summaryrefslogtreecommitdiff
path: root/src
AgeCommit message (Collapse)Author
2012-06-21rpm: include RPM macro definition for .spec snippetsLennart Poettering
Let's try to standardize a bit the RPM macros used for installing/uninstalling services. This only covers the non-SysV compat bits, since that tends to vary widely between the various distros. Usage: Add %{?systemd_requires} to the header of the spec file. And then: %post %systemd_post foobar.service %preun %systemd_preun foobar.service %postun %systemd_postun foobar.service And, instead of the latter, in case the service shall be restarted on updates: %postun %systemd_postun_restart foobar.service
2012-06-21login: wrap CanTTY and CanGraphical seat attributes in libsystemd-loginLennart Poettering
2012-06-21logind: introduce a state for session, being one of online, active, closingLennart Poettering
online = logged in active = logged in and session is in the fg closing = nominally logged out but some left-over processes still around Related to: https://bugzilla.gnome.org/show_bug.cgi?id=677556
2012-06-21logind: expose CanGraphical and CanTTY properties on seat objectsLennart Poettering
Since we boot so fast now that gdm might get started before the graphics drivers are properly loaded and probed we might end up announcing seat0 to gdm before it has graphics capabilities. Which will cause gdm/X11 cause to fail later on. To fix this race, let's expose CanGraphical and CanTTY fields on all seats, which clarify whether a seat is suitable for gdm resp, suitable for text logins. gdm then needs to watch CanGraphical and spawn X11 on it only if it is true. This way: USB graphics seats will expose CanGraphical=yes, CanTTY=no Machines with no graphics drivers at all, but a text console: CanGraphical=no, CanTTY=yes Machines with CONFIG_VT turned off: CanGraphical=yes, CanTTY=no And the most important case: seat0 where the graphics driver has not been probed yet boot up with CanGraphical=no, CanTTY=yes first, which then changes to CanGraphical=yes as soon as the probing is complete.
2012-06-20update TODOLennart Poettering
2012-06-20tmpfiles: exclude the first level directories in /run/user from automatic ↵Lennart Poettering
clean up It's logind's job to maintain those user dirs, so avoid automatic clean up for them. However, we do cover everything within them.
2012-06-19analyze: Python 3 compatibilityShawn Landden
Plot verb doesn't currently work b/c Cairo bug sending to stdout in python 3.
2012-06-19login: initialize XDG_RUNTIME_DIR as /run/user/$UID instead of /run/user/$USERLennart Poettering
The sssd folks would like to place the kerberos credential cache in XDG_RUNTIME_DIR, but need to do that in the PAM auth hooks, before pam_systemd is run as part of the PAM session setup. Hence, in order to make this easy for them: avoid usage of usernames, and use user IDs instead thus making an additional NSS lookup unnecessary in the kerberos bits, but still have the directory well-defined so that the kerberos bits can determine it before pam_systemd is run.
2012-06-19vconsole-setup: enable utf-8 mode explicitlyTom Gundersen
Rather than assuming the console is in utf-8 mode if nothing else is specified, be a bit more robust and enable it explicitly. This fixes a regression compared with Arch's initscripts when not using a framebuffer as the old VGA console would not be in utf-8 mode by default. Furthermore, this would allow vconsole-setup to be used after boot to change the vconsole into utf-8 mode in case it has been set to non-utf-8 mode for whatever reason. I.e, the following would leave the console in utf-8 mode as expected: # export LANG=en_US.ISO-8859-1 # /usr/lib/systemd/systemd-vconsole-setup # export LANG=en_US.UTF-8 # /usr/lib/systemd/systemd-vconsole-setup Reported-by: Xyne <xyne@archlinx.ca> Reported-by: Thomas Bächler <thomas@archlinux.org> Cc: Dave Reisner <dreisner@archlinux.org>
2012-06-19systemd: return error when asked to stop unknown unitMichal Sekletar
Command systemctl stop foo.service, will print error message, when foo.service is unknown to systemd, i.e. there is no unit file loaded for this service. https://bugzilla.redhat.com/show_bug.cgi?id=732874
2012-06-18libudev: clarify that udev_monitor_receive_device() is non-blockingKay Sievers
Based on a patch from: Sam Thursfield <ssssam@gmail.com> https://bugs.freedesktop.org/show_bug.cgi?id=51202
2012-06-17journal: fix monotonic seekingLennart Poettering
2012-06-17journal: rotate on SIGUSR2Lennart Poettering
2012-06-17journal: fix SD_JOURNAL_SYSTEM_ONLY flagLennart Poettering
2012-06-17journal: expose and make use of cutoff times of journalLennart Poettering
This helps explaining when the log output of "systemctl status" is incomplete because the logs got rotated since the service was started.
2012-06-17journal: correct list link up on hash collisionsLennart Poettering
2012-06-17journal: add basic object definition for signaturesLennart Poettering
2012-06-17journal: replace arena offset by header sizeLennart Poettering
2012-06-15logind: more robust handling of VT-less systemsMichal Schmidt
Restore the check that was removed in commit 74afee9c. Its removal caused a regression on some s390x systems where for whatever reason the device node /dev/tty0 exists and makes the preceding access() check pass.
2012-06-15service: timeout for oneshot servicesLukas Nykryn
Add possibility to specify timeout for oneshot services. [ https://bugzilla.redhat.com/show_bug.cgi?id=761656 Added minor fixups. -- michich ]
2012-06-14systemctl: clearer error message for missing install informationMichal Schmidt
Some users found it difficult to understand what systemctl was telling them. Instead of "install information" talk about "[Install] section", which is more likely to ring a bell. And suggest that it is intentional, so that users do not attempt to "correct" the unit files. https://bugzilla.redhat.com/show_bug.cgi?id=817033
2012-06-14tmpfiles: create char devices with correct SELinux contextMichal Schmidt
https://bugzilla.redhat.com/show_bug.cgi?id=824059
2012-06-13unit-name: introduce unit_dbus_path_from_name()Michal Schmidt
Use the same function in core and in systemctl. get_unit_path() in systemctl becomes unnecessary.
2012-06-13systemctl: warn about all active triggers, not just socketsMichal Schmidt
2012-06-13systemctl: fix iteration in check_listening_sockets()Michal Schmidt
2012-06-13systemctl: remove is_socket_listeningMichal Schmidt
We can use the functionality of check_unit(). Factor out check_one_unit().
2012-06-13systemctl: style fixes for the previous patchMichal Schmidt
Use the usual indentation, bracketing style, and no assignments in ifs. Since check_listening_sockets provides just optional hints for the user, don't pass its DBusErrors to the caller.
2012-06-13systemctl will print warning when stopping unitMichal Sekletar
systemctl now prints warning and list of sockets in listenning state which can trigger start of service which is about to be stopped
2012-06-12journal-file: fix mmap leakMichal Schmidt
https://bugzilla.redhat.com/show_bug.cgi?id=831132
2012-06-12logind: fix check for multiple sessionsMichal Schmidt
The "$action-multiple-sessions" polkit actions are defined as "$action while other users are logged in". To me this implies that the following sessions should not count: - greeter sessions - user sessions belonging to the same user as the one who's asking Not sure how to treat class SESSION_LOCK_SCREEN. I never have these. I just ignore every class that's not SESSION_USER. https://bugzilla.redhat.com/show_bug.cgi?id=814424
2012-06-10udev: support "udevadm info /dev/sda; udevadm info /sys/class/block/sda"Kay Sievers
2012-06-10tmpfiles: fix error messageKay Sievers
2012-06-10tmpfiles: print error if basename lookup fails; document it in manpageKay Sievers
2012-06-10tmpfiles: allow to specify basename only: systemd-tmpfiles <program.conf>Dave Reisner
Allow passing of basename only, instead of the absolute path; letting systemd-tmpfiles perform a path lookup for the proper fragment path in the config directories. This allows distributions to call: systemd-tmpfiles <program.conf> on upgrade of a package, with respecting the possibly overriden (or even masked) tmpfile.
2012-06-05readahead: Add tool to analyze the contents of the pack file. v3.Auke Kok
This patch adds code to compile 'systemd-readahead-analyze' and install it into $bindir. Use this program to parse the contents of the readahead pack file, or an arbitrary pack file and display which files are listed in it, and how much of the files are requested to be readahead. This code is not new - it's partially taken from sreadahead (formerly maintained by Arjan van der Ven and me, and was originally written by me), and adapted with the right bits to parse the systemd readahead pack files, which are slightly different in format. v2 adds a common READAHEAD_PACK_FILE_VERSION used in all the code to provide a quick way to assure all these programs are always synchronized. v3 fixes the integer math.
2012-06-05udev: remove remaining selinux labeling for file in /runKay Sievers
2012-06-04udev: kernel cmdline - accept rd.* parametersKay Sievers
2012-06-04silence gcc warning on 32 bitKay Sievers
2012-06-04systemctl: rename "man" to "help"Kay Sievers
2012-06-04udev: reset /proc/self/oom_score_adj in worker processesKay Sievers
Only the main daemon process should be excluded from OOM handling, not the worker processes or their child processes.
2012-06-04remove support for deprecated /proc/self/oom_adjKay Sievers
2012-06-04udev: limit minimum worker count to 16Kay Sievers
2012-06-04udev: lower the starting children_max -> drop the minimum required RAM from ↵Lauri Kasanen
23mb to 5mb Udev was the limiting factor for us on low-RAM systems. Given an average RSS of 180kb, 128 workers would require ~23mb of RAM. Now, please consider what happens when there is only, say, 15mb free. Udev protects itself from OOM, and the kernel can do nothing but panic. 28 workers * 0.18mb = ~5mb. This change should not affect more powerful systems much, given that they still get the addition from the amount of RAM.
2012-06-04fstab-generator: avoid mangling of mount source stringKay Sievers
This is a valid source entry in /etc/fstab: //192.168.6.10/data /data cifs noperm,auto On Mon, Jun 4, 2012 at 2:04 PM, Dave Reisner <d@falconindy.com> wrote: > On Mon, Jun 04, 2012 at 12:57:47PM +0200, Kay Sievers wrote: >> >> Changed it to use: >> path_is_absolute() >> instead of: >> is_path(), >> so that we still sanitize the input we might match against. >> >> Let me know, if you think that could still cause any problems? > Yes, this will still break CIFS shares.
2012-06-04unit-name: never create a unit name with a leading '.'Kay Sievers
Supposed to prevent creating unit files like:   ├── dev-sda1.device.wants   │   └── .dot.mount -> /run/systemd/generator/.dot.mount   ├── .dot.mount from: # cat /etc/fstab /dev/sda1 /.dot vfat ro 1 3 which we later skip reading because of the leading '.'.
2012-06-04logind: punt duplicate definition of InhibitWhatMalte Starostik
Trivial fix for: src/login/logind-inhibit.h:37:3: error: redefinition of typedef 'InhibitWhat' src/login/logind-inhibit.h:26:26: note: previous declaration of 'InhibitWhat' was here
2012-06-04fstab-generator: avoid mangling of non-path mount source and destKay Sievers
This can invalidate otherwise valid source paths with trailing slashes, such as "host:/" in the case of a network mount. Based on a patch from Dave Reisner <dreisner@archlinux.org>, which removed the slash mangling entirely. Changed it to match on the leading slash to exclude non-path values.
2012-06-04keymaps: keyboard-force-release - fix shebangDave Reisner
This script will still run without the shebang, but we won't get the intended effect of the errexit flag in the interpreter line.
2012-06-04systemd-detect-virt: fix "option '--quiet' requires an argument"Kay Sievers
https://bugs.freedesktop.org/show_bug.cgi?id=50671
2012-06-02keymaps: replace dir tokens in keyboard-force-releaseDave Reisner
Use the correct udev libexec dir as well, not systemd's.