51% Attack - Countermeasure Roundup

[vinced]

You have copies of all the blocks even if there is an attack, so you can always go back to the last block before the attack started.

This doesn't help exchanges and merchants who suffer loss in physical goods or other currencies during an attack.

Yes, agreed. In the long term we really need more hash power which we will get with merged mining and wider adoption.

[johntobey253]

If a real 51% attach starts we need a way to exclude or disadvantage the chain built by the attacking miners. Any idea how we would do that? We still want to be an open system while defending.

Thanks for asking. I'd like us to acknowledge that after 19200, we are ultimately sunk if more hashing power (in the BTC network) wants to destroy Namecoin than wants to support it. Several people are *very* suspicious of merged mining and Namecoin and would, I am sure, support an attack intended to kill it and suppress any future merged-mining alt chains. Personally, I think this would set Satoshi's dream back five or ten years, maybe longer. Therefore, our best and only long-term defense is friendly public relations with BTC pools/miners.

I respect your and khal's personal time but would feel safer with another committer or two capable of releasing officialish upgrades in a pinch.

There are some holes to be filled, especially the "time warp" that doublec has been mentioning. I'd suggest reading up on every weakness mentioned by ArtForz.

I'll bet we could come up with heuristics for namecoind to detect "unusual" network activity and notify the user. Examples would include long side chains and abrupt changes in network hash rate.

Other alt chains (SolidCoin, Geist Geld, ...) have dealt with or are dealing with real or theoretical attacks, and their solutions may interest us.

[doublec]

I'll bet we could come up with heuristics for namecoind to detect "unusual" network activity and notify the user. Examples would include long side chains and abrupt changes in network hash rate.

I do this on my exchanges now. I detect if an attempt is made to rewrite the block chain past the deposit confirmation limit for the exchange. When this happens the exchange stops processing blocks. Then I manually work through the two chains to see if the attempt was to double spend on of the deposits. This is the current state of the i0coin exchange, where the 20 confirmation limit was exceeded.

[vinced]

...
I respect your and khal's personal time but would feel safer with another committer or two capable of releasing officialish upgrades in a pinch.
...
There are some holes to be filled, especially the "time warp" that doublec has been mentioning. I'd suggest reading up on every weakness mentioned by ArtForz.

If a trusted person steps forward I can add them to the github repository as commiter.

The timetravel fix was released yesterday.

[cosurgi]

If a trusted person steps forward I can add them to the github repository as commiter.

I suggest nodemaster

[BitcoinEXpress]

I did not go crazy. Shinning the light on a plan is the best way to determine it's merit. I talked to other people in the community and conclude that the plan was flawed.

Umm.. yeah. Must be the reason why you was the only one mining on the wrong blockchain after all. Anyway thank you very much for your integrity handling the issue secretly as you was asked by everybody. I really appreciate your trustworthyness and keep that in mind.

Hey nodemaster,
First time I have seen this thread and it pisses me off.

Why are you beating up on Davinci? He was one of the main reasons I decided to stop this.

Nodemaster, you are an idiot to say the least. You do realize that the whole 51% attack was to distract you and The Time Travel EXploit is the real danger. NMC has peculiarities that make it EXCEPTIONALLY prime for this attack and less than 20% is needed.

"Forking the Chain" only gives more momentum to someone in a 51% attacking LOL...you really don't understand it at all do you and I know sure as hell you have no clue about The Time Travel EXploit. Forking the chain to defend against a 51% attack is like throwing gas on a burning fire to put it out!

Here's another newsflash for you, Merged Mining will make "Lock In Points" meaningless.

It's good thing I don't go back on my word or I would make you a "true believer" the same way I made Coinhunter after he made his public challenge. Notice, SC kinda a took a dive after he asked me to take my best shot.

Keep in mind, you guys haven't fixed anything, you're as wide open as you were a month ago.

[doublec]

It's good thing I don't go back on my word or I would make you a "true believer" the same way I made Coinhunter after he made his public challenge.

You need some new material BitcoinEXpress, your trolling's getting ordinary.

[nodemaster]

If a trusted person steps forward I can add them to the github repository as commiter.

I suggest nodemaster

IMHO we should have at least three persons who are able to release new versions. Perhaps we should as well ask doublec and/or khal? If no one is against I'm willing to take this challenge, but I'd also want to make sure, that I'll only make use of the rights if there is imminent danger for the blockchain, we need an emergency release and vinced is not available for whatever reason. During normal development process vinced should be the only one who is pulling into his repo.

We should develop and agree on a process for emergency releases like:

1. Emergency fix is done on fork by whoever provide it
2. After testing it is pulled to vinceds repo
3. Version is advanced
4. Announcement is made on Forum. PGP signed by at least 2 out of n persons who can commit to the repo.
5. Binaries are created and made available on the main page.

I'm just working on automatic binary creation for linux and packaging for debian. Perhaps someone is able to do this for windows and Mac OS X? It's really a good idea to tighten the process of doing emergency releases. This is one of the weaknesses we identified during the 51% attack and I think we should address this. What do you think?

[khal]

I splitted the discussion on an official tree here :
- Creating an official namecoin tree ?

[BitcoinEXpress]

@doubleC

No problem DoubleC

You want it like that. No problem.

Your exchanges and pools are my primary target effective immediately. I will also reorganize the Namecoin attack and will begin as soon as I can get it assembled. We'll see who's trolling.

BCX