btcrecover
btcrecover copied to clipboard
3rdIteration
Improve pre-start dialogue/messages & add ability to skip/shorten (Search stuck at 0 for some time with some types of search)
Hello, I'm using btcrecover to recover eth seed phrase. I know some of the words, and for the rest I have a few guesses each, so I use a tokenlist which looks like this:
+ ^1^word1 ^2^word2-1 ^2^word2-2 ^2^word2-3 + ^3^word3 ... ^12^word12
I'm running it like this:
python seedrecover.py --no-dupchecks --mnemonic-length 12 --language EN --dsw --wallet-type ethereum --addr-limit 1 --addrs -address- --tokenlist ./tokens.txt --no-eta
I get this:
Starting seedrecover 1.3.0-CryptoGuide, btcrecover 1.3.0-Cryptoguide on Python 3.7.3 64-bit, 21-bit unicodes, 64-bit ints Using the 'en' wordlist. 2020-07-19 14:00:58 : Phase 1/1: up to 12 mistakes, 12 of which can be an entirely different seed word. 2020-07-19 14:00:58 : Using 8 worker threads | 0 elapsed: 0:00:00 rate: 0.00 P/s
And it gets stuck here. Ctrl-C doesn't work. I've tried both with and without --no-eta, without it it just gets stuck after Phase 1/1
My tokens list has 47 lines, only 3 of those are + If I significantly reduce the amount of tokens it starts working.
It starts going after a pretty long time. If it's taking this time to precalculate the combinations or whatever, I believe there should be an indication.
As an additional note, I'm not exactly sure what it's doing under the hood as I haven't explored the sources, but I ended up writing a small script of my own for this task specifically and the precalculation step only took some minutes. I wasn't able to use btcrecover at all with 56 tokens, as it was still stuck after I left it do its thing for hours.
hi, im having the same issue while using this position anchor like you did. it can only run up to 126 words, while i need to run 198 words, it stuck at 0elapse also
hi, im having the same issue while using this position anchor like you did. it can only run up to 126 words, while i need to run 198 words, it stuck at 0elapse also
I can share my script if this is still relevant, it might not suit your needs though.
i had found the solution to run it. you have to arrange the words in your txt file like below:
example: +^1^word1 ^2^word2-1 ^2^word2-2 ^2^word2-3 +^3^word3 ... ^12^word12
@3rdIteration, this does not solve my problem, you can run 100,000 lines of passphrases-list and wait for 1 hour - nothing will change, although you claimed that it will take 15 minutes with 1 million lines.
@3rdIteration, this does not solve my problem, you can run 100,000 lines of passphrases-list and wait for 1 hour - nothing will change, although you claimed that it will take 15 minutes with 1 million lines.
That's exactly this issue. 1 million passphrases to check with each seed will take about 1000 second per seed checked, so about 15 minutes per seed. (Even in the pre-start benchmark step where a number of seeds are checked, but there is no updates in the UI while this is happening)
@3rdIteration, I use only one mnemonic, and it seems that I do not understand something, what do you mean by seed - result 1 mnemonic + 1 passphrase ? And each result takes 15 minutes of time?
If you are only using one mnemonic then just use BTCRecover.py and do a normal passphrase recovery.
@3rdIteration, I just gave an example that I only use one mnemonic right now (testing program). In fact, there are many more options. And yes, btcrecover does not support BIP39 + OpenCL
OpenCl isn't being used for the passphrase checks in any meaningful way anyway, so if you only have one mnemonic, then you might as well use that.
On Tue, Feb 2, 2021 at 11:09 AM urbanurba [email protected] wrote:
@3rdIteration https://github.com/3rdIteration, I just gave an example that I only use one mnemonic right now (testing program). In fact, there are many more options. And yes, btcrecover does not support BIP39 + OpenCL
— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/3rdIteration/btcrecover/issues/49#issuecomment-771746283, or unsubscribe https://github.com/notifications/unsubscribe-auth/AARAQLTKGVD4S6ICJSBYSHLS5APR3ANCNFSM4PBG4X6Q .
This behavior is now improved for running seedrecover with large numbers of passphrases as of https://github.com/3rdIteration/btcrecover/commit/2743e9587d0cd84e12dea50a7dfb609d777b5f3e
@3rdIteration, why not use this kernel - https://github.com/openwall/john/blob/bleeding-jumbo/run/opencl/pbkdf2_hmac_sha512_kernel.cl ? salt = passphrase
@3rdIteration, why not use this kernel - https://github.com/openwall/john/blob/bleeding-jumbo/run/opencl/pbkdf2_hmac_sha512_kernel.cl ? salt = passphrase
It's not offering any functionalities that the current kernel doesn't have...
@3rdIteration, but what about salt? As far as I understood this is equivalent to a passphrase And the current version of the kernel does not support brute-force passphrases (well, you know that yourself)
The current kernel does support checking a list of seeds where as given passphrase is in use, just like the one you linked. What neither of them support is taking a list of passphrases for a given seed.
There is a feature currently being sponsored whereby GPU acceleration for passphrase recovery will be available.
@3rdIteration, this is amazing news! When is the release expected? Unfortunately, now there are no free funds, but I could sponsor you if I could restore my wallets with this