Trouble recovering Multibit Classic Keys

[HCP]

Right now, I decided to specify all of my keyboard characters with the token below:
%0,9[aAbBcCdDeEfFgGhHiIjJkKlLmMnNñÑoOpPqQrRsStTuUvVwWxXyYzZ0123456789ºª\!|"@·#$~%%€&¬/()=¿?'¡`+^*[´¨{}Çç,;.:-_ ]
Because I must not have more than 9 characters... but i got this error
Starting btcrecover 0.17.10 on Python 2.7.17 64-bit, 16-bit unicodes, 32-bit ints
btcrecover.py: notice: use --android-pin to recover the spending PIN of
    a Bitcoin Wallet for Android/BlackBerry backup (instead of the backup password)
btcrecover.py: error: on line 1: invalid wildcard (%) syntax (use %% to escape a %)

i saved the file as utf-8 and included --utf8 on comand

Yes, you only specify %% when you a creating a "single" char token... when you're specifying chars in [...] format... you just use a single %. I've just tested it... and created the following token file:

Code: (myTokens.txt)

%0,3[ab%]

And then tested using:

Code:

btcrecover.py --tokenlist myTokens.txt --listpass

It output:

Code:

a
b
%
aa
ab
a%
ba
bb
b%
%a
%b
%%
aaa
aab
aa%
aba
abb
ab%
a%a
a%b
a%%
baa
bab
ba%
bba
bbb
bb%
b%a
b%b
b%%
%aa
%ab
%a%
%ba
%bb
%b%
%%a
%%b
%%%

In addition to that... there is a weird character just before your final "]" which is actually what is causing your error... you need to delete that and the extra %... so your token file should looks something like this:

Code:

%0,9[aAbBcCdDeEfFgGhHiIjJkKlLmMnNñÑoOpPqQrRsStTuUvVwWxXyYzZ0123456789ºª\!|"@·#$~%€&¬/()=¿?'¡`+^*[´¨{}Çç,;.:-_]

The next problen will be when i would solved the error, it could take weeks, months, years¿? for %0,9 characters
it is crazy, i have read about doing it on amazon web services, although I have to research about it, I think it could be more complicated than download python and following tutorials...

What do you think?

Yes... a 9 character password, with that many special chars will take a LONG time to test... With 104 possible characters, it's something ridiculous like 1,423,311,812,421,484,500 possible combinations just for the 9 character passwords... then you have the 8 char passwords, 7 chars, 6 chars etc etc. All up, (with my bad maths) it is something like 1,437,130,373,707,129,900 combinations

If you can check 1,000,000,000 passwords/second... You would need: ~1437130373.73 seconds to check them all...
= ~23952172.9 minutes
= ~399202.9 hours
= ~16633.5 days
= ~45.6 years

You might be better off trying to exclude some characters which you're fairly sure you would not have used to reduce the total search space to a more reasonable amount.

You can also try and do it using Amazon Web Services, but the costs can get prohibitive... so unless you have a significant amount of money in this wallet, that might not be the most cost effective way of doing it. Again, talking to wallet recovery experts who already have the equipment and code setup and ready to go might be a better option.

[1715337850]

1715337850

[1715337850]

1715337850

[1715337850]

1715337850

[1715337850]

1715337850

[1715337850]

1715337850

[oaken]

Right now, I decided to specify all of my keyboard characters with the token below:
%0,9[aAbBcCdDeEfFgGhHiIjJkKlLmMnNñÑoOpPqQrRsStTuUvVwWxXyYzZ0123456789ºª\!|"@·#$~%%€&¬/()=¿?'¡`+^*[´¨{}Çç,;.:-_ ]
Because I must not have more than 9 characters... but i got this error
Starting btcrecover 0.17.10 on Python 2.7.17 64-bit, 16-bit unicodes, 32-bit ints
btcrecover.py: notice: use --android-pin to recover the spending PIN of
    a Bitcoin Wallet for Android/BlackBerry backup (instead of the backup password)
btcrecover.py: error: on line 1: invalid wildcard (%) syntax (use %% to escape a %)

i saved the file as utf-8 and included --utf8 on comand

Yes, you only specify %% when you a creating a "single" char token... when you're specifying chars in [...] format... you just use a single %. I've just tested it... and created the following token file:

Code: (myTokens.txt)

%0,3[ab%]

And then tested using:

Code:

btcrecover.py --tokenlist myTokens.txt --listpass

It output:

Code:

a
b
%
aa
ab
a%
ba
bb
b%
%a
%b
%%
aaa
aab
aa%
aba
abb
ab%
a%a
a%b
a%%
baa
bab
ba%
bba
bbb
bb%
b%a
b%b
b%%
%aa
%ab
%a%
%ba
%bb
%b%
%%a
%%b
%%%

In addition to that... there is a weird character just before your final "]" which is actually what is causing your error... you need to delete that and the extra %... so your token file should looks something like this:

Code:

%0,9[aAbBcCdDeEfFgGhHiIjJkKlLmMnNñÑoOpPqQrRsStTuUvVwWxXyYzZ0123456789ºª\!|"@·#$~%€&¬/()=¿?'¡`+^*[´¨{}Çç,;.:-_]

The next problen will be when i would solved the error, it could take weeks, months, years¿? for %0,9 characters
it is crazy, i have read about doing it on amazon web services, although I have to research about it, I think it could be more complicated than download python and following tutorials...

What do you think?

Yes... a 9 character password, with that many special chars will take a LONG time to test... With 104 possible characters, it's something ridiculous like 1,423,311,812,421,484,500 possible combinations just for the 9 character passwords... then you have the 8 char passwords, 7 chars, 6 chars etc etc. All up, (with my bad maths) it is something like 1,437,130,373,707,129,900 combinations

If you can check 1,000,000,000 passwords/second... You would need: ~1437130373.73 seconds to check them all...
= ~23952172.9 minutes
= ~399202.9 hours
= ~16633.5 days
= ~45.6 years

You might be better off trying to exclude some characters which you're fairly sure you would not have used to reduce the total search space to a more reasonable amount.

You can also try and do it using Amazon Web Services, but the costs can get prohibitive... so unless you have a significant amount of money in this wallet, that might not be the most cost effective way of doing it. Again, talking to wallet recovery experts who already have the equipment and code setup and ready to go might be a better option.

thanks a lot for your quality responses
with those changes it works but the final character was a spacebar...

yes if i can't find a solution i could try with any btc recovering services...
But I prefer to try it first

[HCP]

thanks a lot for your quality responses
with those changes it works but the final character was a spacebar...

You can't put the "space" in the [...]'s... If you do that, it thinks that is the "end" of your token (and why you get the error).

So, if you think that you might have used a space in your password... you'll need to make a slightly more complex token file... something like this:

Code:

%0,1[aAbBcCdDeEfFgGhHiIjJkKlLmMnNñÑoOpPqQrRsStTuUvVwWxXyYzZ0123456789ºª\!|"@·#$~%€&¬/()=¿?'¡`+^*[´¨{}Çç,;.:-_] %s
%0,1[aAbBcCdDeEfFgGhHiIjJkKlLmMnNñÑoOpPqQrRsStTuUvVwWxXyYzZ0123456789ºª\!|"@·#$~%€&¬/()=¿?'¡`+^*[´¨{}Çç,;.:-_] %s
%0,1[aAbBcCdDeEfFgGhHiIjJkKlLmMnNñÑoOpPqQrRsStTuUvVwWxXyYzZ0123456789ºª\!|"@·#$~%€&¬/()=¿?'¡`+^*[´¨{}Çç,;.:-_] %s
%0,1[aAbBcCdDeEfFgGhHiIjJkKlLmMnNñÑoOpPqQrRsStTuUvVwWxXyYzZ0123456789ºª\!|"@·#$~%€&¬/()=¿?'¡`+^*[´¨{}Çç,;.:-_] %s
%0,1[aAbBcCdDeEfFgGhHiIjJkKlLmMnNñÑoOpPqQrRsStTuUvVwWxXyYzZ0123456789ºª\!|"@·#$~%€&¬/()=¿?'¡`+^*[´¨{}Çç,;.:-_] %s
%0,1[aAbBcCdDeEfFgGhHiIjJkKlLmMnNñÑoOpPqQrRsStTuUvVwWxXyYzZ0123456789ºª\!|"@·#$~%€&¬/()=¿?'¡`+^*[´¨{}Çç,;.:-_] %s
%0,1[aAbBcCdDeEfFgGhHiIjJkKlLmMnNñÑoOpPqQrRsStTuUvVwWxXyYzZ0123456789ºª\!|"@·#$~%€&¬/()=¿?'¡`+^*[´¨{}Çç,;.:-_] %s
%0,1[aAbBcCdDeEfFgGhHiIjJkKlLmMnNñÑoOpPqQrRsStTuUvVwWxXyYzZ0123456789ºª\!|"@·#$~%€&¬/()=¿?'¡`+^*[´¨{}Çç,;.:-_] %s
%0,1[aAbBcCdDeEfFgGhHiIjJkKlLmMnNñÑoOpPqQrRsStTuUvVwWxXyYzZ0123456789ºª\!|"@·#$~%€&¬/()=¿?'¡`+^*[´¨{}Çç,;.:-_] %s

This file basically tells btcrecover to create a password created from 9 tokens as follows

Token 1 = Nothing OR 1 of the chars in [] OR space
Token 2 = Nothing OR 1 of the chars in [] OR space
...
Token 9 = Nothing OR 1 of the chars in [] OR space

So, it *should* create all combinations of 0-9 character passwords, where each char can be any of the ones you have specified OR space

Also, you can make it a little more tidy by using: the "a-z" and "0-9" notations...

Code:

%0,1i[a-zñÑ0-9ºª\!|"@·#$~%€&¬/()=¿?'¡`+^*[´¨{}Çç,;.:-_] %s
%0,1i[a-zñÑ0-9ºª\!|"@·#$~%€&¬/()=¿?'¡`+^*[´¨{}Çç,;.:-_] %s
%0,1i[a-zñÑ0-9ºª\!|"@·#$~%€&¬/()=¿?'¡`+^*[´¨{}Çç,;.:-_] %s
%0,1i[a-zñÑ0-9ºª\!|"@·#$~%€&¬/()=¿?'¡`+^*[´¨{}Çç,;.:-_] %s
%0,1i[a-zñÑ0-9ºª\!|"@·#$~%€&¬/()=¿?'¡`+^*[´¨{}Çç,;.:-_] %s
%0,1i[a-zñÑ0-9ºª\!|"@·#$~%€&¬/()=¿?'¡`+^*[´¨{}Çç,;.:-_] %s
%0,1i[a-zñÑ0-9ºª\!|"@·#$~%€&¬/()=¿?'¡`+^*[´¨{}Çç,;.:-_] %s
%0,1i[a-zñÑ0-9ºª\!|"@·#$~%€&¬/()=¿?'¡`+^*[´¨{}Çç,;.:-_] %s
%0,1i[a-zñÑ0-9ºª\!|"@·#$~%€&¬/()=¿?'¡`+^*[´¨{}Çç,;.:-_] %s

[oaken]

it works

I have checked out any possibility between 0 and 4 characters
Now it is counting the 14.025.517.307 combinations for 5 characters, and then it will check them, therefore in 36-48 of mid-work for dont force it, i will get the result

next step?
i will try it with 6 characters, how much time will it get? about 15-20 days

after that i will reconsider any other possiblities  

[falk0069]

Hi all - I just want to share my experience.  I set up MultiBit v0.15.16 on Linux back in Dec 2013 and upgraded to v0.15.18 Apr 2014.  Prior to upgrading I added a second receiving address to my wallet, encrypted it, and I added some funds.  I then have been pretty much hands off for the last 10 years.  I did upgrade my system a couple times but just copied the MultiBit folders over and made sure it still started and showed the wallet info.  Then a few days ago I decided I wanted to switch to a completely offline version and needed the private key.  This is where my adventures started.

I was positive I knew the password.  Ten years ago I didn't use a password manager and only had a handful of passwords that I'd use with just some variety.  When I ran the openssl command

Code:

openssl enc -d -p -aes-256-cbc -md md5 -a -in multibit-yyyymmddhhmmss.key

it presented no errors but it was just gibberish.  I thought for sure I was facing a MultiBit bug mentioned in this thread and I had a corrupt key.  Also, when I ran the decrypt_multibit_classic_keys.py from https://github.com/HardCorePawn/multibit_recovery I also saw the same gibberish.

The fact that my password was sort of working and one that I was positive I would have used 10 years ago, really prolonged my adventure.  It turns out it there is a really good change a bad password will still successful decrypt--just not correctly.

I also did start trying to use btcrecover.py only to realize it will take years to decrypt a password.  Fortunately, for me when I realize that the password might just incorrect, I was able to rack my brain enough to figure out what password I used.  

Also in my case, I had the main wallet file, 3 key backups, 3 wallet backups, and a wallet.cipher file.  My 'bad' password 'incorrectly' worked on one of the files.  What are the odds that seven files are really corrupts and one file mostly corrupt.  Once I figured out the right password, I ran all of HardCorePawn awesome scripts on all 8 files and got the expected results.

So moral of the story is if you think you have a corrupt file, try all of HardCorePawn scripts on all the different backup files that MultiBit creates and see if you can get any of them to work.