Search content
Sort by

Showing 7 of 7 results by nosferatu8701
Post
Topic
Board Development & Technical Discussion
Re: Does a multi-sig wallet protect from random private key attacks?
by
nosferatu8701
on 11/05/2022, 14:15:18 UTC
Quote
Did you run it randomly or on very specific range?

I ran it randomly. I'm not concerned that someone will guess my private key with the intention of guessing "my" private key. I'm concerned about somebody stumbling upon my private key by running these softwares at scale (one instance can run 1 million combinations per second. If someone were to run 10,000 instances they will most likely come across keys which have bitcoin in them)
Post
Topic
Board Development & Technical Discussion
Re: Does a multi-sig wallet protect from random private key attacks?
by
nosferatu8701
on 11/05/2022, 04:53:31 UTC
Is the above true? If an attacker were to randomly come across my private key, he can move the funds without requiring the origin keys that resulted in the multi sig?
I never heard of a single case of anyone losing coins with multisig setup with attack like you mentioned, and I couldn't find anything about reddit topic talking about this, so maybe you should post a link for us to see.
I know that more more co-signers you have in multisig setup, the harder it will be for attacker to stole your coins, and I don't see any real threat with this.
With new taproot addresses all transactions like the same, so there is no way you could know if transaction is single or multi sig, but that is not the case with older address types.

This is the post and the other comments that follow it.

My primary concern is dictionary attacks. I know and have tried using rotorcuda and fialka to run random private key attacks and trying to find private keys. In fact, I have already found a few private keys (unfortunately they were already emptied before by someone else). However, this is very much a possibility. The fact that me, an individual can run such brute force attacks for random keys with little knowledge concerns me. I'm sure that North Korea and other big malicious actors would be running far bigger operations to brute force random keys. I may go so far as to even say that these whale alerts that we see on twitter (that some bitcoin was moved after 10-11 years) may be such crackers stumbling on these private keys.

I want to protect myself from such attacks by using multi sig. My assumption was that the Bitcoin chain requires the 2 signatures and this enforcement is done on chain. However those reddit comments and the ones in this thread too suggest otherwise.
Post
Topic
Board Development & Technical Discussion
Re: Does a multi-sig wallet protect from random private key attacks?
by
nosferatu8701
on 08/05/2022, 10:15:04 UTC
> But, that's true for everything. If the attacker randomly comes across your private key, he's also the owner of that public key. Multi-sig or not.

I used to think that multi sig is enforced on chain and the chain would require signature of both keys to move the funds.
Post
Topic
Board Development & Technical Discussion
Merits 3 from 2 users
Topic OP
Does a multi-sig wallet protect from random private key attacks?
by
nosferatu8701
on 08/05/2022, 08:52:00 UTC
⭐ Merited by Welsh (2) ,vapourminer (1)
I am currently using a multi-sig setup for my bitcoin wallet (Sparrow wallet).

Does using a multi-sig wallet protect me from random private key guessing attacks.

Someone on reddit pointed it to me that "Multisigs exist in the same pool of 2^256 keys. The resulting key size isn't bigger just because it's multisig. You can find a non-multisig that is the same as your resulting key from a multisig wallet."

Is the above true?
Post
Topic
Board Bitcoin Technical Support
Re: Bitcoin censorship for Russian addresses?
by
nosferatu8701
on 26/02/2022, 13:54:30 UTC
There's a more fundamental problem with this: governments long wanted control over Bitcoin. If they manage to force miners to do what they want, Bitcoin is basically over.
Haven't they done everything they could in the last over 10 years? Bitcoin is going nowhere, not in the near future. In fact bitcoin is the solution for the new economy which many government do not want at all especially the US. These centralized exchanges are playing a key role for government censorship. Bitcoin community needs to do something to eliminate these centralized exchanges to enjoy the true bitcoin experience.

Apologies if my questions are too basic. If there are reading materials that explain in detail how the Bitcoin network works relating to some of the questions I posted above, I'd appreciate it if you can point me in that direction.
LoyceV already said what I was planning to write. Just a little suggestions - if you have bitcoin then don't keep them in an exchange. Move them to your own wallet.


Thank you. I have been hearing about Bitcoin since 2009 but mental inertia stopped me from giving a damn about it. I dismissed it as farmville money for 12 years. Never even bothered researching about it.
Finally made the leap in 2021 for purposes I'd like to keep out of this thread  Wink

Yes, I'm using a Ledger but trying to move to multi-sig. I will have a new thread coming up on that today about questions I have regarding that.
Post
Topic
Board Bitcoin Technical Support
Re: Bitcoin censorship for Russian addresses?
by
nosferatu8701
on 26/02/2022, 13:47:26 UTC
Thanks for your answers.

To clarify,

"Russian Bitcoin addresses" meaning addresses which have been identified by the US government / other world governments to belong to rogue countries.

I still am unclear about what happens if miners are told to censor transactions originating from certain addresses.

Let's assume that the Russian government sends a 1500 BTC transaction to China. This transaction goes to the mempool.

Q. If there are 100 miners around the world and 100,000 transactions in the mempool, do all the 100 miners pick up the same 4000 txns in the block sorted by fee? (I read online that a block has approx 4000 txns). I would imagine that all 100 miners would pick up different transactions? If so, I guess there would be different blocks. But the same txn could be picked up by different miners in different blocks. When a block is solved, do other miners drop the block they are solving if the most recent solved block contains a txn that is in their block? If so, does the other non-common txns return back to the mempool?

Now, Let's take an example of Hut8 and Marathon Digital (BTC mining companies in USA).

Q. If the USA government controls these miners and tells them to not pick up transactions from the mempool that belong to the addresses blacklisted by the US government, can't a Russian miner create their own block with the transactions from the mempool ?

Q. What happens Hut8 does not pick up the txns in the block but an Australian miner does?

Post
Topic
Board Bitcoin Technical Support
Merits 21 from 6 users
Topic OP
Bitcoin censorship for Russian addresses?
by
nosferatu8701
on 26/02/2022, 12:59:09 UTC
⭐ Merited by Royse777 (5) ,LoyceV (4) ,o_e_l_e_o (4) ,BlackHatCoiner (4) ,ETFbitcoin (3) ,Husna QA (1)
I read this article today that talks about the possibility of bitcoin censorship.

Post this, I had some fundamental questions regarding how the bitcoin network works. Would appreciate if someone could help me understand this.

As per this article: https://tftc.io/martys-bent/issue-1170,

Quote
If Russians are able to get transactions included in blocks alongside individuals from countries who are still connected to SWIFT, it sort of makes the chord cutting mute. In an attempt to prevent this from happening it is totally possible that the US government and other NATO governments would try to thrust regulations on the mining industry to keep a blacklist of Russian bitcoin addresses at all times and never mine a block with a transaction that is sent from any of those addresses lest they want to be subjected to harsh punishments for violating sanctions. Worse yet, they could even try to force a whitelist of approved addresses tied to the identities of individuals and make it so mining pools are only allowed to interact with those addresses.

Question: 

  • How do transactions get included in a block?
  • Are miners responsible for deciding which transactions are included in a block? I thought miners work towards finding the solution for the block.
  • If so, the block must be already created ? Who created this block ?
  • When I send a bitcoin transaction, does it get sent to a miner ? How does my client (such as electrum) know whom to send the transaction to?
  • What other parameters are taken into consideration while including a transaction into a block for eg. Fee? If so, is it possible that Russian bitcoin addresses can give an incredibly large fee to incentivise miners to include the transaction?