All of lore.kernel.org
 help / color / mirror / Atom feed
From: Milan Broz <gmazyland@gmail.com>
To: Julien Lepiller <julien@lepiller.eu>, dm-crypt@saout.de
Subject: Re: [dm-crypt] unknown version
Date: Sat, 23 Apr 2016 18:51:52 +0200	[thread overview]
Message-ID: <571BA828.6060405@gmail.com> (raw)
In-Reply-To: <20160423151734.57ff2a5d@polymos>

On 04/23/2016 03:17 PM, Julien Lepiller wrote:
> On Sat, 23 Apr 2016 10:45:52 +0200
> Milan Broz <gmazyland@gmail.com> wrote:
> 
>> On 04/22/2016 02:01 PM, Julien Lepiller wrote:
>>> Hello,
>>>
>>> I am trying to use cryptsetup with a disk that has been encrypted
>>> some time ago. I'm using Linux From Scratch, and built cryptsetup
>>> myself. What I see when I run luksOpen is the following (all
>>> commands are run as root) :
>>>
>>> # cryptsetup 1.7.1 processing "cryptsetup --debug luksOpen /dev/sda1
>>> hdd"  
>> ...
>>> # Activating volume hdd [keyslot -1] using [none] passphrase.
>>> # dm version   OF   [16384] (*1)
>>> # device-mapper: version ioctl on  failed: Permission denied  
>>
>> This looks like you cannot access something (/dev/mapper/control?)
>> and then it just fails because of this initial failure.
>>
>> Do you have SElinux switched on?
>>
>> What is output of "dmsetup version" - does it work?
>>
>> Milan
> 
> Thank you for your answer, the output of the command is:
> 
> Library version:   1.02.121 (2016-04-01)
> Driver version:    4.34.0
> 
> Selinux is switched off (I tried to use it, so I have the libraries,
> but it just does not work at all), and their is nothing in journald.

Ok, so device mapper works.

So if it is not SElinux, something is preventing ioctl to run.
(The error is internal libdevmapper error.)

Do you compile libgcrypt yourself with POSIX capabilities enabled?

If so, gcrypt drops privileges for the whole calling process...
see comment in lib/crypto_backend/crypto_gcrypt.c:

/* FIXME: If gcrypt compiled to support POSIX 1003.1e capabilities,
 * it drops all privileges during secure memory initialisation.
 * For now, the only workaround is to disable secure memory in gcrypt.
 * cryptsetup always need at least cap_sys_admin privilege for dm-ioctl
 * and it locks its memory space anyway.
 */
#if 0


If it is your case, please do not use capabilities in gcrypt or
try to change define above in cryptsetup source as a workaroung and recompile it.

Thanks,
Milan

  reply	other threads:[~2016-04-23 16:51 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-04-22 12:01 [dm-crypt] unknown version Julien Lepiller
2016-04-23  8:45 ` Milan Broz
2016-04-23 13:17   ` Julien Lepiller
2016-04-23 16:51     ` Milan Broz [this message]
2016-04-23 19:22       ` Julien Lepiller

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=571BA828.6060405@gmail.com \
    --to=gmazyland@gmail.com \
    --cc=dm-crypt@saout.de \
    --cc=julien@lepiller.eu \
    /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.