summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2011-09-23readahead: opt out of bridge logging for readahead in order not to delay ↵Lennart Poettering
things unnecessarily
2011-09-23service: fix up std output/error before we add dependencies to the bridge socketLennart Poettering
2011-09-23virt: add missing header inclusionLennart Poettering
2011-09-23condition: optionally test against type of virtualization (vm vs. container)Lennart Poettering
2011-09-23util: move virtualization detection into its own files, and extend return codesLennart Poettering
2011-09-23analyze: report startup time in plot mode as wellKoen Kooi
It now prints something like "Startup finished in 1507ms (kernel) + 850ms (userspace) = 2357ms" below the legend.
2011-09-23analyze: always draw 1s marker for scaleKoen Kooi
In situations like this: root@omap4430-panda:~# systemd-analyze Startup finished in 1499ms (kernel) + 916ms (userspace) = 2416ms The svg plot will only have the 0s marker and no subsequent markers for scale. This patch forces the 1s marker to always be drawn.
2011-09-23unit: fix complementing of requirement deps with After deps for targetsMichal Schmidt
'man systemd.target' says: Unless DefaultDependencies= is set to false, target units will implicitly complement all configured dependencies of type Wants=, Requires=, RequiresOverridable= with dependencies of type After= if the units in question also have DefaultDependencies=true. It did not work because of a forgotten negation.
2011-09-23util: don't fail if no id was passed to detect_container()Lennart Poettering
2011-09-23util: detect systemd-nspawn without relying on ns cgroup treeLennart Poettering
2011-09-23nspawn: set env var container=systemd-nspawn, following the scheme lxc ↵Lennart Poettering
introduced
2011-09-23loginctl: fix typoLennart Poettering
2011-09-23llvm-analyze: change a few things to make llvm-analyze show fewer false ↵Lennart Poettering
positives
2011-09-23llvm-analyze: fix some bugs found by llvm-analyzeLennart Poettering
2011-09-23util: improve signal_to_string, signal_from_stringMichal Schmidt
signal_to_string: Produce names for SIGRTMIN+n. Never give an "n/a" result. In the worst case give the number itself as a string. signal_from_string: Parse "RTMIN+n". Parse any valid signal number.
2011-09-23manager: fix job mode for SIGRTMIN+1, +2Michal Schmidt
The test did not work as intended and always resulted in JOB_REPLACE.
2011-09-23condition: fix reversed tests if path does not exist at allMichal Schmidt
CONDITION_PATH_IS_DIRECTORY, CONDITION_PATH_IS_SYMBOLIC_LINK and CONDITION_FILE_IS_EXECUTABLE gave reversed results when the path did not exist at all.
2011-09-23coverity: change a few things so that coverity doesn't show so many false ↵Lennart Poettering
positives
2011-09-23coverity: fix a couple of bugs found by coverityLennart Poettering
2011-09-22units: fix rescue.service race with plymouthMichal Schmidt
It was possible for the "ExecStartPre=-/bin/plymouth quit" to race with plymouth-start.service which is pulled in indirectly by basic.target -> sysinit.target. The race left plymouth running on the terminal, making it unusable for rescue purposes. https://bugzilla.redhat.com/show_bug.cgi?id=710487
2011-09-22man: document that PathChanged= does not care for simple writes but only for ↵Lennart Poettering
close()-after-write()
2011-09-22unit: don't recheck conditions when a unit is already starting but ↵Lennart Poettering
unit_start() is invoked
2011-09-22man: document pkg-config line for sd-daemonLennart Poettering
2011-09-22man: document that sd-daemon.[ch] is now available as shared libraryLennart Poettering
2011-09-22service: warn if a service fails to write its PID fileMichal Schmidt
Warn if a service promises to write a PID file (using 'PIDFile=' in the unit file or '# pidfile:' in SysV header), but fails to keep the promise. This warning will likely trigger also for the forking services with a racy daemonization, which exit the original process before the PID file is written.
2011-09-22update TODOLennart Poettering
2011-09-22selinux: reopen log when loading the SELinux policy failsLennart Poettering
2011-09-22service: if StandardInput=socket and StandardOutput=inherit imply socket for ↵Lennart Poettering
output, don't imply default output This is useful for inetd-style per-connection services, so that they again can simply specify StandardOutput=socket to connect all three fds to the socket.
2011-09-22udev: asynchronously mount fuse/confifgs as soon as the kmods are loadedLennart Poettering
2011-09-22units: introduce unit file for FUSE fsLennart Poettering
FUSE and configfs is very very similar, so handle both the same way.
2011-09-21units: In MeeGo there was more than one ExecStart lines present that caused ↵Marko Saukko
systemd to refuse the rescue.service
2011-09-21mount: pull in quota tools from fstab lines with 'quota' option, tooLennart Poettering
https://bugzilla.redhat.com/show_bug.cgi?id=736360
2011-09-21job: after converting a job from restart to start readd it to run queueLennart Poettering
When we converted a restart job it's time to look again whether any of its dependencies are now runnable. https://bugzilla.redhat.com/show_bug.cgi?id=735013
2011-09-21update TODOKay Sievers
2011-09-21units: we don't need to check whether /sys/kernel/security is already ↵Lennart Poettering
mounted since we'll enumerate that fact anyway
2011-09-21update TODOLennart Poettering
2011-09-21units: drop automount unit for /dev/hugepagesLennart Poettering
Since hugetlbfs cannot be compiled as kernel module there's little point in doing on-demand mounting via autofs for it.
2011-09-21units: drop automount unit for /dev/mqueueLennart Poettering
Since the mqueue support cannot be built as a module there's little benefit in having an autofs mount point set up for this.
2011-09-21units: drop automount unit for /sys/kernel/securityLennart Poettering
Since securityfs cannot be build as module there's little value in having an on-demand autofs mount point for it.
2011-09-21units: drop automount unit for /sys/kernel/debugLennart Poettering
Since debugfs cannot be compiled as module there's little benefit in having it as autofs mount point.
2011-09-21units: drop /sys/kernel/config automount unitLennart Poettering
The mount point directory /sys/kernel/config is only created after the module is loaded, hence there's little value in having this an automount unit: the runtime penalty for mounting an autofs here should be the same as for a real mount.
2011-09-21modules-load: filter out double modulesLennart Poettering
2011-09-21service: minor change in service_load_pid_file return valueMichal Schmidt
Return 0 only if the PID was really loaded. If no PIDFile= is defined, return -ENOENT. Only one caller cares about the return value of this function and this change makes the usage nicer.
2011-09-21condition: add ConditionPathIsSymbolicLinkMichal Schmidt
2011-09-21condition: always follow symlinks for condition checks, to mimic testLennart Poettering
2011-09-21service: handle forking services that move to a new PIDMichal Schmidt
When some forking daemons receive a SIGHUP, they re-execute themselves and consequently change to a new main PID. As long as they update the PID file in the right order (before exiting the old PID), we can detect that and avoid killing them.
2011-09-21units: don't initialize automount point for securityfs if the initrd already ↵Lennart Poettering
mounted the directory
2011-09-21unit: introduce ConditionPathIsMountPoint=Lennart Poettering
2011-09-20man: document which condition tests follow symlinksMichal Schmidt
ConditionPathExists and ConditionFileIsExecutable follow symlinks. ConditionPathIsDirectory does not follow symlinks. (XXX: is this actually the desired behavior?)
2011-09-20units: direct stdout/stderr of rescue shells to ttyMichal Schmidt
With output of services going to syslog by default now, the rescue shell units need to direct their output to tty explicitly. Specify stderr too, just in case.