john-users - Re: john and muti-core settings and progress display

Openwall GNU/*/Linux 3.0 - a small security-enhanced Linux distro for servers
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Tue, 1 Jan 2013 11:12:09 +0100
From: Frank Dittrich <frank_dittrich@...mail.com>
To: john-users@...ts.openwall.com
Subject: Re: john and muti-core settings and progress display

On 01/01/2013 12:13 AM, John Hall wrote:
> 
> When using john with the jumbo patch and enabling multi-core on a Fedora 17 OS on vmWare VMworkstation with 2 cores enabled per below, I noticed that pressing "any key" when john is grinding away produces a different progress display that I saw with single core.
> 
> With single core I saw something like "...trying: Initialguessabc123"
> 
> With multicore I saw something like "...trying Initialguessabc123 - Initialguessxyz456"
> 
> So is john actually running 2 threads and each separately is attempting to discover the hash code?

Actually, "...trying Initialguessabc123 - Initialguessxyz456" doesn't
necessarily mean that just 2 passwords candidates are tried simultaneously.
How many passwords are tried simultaneously, depends on the format and
on compile time settings.

It also depends on the cracking mode. Many formats use a smaller number
of password candidates cracked simultaneously in --single mode.

So, "Initialguessabc123 - Initialguessxyz456" is indicating a range of
passwords tried simultaneously. The first candidate and the last
candidate get printed.

If you want to know how many candidates are tried simultaneously (and
much more format specific information), you can use

./john --list=format-all-details|less

Not all of this information might be interesting for a user.

Here is he information which gets printed for the zip format (close to
the end of the output) for my 32bit Linux system:

Format label                            zip
Max. password length in bytes           64
Min. keys per crypt                     1
Max. keys per crypt                     96
Flags
 Case sensitive                         yes
 Supports 8-bit characters              yes
 Converts 8859-1 to UTF-16/UCS-2        no
 Honours --encoding=NAME                no
 False positives possible               yes
 Uses a bitslice implementation         no
 The split() method unifies case        no
 A $dynamic$ format                     no
Number of test cases for --test         2
Algorithm name                          32/32
Format name                             WinZip PBKDF2-HMAC-SHA-1
Benchmark comment
Benchmark length                        -1
Binary size                             2
Salt size                               512

In this case, 96 password candidates will be tried simultaneously:
Max. keys per crypt                     96

Frank