Your message dated Wed, 21 Jul 2010 11:09:31 +0200
with message-id
<1279703371.25386.29.ca...@etppc09.garching.physik.uni-muenchen.de>
and subject line Re: [pkg-cryptsetup-devel] Bug#589641: Bug#589686: cryptsetup:
split out keyscript in separate packages
has caused the Debian Bug report #589686,
regarding cryptsetup: split out keyscript in separate packages
to be marked as done.
This means that you claim that the problem has been dealt with.
If this is not the case it is now your responsibility to reopen the
Bug report if necessary, and/or fix the problem forthwith.
(NB: If you are a system administrator and have no idea what this
message is talking about, this may indicate a serious mail system
misconfiguration somewhere. Please contact [email protected]
immediately.)
--
589686: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=589686
Debian Bug Tracking System
Contact [email protected] with problems
--- Begin Message ---
Package: cryptsetup
Version: 2:1.1.3-1
Severity: wishlist
Hi Jonas.
Not sure whether I've already suggested this and it was just reject... so simply
close it if so.
We should perhaps consider, to split out the keyscripts in separate packages.
I mean e.g. something like:
cryptsetup-openct
cryptsetup-opensc
cryptsetup-openssl
etc.
(not yet sure about passdev...)
Of course one might argue that these (usually) two small scripts are overkill
for own packages...
but the main benefit I see at the moment is, that each of this minor packages
could depend on
what it needs, e.g. openssl, openct, and so on.
Which is in cryptsetup itsefl not really possible,... as it would be too
annyoing to depend on
just everything any single keyscript would need.
I think, that each such package would have to depend on cryptsetup itself,
because most keyscripts
and their hook scripts will require it directly (by depending on cryptroot or
looking for it in the
hookscript) or indirectly (e.g. they use askpass).
cryptsetup itself should suggest each such minor package.
A minor benefit could be to make
/usr/share/initramfs-tools/conf-hooks.d/cryptsetup less demanding.
I guess we always have to set BUSYBOX=y because cryptroot already needs it,
right?
But maybe there are keyscripts which do not strictly need KEYMAP=y, as they
don't require a passphrase
or so.
This could also make life easier in solving #589641. If the "ugly" solution
would be the one to go,
one could perhaps make use of triggers.
But I must admit that I do not yet fully understand the concept of triggers,
and whether they'd
actually help us.
Cheers,
Chris.
--- End Message ---
--- Begin Message ---
Seems I've misunderstood how the control server works,... therefore
closing them this way, for the "reasons" laid out above.
Chris.
--- End Message ---
