All of lore.kernel.org
 help / color / mirror / Atom feed
From: Alexander Holler <holler@ahsoftware.de>
To: Bastien Nocera <hadess@hadess.net>,
	"linux-bluetooth@vger.kernel.org"
	<linux-bluetooth@vger.kernel.org>
Cc: Szymon Janc <szymon.janc@tieto.com>
Subject: Re: Why doesn't plugins/sixaxis.c set devices as Trusted?
Date: Tue, 07 Jul 2015 08:49:36 +0200	[thread overview]
Message-ID: <559B7680.5040004@ahsoftware.de> (raw)
In-Reply-To: <559B64FA.2070200@ahsoftware.de>

Am 07.07.2015 um 07:34 schrieb Alexander Holler:
> Am 07.07.2015 um 07:11 schrieb Bastien Nocera:
>> On Tue, 2015-07-07 at 06:31 +0200, Alexander Holler wrote:
>>> Am 06.07.2015 um 15:17 schrieb Bastien Nocera:
>>>> Hey,
>>>>
>>>> I don't understand the reason why plugins/sixaxis.c doesn't set the
>>>> device as trusted when plugged in.
>>>
>>> It's because of security. If you trust a bluetooth device on Linux,
>>> you're trusting it for all services. In case of the sixaxis it means
>>> you're not only trusting it (the BT-MAC) as an input device, but also
>>> as
>>> a network device.
>>>
>>> Now if you trust any plugged in device which says it's a sixaxis, I
>>> would tell my arduino to say it's an sixaxis with a MAC from one of
>>> my
>>> BT-dongles to get a magic device which gives me wireless remote
>>> access
>>> on every linux box with BT when I plug it in once.
>>>
>>> That means you want user interaction, besides just plugging in a
>>> device.
>>
>> What should the pairing process look like then? Because the current
>> workflow is absolutely dreadful.
>
> Pairing is something different than trusting a bluetooth device.
>
> No idea what's your problem. If you remove the necessary user
> interaction to trust a device, you remove the security. Just plugging in
> an (anonymous) usb-device isn't usable as trust.
>
> If Sony decided that's ok for the PS3 (a game console), it's one thing.
>
> But you don't want a wireless remote connected second keyboard or even a
> network device if some just plugged in an anonymous usb-device which
> might even look totally different than as what it presents itself to the
> system.

Just to say it in more clear words. The trust you wanted to give 
automatically does not disappear when the the plugged in device will be 
removed. And furthermore the trust is for a different, wireless 
connected, device, you don't see if you examine the computer.

That's why there is absolutely a need to make the user aware that he is 
going to trust a wireless device.

Alexander Holler

  reply	other threads:[~2015-07-07  6:49 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-07-06 13:17 Why doesn't plugins/sixaxis.c set devices as Trusted? Bastien Nocera
2015-07-07  4:31 ` Alexander Holler
2015-07-07  5:11   ` Bastien Nocera
2015-07-07  5:34     ` Alexander Holler
2015-07-07  6:49       ` Alexander Holler [this message]
2015-07-07  5:35     ` Bastien Nocera
2015-07-07 14:16   ` Harald Schmitt
2015-07-09 20:26     ` Alexander Holler

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=559B7680.5040004@ahsoftware.de \
    --to=holler@ahsoftware.de \
    --cc=hadess@hadess.net \
    --cc=linux-bluetooth@vger.kernel.org \
    --cc=szymon.janc@tieto.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.