David Herrmann [2013-12-01 16:57 +0100]:
> Screen can be fixed to call:
> pam_start(&pamh)
> pam_open_session(pamh)
>
> and during shutdown:
> pam_close_session(pamh)
> pam_end(pamh)
Please not; screen has no business interfering with the PAM stack, it
does not start login sessions by its
https://bugzilla.redhat.com/show_bug.cgi?id=1036462
___
systemd-devel mailing list
systemd-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/systemd-devel
On Sun, Dec 1, 2013 at 3:59 PM, Zbigniew Jędrzejewski-Szmek
wrote:
> On Sun, Dec 01, 2013 at 02:50:15PM -0800, Shawn Landden wrote:
>> ---
>> src/shared/strv.c | 2 +-
>> 1 file changed, 1 insertion(+), 1 deletion(-)
>>
>> diff --git a/src/shared/strv.c b/src/shared/strv.c
>> index 607c221..cc6ad
On Wed, Nov 20, 2013 at 12:35:04AM -0800, Shawn Landden wrote:
> ---
> src/core/service.c | 3 +--
> src/core/service.h | 2 --
> 2 files changed, 1 insertion(+), 4 deletions(-)
>
> diff --git a/src/core/service.c b/src/core/service.c
> index c0ee114..24f7a42 100644
> --- a/src/core/service.c
> +
---
src/nspawn/nspawn.c | 19 +++
1 file changed, 7 insertions(+), 12 deletions(-)
diff --git a/src/nspawn/nspawn.c b/src/nspawn/nspawn.c
index dd7337b..f400a65 100644
--- a/src/nspawn/nspawn.c
+++ b/src/nspawn/nspawn.c
@@ -481,10 +481,8 @@ static int setup_timezone(const char *de
On Sun, Dec 01, 2013 at 02:50:15PM -0800, Shawn Landden wrote:
> ---
> src/shared/strv.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/src/shared/strv.c b/src/shared/strv.c
> index 607c221..cc6adfa 100644
> --- a/src/shared/strv.c
> +++ b/src/shared/strv.c
> @@ -84,7 +8
the whitelist of dynamic linker paths comes from clang
---
man/systemd-nspawn.xml | 8 +--
src/nspawn/elf.c | 161 +
src/nspawn/elf.h | 14 +++-
src/nspawn/nspawn.c| 191 ++---
src/shared/util.c
the whitelist of dynamic linker paths comes from clang
---
man/systemd-nspawn.xml | 8 +--
src/nspawn/elf.c | 162 +
src/nspawn/elf.h | 14 +++-
src/nspawn/nspawn.c| 191 ++---
src/shared/util.c
nspawn has been called "chroot on steroids".
Continue that tradition by supporting target directories that
are not root directories.
This patch handles the simple case: a static binary.
---
Makefile.am| 2 +
man/systemd-nspawn.xml | 11 +
src/nspawn/elf.c | 131 +
---
src/shared/strv.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/src/shared/strv.c b/src/shared/strv.c
index 607c221..cc6adfa 100644
--- a/src/shared/strv.c
+++ b/src/shared/strv.c
@@ -84,7 +84,7 @@ char **strv_copy(char * const *l) {
return r;
}
-unsigned strv
---
src/nspawn/nspawn.c | 19 +++
1 file changed, 7 insertions(+), 12 deletions(-)
diff --git a/src/nspawn/nspawn.c b/src/nspawn/nspawn.c
index dd7337b..0151cf3 100644
--- a/src/nspawn/nspawn.c
+++ b/src/nspawn/nspawn.c
@@ -481,10 +481,8 @@ static int setup_timezone(const char *de
Hi
On Sun, Dec 1, 2013 at 4:13 PM, Colin Walters wrote:
> On Sun, 2013-12-01 at 01:26 +0100, Zbigniew Jędrzejewski-Szmek wrote:
>
>> Actually I don't think we need to totally forbid declarations after
>> statements.
>
> I don't have an opinion myself on making -Wdeclaration-after-statement
> an
See systemd src/core/execute.c:1299-1312, especially lines 1300 and
1307.
___
systemd-devel mailing list
systemd-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/systemd-devel
What about DISPLAY being different for different graphical sessions?
___
systemd-devel mailing list
systemd-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/systemd-devel
On Sun, Dec 1, 2013 at 8:25 PM, Hristo Venev wrote:
> User is given permissions to their user@*.service cgroup so that user mode
> systemd can run. session-*.scope cgroup permissions are required for
> session mode.
Systemd cannot delegate access to cgroupfs, it all has to happen
inside of PID1.
On Sun, Dec 1, 2013 at 8:25 PM, Hristo Venev wrote:
> I've implemented session mode for systemd. I am currently using it and it
> works
> pretty okay. I just had to generate D-Bus user services for D-Bus activation
> to
> work.
>
> A new variable, XDG_SESSION_DIR, has been added. It defaults to
On Sun, Dec 1, 2013 at 11:25 AM, Hristo Venev wrote:
> I've implemented session mode for systemd. I am currently using it and it
> works
> pretty okay. I just had to generate D-Bus user services for D-Bus activation
> to
> work.
>
> A new variable, XDG_SESSION_DIR, has been added. It defaults to
On Sat, Nov 30, 2013 at 07:48:26PM +0100, Lennart Poettering wrote:
> On Sat, 30.11.13 19:08, Lennart Poettering (lenn...@poettering.net) wrote:
>
> > < 0 → temporarily const, deep copies necessary to keep around, no free() on
> > free
> > 0 → forever const, shallow copy OK, no free() on free
>
On Sun, 2013-12-01 at 01:26 +0100, Zbigniew Jędrzejewski-Szmek wrote:
> Actually I don't think we need to totally forbid declarations after
> statements.
I don't have an opinion myself on making -Wdeclaration-after-statement
an error or not, but presently with GCC 4.7 as in gnome-continuous,
we
---
src/run/run.c | 16 +++-
1 file changed, 15 insertions(+), 1 deletion(-)
diff --git a/src/run/run.c b/src/run/run.c
index 537b725..c5c5c77 100644
--- a/src/run/run.c
+++ b/src/run/run.c
@@ -35,6 +35,7 @@ static bool arg_remain_after_exit = false;
static const char *arg_unit = NUL
Replace the conditions of some ifs that make no difference now but are
the correct behavior if session mode is added
---
src/core/dbus.c | 2 +-
src/core/main.c | 4 ++--
src/core/manager.c| 2 +-
src/core/service.c| 2 +-
src/core/unit.c | 2 +-
src/s
systemctl --session restart gnome-settings-daemon
Add a new environment variable:
XDG_SESSION_DIR=/run/session/$XDG_SESSION_ID
The session instance runs in session-*.scope and is started as a normal
process inside a session. The socket is stored in
$XDG_SESSION_DIR/systemd/private
It would be
---
src/core/unit.c | 14 ++
src/shared/path-lookup.c | 34 ++
src/shared/path-lookup.h | 2 ++
3 files changed, 46 insertions(+), 4 deletions(-)
diff --git a/src/core/unit.c b/src/core/unit.c
index c963870..029f7ee 100644
--- a/src/core/unit.
I've implemented session mode for systemd. I am currently using it and it works
pretty okay. I just had to generate D-Bus user services for D-Bus activation to
work.
A new variable, XDG_SESSION_DIR, has been added. It defaults to
/run/session/$XDG_SESSION_ID and is to be used for session-specific
More predictable names, easier to find if debugging, deleted on
session/user termination
---
src/core/manager.c | 35 ---
1 file changed, 28 insertions(+), 7 deletions(-)
diff --git a/src/core/manager.c b/src/core/manager.c
index ada62d0..617fc9f 100644
--- a/src/c
User is given permissions to their user@*.service cgroup so that user mode
systemd can run. session-*.scope cgroup permissions are required for
session mode.
---
src/core/dbus-scope.c | 6 ++
src/core/scope.c | 16
src/core/scope.h | 2 ++
src/login/
SYSTEMD_USER/SYSTEMD_SYSTEM is far more clear than true/false
---
src/analyze/analyze.c | 8
src/hostname/hostnamectl.c| 2 +-
src/libsystemd-bus/bus-util.c | 39 ---
src/libsystemd-bus/bus-util.h | 5 +++--
src/libsystemd-bus/busctl.c
> There is no reason to lose valuable debugging information. All output
> is collected by systemd and is available via journal. Hiding it makes
> really no sense.
...
> Show quoted textThis is not a shell, and you cannot reference $PATH like
> this. Full
> content of the variable has to included i
On Sun, Dec 1, 2013 at 3:43 AM, David Herrmann wrote:
> logind has no concept of session ordering. Sessions have a unique name,
> some attributes about the capabilities and that's already it. There is
> currently no stable+total order on sessions. If we use the logind API to
> switch between sessi
On Mon, Nov 25, 2013 at 11:13:35AM +0100, Daniele Medri wrote:
> Dear systemd maintainers,
>
> I've made a pull request on github with some patches:
>
> - Add new po/it.po (Italian translation)
> - Revision proposals (wording) on english labels
>
> http://github.com/systemd/systemd/pull/7
Catalo
Hi
>>> But in the case of screen I'm specifically asking for a new, stand alone
>>> session.
>>
>> I'd agree; but the fix would be fairly invasive for screen. I think
>> it'd have to become setuid root, so it could request a new session.
>
> Yeah that was my fear too.
>
> Although perhaps this is
Hi
On Sun, Dec 1, 2013 at 4:13 PM, Zbigniew Jędrzejewski-Szmek
wrote:
> On Sun, Dec 01, 2013 at 12:43:35PM +0100, David Herrmann wrote:
>> This patch introduces session "positions". A position is a simple integer
>> assigned to a session which is never changed implicitly (currently, we
>> also do
On Sun, Dec 01, 2013 at 12:10:36AM -0800, jen...@promessage.com wrote:
> I'm using an openvpn unit,
>
> cat openvpn.service
> [Unit]
> Description=OpenVPN
> After=syslog.target network.target
> Before=openvpn.target
>
> [
On Sun, Dec 01, 2013 at 12:43:35PM +0100, David Herrmann wrote:
> This patch introduces session "positions". A position is a simple integer
> assigned to a session which is never changed implicitly (currently, we
> also don't change it explicitly, but that may be changed someday). For
> seats with
Hi
On Sun, Dec 1, 2013 at 3:48 PM, Zbigniew Jędrzejewski-Szmek
wrote:
> On Sun, Dec 01, 2013 at 03:32:14PM +0100, Zbigniew Jędrzejewski-Szmek wrote:
>> On Sun, Dec 01, 2013 at 10:05:49AM +0100, David Herrmann wrote:
>> > On Sun, Dec 1, 2013 at 6:28 AM, Zbigniew Jędrzejewski-Szmek
>> > > Also, if
On Sun, Dec 01, 2013 at 03:32:14PM +0100, Zbigniew Jędrzejewski-Szmek wrote:
> On Sun, Dec 01, 2013 at 10:05:49AM +0100, David Herrmann wrote:
> > On Sun, Dec 1, 2013 at 6:28 AM, Zbigniew Jędrzejewski-Szmek
> > > Also, if the font was embedded in systemd, distributions would then
> > > remove it in
On Sun, Dec 01, 2013 at 10:05:49AM +0100, David Herrmann wrote:
> Hi
>
> On Sun, Dec 1, 2013 at 6:28 AM, Zbigniew Jędrzejewski-Szmek
> wrote:
> > On Wed, Nov 27, 2013 at 07:48:40PM +0100, David Herrmann wrote:
> >> As a first step, we add the required header+build-chain and add the
> >> font-hand
В Sun, 01 Dec 2013 00:10:36 -0800
jen...@promessage.com пишет:
I cannot answer why iptables do not work, but general comment
> with the ExecStartPre= script,
>
> cat /usr/local/etc/openvpn/up.script
>
> #!/bin/sh
> /usr/local/sbin/openvpn --rmtun --dev tun1 > /
logind has no concept of session ordering. Sessions have a unique name,
some attributes about the capabilities and that's already it. There is
currently no stable+total order on sessions. If we use the logind API to
switch between sessions, we are faced with an unordered list of sessions
we have no
Hi
On Sun, Dec 1, 2013 at 6:21 AM, Zbigniew Jędrzejewski-Szmek
wrote:
> On Wed, Nov 27, 2013 at 07:48:46PM +0100, David Herrmann wrote:
>> The test-gl helper shows how sd_gfx_card can be used to get a full OpenGL
>> context on the device. It is not added to the build-tools as it requires
>> mesa
Hi Lennart,
Thanks for pointing this out...
> There are standard dbus PropertiesChanged signals sent out for ActiveState
> changes, which invalidate the properties when they change in released
> versions of systemd, and which carry the new values along in git.
>
> We probably should document wh
Hi
On Sun, Dec 1, 2013 at 6:28 AM, Zbigniew Jędrzejewski-Szmek
wrote:
> On Wed, Nov 27, 2013 at 07:48:40PM +0100, David Herrmann wrote:
>> As a first step, we add the required header+build-chain and add the
>> font-handling. To avoid heavy font-pipelines in systemd, we only provide
>> a staticall
I'm using an openvpn unit,
cat openvpn.service
[Unit]
Description=OpenVPN
After=syslog.target network.target
Before=openvpn.target
[Service]
PrivateTmp=true
Environment=PATH="/u
43 matches
Mail list logo
