2011-09-13 12:29:00.000000000 +0200
@@ -454,6 +454,7 @@
obj.push_back(Pair("keypoololdest", (boost::int64_t)pwalletMain->GetOldestKeyPoolTime()));
obj.push_back(Pair("paytxfee", ValueFromAmount(nTransactionFee)));
obj.push_back(Pair("errors", GetWarnings("statusbar")));
+ obj.push_back(Pair("merged-mine", 18950 ));
return obj;
}
So where do we go from here? Either go back to being on the official rules, or stay rogue but realize you are now rogue. btw, what was the actual block count at when you started to run this patch? |
|
|
|
But that does that mean you forked the chain ?
Your namecoind software is useless you need a new version. One that does not exit yet. Currently the version people are supposed to be upgrading to is 0.3.24.62 and any version from .60 to .62 would work together and support the last dot-bit official changes. https://github.com/vinced/namecoin/ for the source, or if you trust my binaries http://www.wuala.com/jbw9/pub/Bitcoin/namecoin/nc0.3.24.62/And how will that stop the 51% attack from taking place. There's still not nearly enough hashing power to prevent that. Only adaption by a few large pools will be able to.
People want bitcoins not namecoins so it's easier to offer bitcoins when all namecoin pools are mining them. I was the only one offering bitcoins while mining namecoins but if all pools can offer bitcoins we have a better chance as I'm not the only one hashing with 100% of the power. I was for a while the only one finding blocks on the namecoin network. [/quote] It may be easier, but you have now in effect attacked the current main namecoins branch.
Getting really confusing now
I've spent the last hour reading the Dot-BIT forum and it seems there's not a real coordinated effort to step up to the challenge....
you are right none of the namecoin developers are speaking to the pool operators, they have ignored my PM as if they are to busy to speak to me. That is ridiculous since the community is to small for this kind of behavior on their part. BTW: However what we did was not good it means anyone who traded in namecoins on the primary chain in the last 12 hrs will have their trades invalidated on our chain. If the developers do not go along with our changes this will get UGLY. Davinci [/quote] It means more then just that, it means either everyone trading and mining on the official client gets screwed, or you guys do.... I hope they don't agree to let to rogue pools force their way over people using the official client. |
|
|
|
Good to see the pool is back up !!! Let's go save NameCoin I am not sure just what DavinciJ15 thinks he is doing now, but he mentioned in irc he enabled merged mining already, prematurely to what the official client was set for. He also has on his newspage Sept 14 Merged mining right now but no payment method for both. Sept 14 New block chain created Pool Down for updates. Since he has created his own fork of the namecoins chain at this point, mining here is attacking namecoins, not defending it now. It may seem that way but there are only 3 pools and a few Gh solo mining on the main chain so it did not make sense to wait we needed to circle the wagons be for he attack. Is it a mistake? Time will tell. Since your blocks are going to be rejected by the official client, it seems a mistake to me to have forked namecoin. Did you get doublec to go along with this as well? That would surprise me. NodeMaster and slush told me that they got doublec to agree and it made sense to me, the only way to get hashing power is to pay bitcoins so forking the block chain seems right, we can deal with the fallout later. I talked to doublec between those posts, seems they lied to you then. He did not change. BTW, there are more then 3 pools. |
|
|
|
Good to see the pool is back up !!! Let's go save NameCoin I am not sure just what DavinciJ15 thinks he is doing now, but he mentioned in irc he enabled merged mining already, prematurely to what the official client was set for. He also has on his newspage Sept 14 Merged mining right now but no payment method for both. Sept 14 New block chain created Pool Down for updates. Since he has created his own fork of the namecoins chain at this point, mining here is attacking namecoins, not defending it now. It may seem that way but there are only 3 pools and a few Gh solo mining on the main chain so it did not make sense to wait we needed to circle the wagons be for he attack. Is it a mistake? Time will tell. Since your blocks are going to be rejected by the official client, it seems a mistake to me to have forked namecoin. Did you get doublec to go along with this as well? That would surprise me. |
|
|
|
Zeitgeist was a shitty movie and 'time-travel attacks' sounds much cooler.
Personally I am going with the time warp attack. |
|
|
|
Good to see the pool is back up !!! Let's go save NameCoin I am not sure just what DavinciJ15 thinks he is doing now, but he mentioned in irc he enabled merged mining already, prematurely to what the official client was set for. He also has on his newspage Sept 14 Merged mining right now but no payment method for both. Sept 14 New block chain created Pool Down for updates. Since he has created his own fork of the namecoins chain at this point, mining here is attacking namecoins, not defending it now. |
|
|
|
When a supposed fix for the exploits being discussed are actually posted publicly so they can be vetted, I will update the fusebox fork of the i0coin chain. But so far it's all talk until the source is openly posted. Frankly since doubleC runs the largest i0coin exchange any submission would essentially need to be a-oked by him otherwise he will have no reason to turn the exchange back on as long as there no coinfidence in the algorithms that were placed in ix/i0 coin to lead to the forks...
That source was published on github yesterday. Edit: think it was in this commit https://github.com/Lolcust/GeistGeld/commit/f9523f33ec22e26b7781f5a545fc74b4ecdc31a6if not it is in one of them anyway |
|
|
|
Store 'em on your computer.
That's what I am trying to do... please see my original post. I am looking for a windows client/wallet. Or is there a trusted online wallet I could use? Guess you could move them into bitparking and trust doublec's monitoring and check point lock in. As for a windows client, only current one I know of is the build at http://www.wuala.com/jbw9/pub/Bitcoin/namecoin/nc0.3.24.61/win32.7z/ , however that has no locked in checkpoints. I would suggest using Multicoin-exp , as an update to the namecoin config file with a lockin will be coming soon. You would need to be a bit comfortable using the cli with either of these. ( http://www.wuala.com/jbw9/pub/Bitcoin/multicoin/MultiCoin-exp/src/ has some *.exe in it for windows built). You would want to make sure you updated to the latest http://exchange.beertokens.info/docs/multicoin/bitcoin.conf.namecoin and be ready to either add or update again when a checkpoint gets added. Maybe Lolcust would be willing to package up his console interface for namecoin usage? |
|
|
|
|
Show Posts
