[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20200426181647.GA583@openwall.com>
Date: Sun, 26 Apr 2020 20:16:47 +0200
From: Solar Designer <solar@...nwall.com>
To: john-users@...ts.openwall.com
Subject: Re: Bitcoin and John the Ripper - Algorithm update. When?
Hi,
On Sun, Apr 26, 2020 at 07:34:06PM +0200, MA40 wrote:
> With John the Ripper you can find the passwords for the "wallet.dat" file
> hashes. The private keys of the Bitcoins that the owner of the file owns
> are encrypted in these files.
>
> The hashes of the "wallet.dat" files are recorded using two different
> algorithms:
>
> - One results in hashes starting with "$bitcoin$96$..." (This is the
> old one).
>
> - And the other, (the modern one), results in hashes starting with
> "$bitcoin$64$...".
This is the same algorithm with a different parameter value: length of
included data. We've switched to including less data as part of Dhiru
Kholia's commit in 2018 (thus, prior to release of JtR 1.9.0-jumbo-1),
"Generate less revealing hashes for Bitcoin wallets (#3290)". We've
started generating even less revealing "hashes" in bleeding-jumbo after
the 1.9.0-jumbo-1 release, but that's a separate story. Those even
newer "hashes" are supposed to also be supported by 1.9.0-jumbo-1 and
even older. We didn't break compatibility.
> The latest version of John the Ripper (v-1.9.0), which is from April 12,
> 2019, only works with the old algorithm.
What makes you think so? My understanding is that 1.9.0-jumbo-1, as
well as the truly latest code in bleeding-jumbo, are both supposed to
work with both of the above kinds of Bitcoin "hashes".
> Can the next version be expected to bring the modern algorithm built-in?
> When?
At least per my understanding, we already support all of this.
Alexander
Powered by blists - more mailing lists
Confused about mailing lists and their use?
Read about mailing lists on Wikipedia
and check out these
guidelines on proper formatting of your messages.
