[arch-projects] [initscripts] [GIT] Arch Linux initscripts repository branch systemd-cryptsetup created. 2012.06.2-8-g6fe2126

[Dave Reisner]

On Wed, Jun 27, 2012 at 07:57:15PM -0400, Tom Gundersen wrote:
> This is an automated email from the git hooks/post-receive script. It was
> generated because a ref change was pushed to the repository containing
> the project "Arch Linux initscripts repository".
> 
> The branch, systemd-cryptsetup has been created
>         at  6fe21269e5d54c52c168eac40225dca12a79c355 (commit)
> 
> - Log -----------------------------------------------------------------
> commit 6fe21269e5d54c52c168eac40225dca12a79c355
> Author: Tom Gundersen <teg at jklm.no>
> Date:   Thu Jun 28 01:55:40 2012 +0200
> 
>     cryptsetup: use systemd-cryptsetup rather than rolling our own
>     
>     WORK IN PROGRESS; COMPLETELY UNTESTED!

Danger indeed. I've fixed this up [1], but it's still unusable on my
VMs. I can't seem to get the tools to pop the password prompt on
whatever console device it is that I'm connected to (would have thought
it was /dev/console at this point). That goes for the "async" method of
running all the unlocks at once followed by the query, as well as a
synchronous request from systemd-cryptsetup.

That said, even if the password prompting weren't a problem, I don't
believe that the async method is viable. If a user enters the wrong
password, you can't get the feedback from the password prompting agent
that the password is bad, and the boot process will continue on.

d

[1] http://paste.xinu.at/vsk/

>     
>     In addition to supporting whatever systemd supports, we also support
>     all our own ways of specifying passphrases.
>     
>     We have to look into how our "options" support differs from systemd's
>     and what we want to do about that.
>     
>     Signed-off-by: Tom Gundersen <teg at jklm.no>
> 
> -----------------------------------------------------------------------
> 
> 
> hooks/post-receive
> -- 
> Arch Linux initscripts repository