Follow @Openwall on Twitter for new release announcements and other news
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Wed, 17 Dec 2014 20:40:13 -0900
From: Royce Williams <>
Subject: Re: bleeding-jumbo: john fails to show status on key press

On Wed, Dec 17, 2014 at 2:27 PM, magnum <> wrote:
> On 2014-12-17 17:19, Royce Williams wrote:
>> john often fails to show status on key press for me.
>> This happens on on multiple systems and platforms.  When I exit with
>> control-C, the full status I'm expecting appears.  When I run "john
>> --status" from another prompt, I get the vanilla (non-GPU-aware)
>> output.  But nothing I do gets the full status output.
>> It happens whether I'm SSH'd in, an xterm window, or on physical console.
>> I've mentioned this before on the list, but I haven't heard if it is
>> happening for anyone else. It has been a problem on so many systems
>> over the past couple of years, I'm pretty sure it's not me.  Pretty
>> sure, anyway. :-)
> Do you get an immediate status line + abort when you press ctrl-c once?

Yes, the full ETA and stats summary, including GPU and fan if the
format uses GPU.

> Doesn't happen here. Well, it *does* happen (and even delays ctrl-c) if you
> have lots of salts loaded (especially with GPU formats), or when using a
> wordlist and a huge number of words are rejected by rules.

Hmm. It happens to me with a single salt, a CPU format, and no wordlist.

With a "busier" run like the one you describe, I've seen exactly the
same delays in status and ctrl-c, and they're understandable.  This
issue seems different in kind.

> you could also try sending it a SIGUSR1 from another shell. Does that work,
> but not pressing a key? If so, we need to investigate.

Yes, SIGUSR1 also works.  Interesting.


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.