Bug#747180: Bug#766237: [systemd-shim] Version property from org.freedesktop.systemd1.Manager

2014-11-04 Thread Martin Pitt
Hello all,

replying to both bugs at the same time, as they are more or less the
same.

Török Edwin [2014-10-21 20:49 +0300]:
 systemd-shim doesn't provide the Version property from 
 org.freedesktop.systemd1.Manager.
 According to the documentation[0] applications shouldn't parse this property, 
 but powerdevil from KDE does
 and hides the Sleep/Hibernate buttons [1].

I'm not avert to adding the Version property to systemd-shim, but
conceptually this doesn't help at all. Merely checking the version
number does not tell you (i. e. powerdevil) in any way whether
suspend/hibernate is possible. What powerdevil should do is to not
query for a particular init system, but rather ask logind:

| $ gdbus call --system -d org.freedesktop.login1 -o /org/freedesktop/login1 -m 
org.freedesktop.login1.Manager.CanSuspend
| ('yes',)
| $ gdbus call --system -d org.freedesktop.login1 -o /org/freedesktop/login1 -m 
org.freedesktop.login1.Manager.CanHibernate
| ('na',)

I. e. on my system suspend works, but hibernate is not available (I
have no swap partition. logind works under any init system, and
hibernation isn't available on some systems regardless of the init
system.

So I suggest we keep #747180 against powerdevil to fix the
implementation of its CanSuspend/Hibernate, and I'll just add the
Version property to -shim (it can't hurt really).

Thanks,

Martin
-- 
Martin Pitt| http://www.piware.de
Ubuntu Developer (www.ubuntu.com)  | Debian Developer  (www.debian.org)


-- 
To UNSUBSCRIBE, email to debian-qt-kde-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: https://lists.debian.org/20141104103151.ga3...@piware.de



Bug#747180: [kde-workspace-bin] due to checkSystemdVersion()

2014-11-04 Thread Martin Pitt
Control: tag -1 fixed-upstream pending

Hey Török,

Török Edwin [2014-10-21 20:31 +0300]:
 +  return g_variant_new_take_string (g_strdup_printf(systemd %d, 
 SYSTEMD_VERSION));

This is wrong, systemd only gives back the version without the
systemd  prefix. I adjusted the patch accordingly and pushed
upstream:

 https://github.com/desrt/systemd-shim/commit/dd8c6a1a7

Thanks!

Martin

-- 
Martin Pitt| http://www.piware.de
Ubuntu Developer (www.ubuntu.com)  | Debian Developer  (www.debian.org)


signature.asc
Description: Digital signature


Bug#611245: libqtserviceframework1: Please drop hal dependency

2011-01-27 Thread Martin Pitt
Package: libqtserviceframework1
Version: 1.1.0~beta2-1

HAL has been deprecated and unmaintained for several years now [1],
and it's time to clean up the remaining usage of it, as it's only
going to bitrot and break more.

This package still depends on hal. It doesn't build depend on it, but
it seems like the code talks to Hal's D-BUS interface directly.

Are there any plans to move that to libudev? Failing that, could that
dependency at least become a Suggests: instead?

Thanks for considering,

Martin

[1] https://wiki.ubuntu.com/Halsectomy
-- 
Martin Pitt| http://www.piware.de
Ubuntu Developer (www.ubuntu.com)  | Debian Developer  (www.debian.org)


signature.asc
Description: Digital signature


Bug#506181: [Pkg-cups-devel] Bug#506181: Various backends have the wrong permissions

2008-11-20 Thread Martin Pitt
Hi Chanoch,

Chanoch (Ken) Bloom [2008-11-19  8:21 -0600]:
 I decided to test this by purging and reinstalling cups. Cups worked out
 of the box. Then I went to reconfigure things the way I had been using
 CUPS (1 printer, shared with the whole subnet) by editing things in the
 KDE Printing control panel. After changing things with Print Server -
 Configure Server..., KDE wrote out a whole different cupsd.conf, hardly
 resembling the original. I can't figure out how that cupsd.conf is
 assigning permissions. kdeprint added the line Group lpadmin, but the
 test backend gives uid=7(lp) gid=0(root) groups=0(root).

Ah, Group lpadmin is wrong. It should be SystemGroup lpadmin,
and it shouldn't change Group.

 I guess this is a kdeprint bug rather than a cups bug, so I'm
 reassigning it. The two versions of cupsd.conf are attached.

Right, thank you!

  I do see the issue with the serial backends. For that we probably need
  to make the serial backend run as root.
 
 Perhaps making the various backends SetGID (or SetUID) to the
 permissions they need would help?

I fixed it in cups now (see attached patch), will upload to
experimental and unstable.

Thanks,

Martin

-- 
Martin Pitt| http://www.piware.de
Ubuntu Developer (www.ubuntu.com)  | Debian Developer  (www.debian.org)
=== modified file 'debian/changelog'
--- debian/changelog	2008-11-20 07:53:16 +
+++ debian/changelog	2008-11-20 07:56:07 +
@@ -9,6 +9,9 @@
 backchannel EOF (select() returns ready for read on EOF). This
 completely broke printing with e. g. HPJetDirect. Thanks to
 Samuel Thibault for tracking down the problem! (Closes: #489045)
+  * debian/rules: Install the serial backend with 0744 permissions to make it
+run as root, since /dev/ttyS* are root:dialout and thus not accessible as
+user lp. Thanks to Chanoch (Ken) Bloom. (One part of #506181)
 
  -- Martin Pitt [EMAIL PROTECTED]  Thu, 20 Nov 2008 08:52:00 +0100
 

=== modified file 'debian/rules'
--- debian/rules	2008-05-23 08:14:05 +
+++ debian/rules	2008-11-20 07:55:48 +
@@ -73,6 +73,10 @@
 	   install -D -m 644 debian/local/apparmor-profile debian/$(cdbs_curpkg)/etc/apparmor.d/usr.sbin.cupsd; \
 	fi
 
+	# Make the serial backend run as root, since /dev/ttyS* are
+	# root:dialout and thus not accessible as user lp
+	chmod go-x debian/$(cdbs_curpkg)/usr/lib/cups/backend-available/serial
+
 binary-post-install/libcupsimage2-dev::
 	rm -r debian/libcupsimage2-dev/usr/share/doc/libcupsimage2-dev
 	ln -s libcupsimage2 debian/libcupsimage2-dev/usr/share/doc/libcupsimage2-dev



signature.asc
Description: Digital signature


Re: Processed: merging 313507 313602, severity of 303608 is important, tagging 303608

2005-06-14 Thread Martin Pitt
Hi!

Adeodato Simó [2005-06-14 20:55 +0200]:
   severity 303608 important
   tags 303608 - security
  Bug#303608: libqt0-ruby1.8: incorrect include in qtruby library badly 
  polutes Module namespace
 
   Martin, was this a typo?

Yes, sorry, I just reopened the bug.

Martin
-- 
Martin Pitthttp://www.piware.de
Ubuntu Developer   http://www.ubuntu.com
Debian Developer   http://www.debian.org


signature.asc
Description: Digital signature


Bug#303608: marked as done (libqt0-ruby1.8: incorrect include in qtruby library badly polutes Module namespace)

2005-06-14 Thread Martin Pitt
reopen 303608
thanks

Hi!

Debian Bug Tracking System [2005-06-14 11:48 -0700]:
 Changes: 
  postgresql-7.4 (1:7.4.8-9) unstable; urgency=low
  .
* debian/rules: Fix POT file generation.
* Added debian/patches/16_libpq_defaultsocketdir.patch:
  - Set default socket directory for libpq to /var/run/postgresql.
Closes: #303608

Darn, that should have been #313608. I reopen this bug now. Severity
and tags are not altered as far as I can see.

I apologize for the inconvenience.

Martin

-- 
Martin Pitthttp://www.piware.de
Ubuntu Developer   http://www.ubuntu.com
Debian Developer   http://www.debian.org


signature.asc
Description: Digital signature