The current check only tries to detect whether /sys/fs/cgroup exists and
whether it is writable or not. But when the init system doesn't mount
cgroups then /sys/fs/cgroup will just be an empty directory. When paired
with unprivileged containers that mount sysfs this will cause misleading
errors to be printed since /sys/fs/cgroup will be owned by user
nobody:nogroup in this case. Independent of this specific problem this
check will also be misleading when the /sys/fs/cgroup exists and is in
fact writable by the init system but isn't actually a mountpoint.
Note from William. "grep -qs" doesn't need to redirect output to
/dev/null since it is completely silent.
This fixes#209.
Add the following variables to expose more arguments that can be passed
to start-stop-daemon or supervise-daemon:
- directory will be passed to --chdir
- error_log will be passed to --stderr
- output_log will be passed to --stdout
- umask will be passed to umask
This is for #184.
Add the ability to force-kill a service if it does not go down
successfully. Also, adjust the default wait time for an s6 service to go
down to 60 seconds.
This is to be used if the service is being supervised and the
supervisor is somehow killed.
Currently, this is very linux specific, but I will expand to other
platforms, patches are welcome.
Instead of looping and sending multiple signals to child processes in
cgroup_cleanup, we send sigterm followed by sleeping one second then
sigkill.
This brings us more in line with systemd's "control group" killmode
setting.
Also, this commit includes several shellcheck cleanups.
The --retry option for supervise-daemon defines how the supervisor will
attempt to stop the child process it is monitoring. It is defined when
the supervisor is started since stopping the supervisor just sends a
signal to the active supervisor.
This fixes#160.
This creates --respawn-delay, --respawn-max and --respawn-period. It was
suggested that it would be easier to follow if the options were
separated.
This is for #126.
Allow limiting the number of times supervise-daemon will attempt to respawn a
daemon once it has died to prevent infinite respawning. Also, set a
reasonable default limit (10 times in a 5 second period).
This is for issue #126.
We do not need to care about the path on the shebang line of a service
script as long as the shebang line ends with "openrc-run".
This fixes#119 and #120.
Supervisor setups break easily when start/stop/status functions are not
default.
Applications that write multiple PIDs to a pidfile (eg HAProxy as
described in bug 601540), can also benefit from being able to call the
default start/stop/status with modified environment variables.
Expose the default start/stop/status functions as
default_start/stop/status, and use them for the defaults
start/stop/status.
Trivial usage example:
```
stop()
{
t=$(mktemp)
for pid in $(cat $pidfile) ; do
echo $pid >$t
pidfile=$t default_stop
done
rm -f $t
}
```
X-Gentoo-Bug: 601540
X-Gentoo-Bug-URL: https://bugs.gentoo.org/601540
Signed-off-by: Robin H. Johnson <robbat2@gentoo.org>
This allows us to avoid the warnings from bash-4.4 about null bytes in
command substitutions.
If you have separate /usr, are not using an initramfs, and have a file
called /proc/self/environ on your root file system, this will break.
X-Gentoo-Bug: 594534
X-Gentoo-Bug-URL: https://bugs.gentoo.org/show_bug.cgi?id=594534
btrfs support is not implemented yet (for q Q v), but at least tmpfiles.sh
no longer chokes about tmpfiles.d lines of recent systemd versions
This fixes#87.
The read builtin in most shells will interpret backslash characters
as escapes, and they are lost when reading binfmt files line-by-line.
This causes magic strings containing backslashes to be mangled and
become invalid, resulting in erroneous 'invalid entry' messages.
The -r option to read disables special handling of backslashes and
keeps all lines intact.
X-Gentoo-Bug: 575114
X-Gentoo-Bug-URL: https://bugs.gentoo.org/show_bug.cgi?id=575114
The s6-svc syntax changed for wait-up, wait-ready, wait-down, and
wait-finished. This changes the s6 handling script to use the current
valid syntax.
This fixes#65.
In the past, OpenRC was a hybrid of a centralized and file-scope
license/copyright structure.
I followed the instructions from the Software Freedom Law Center [1] to
convert to a Centralized structure where possible, for easier future
maintenance.
[1] https://softwarefreedom.org/resources/2012/ManagingCopyrightInformation.html
- gendepends.sh needs to read this directory to allow dependencies to be
overridden
- init.sh for Linux and Bsd need to read it to allow config settings
they use to be overridden.
The want dependency is similar to the use dependency. If a service
script, for example called service1, adds "want service2" to its depend
function, OpenRC will attempt to start service2, if it exists on the
system, when service1 is started.
However, service1 will start regardless of the status of
service2.
X-Gentoo-Bug: 406021
X-Gentoo-Bug-URL: https://bugs.gentoo.org/show_bug.cgi?id=406021
This makes it possible to override settings in rc.conf by adding a
directory @SYSCONFDIR@/rc.conf.d and putting files in this directory.
The files will be processed in lexical order, and the last setting in
these files will be used.
We were starting the value we write to the cgroup setting file with
leading spaces and this was causing issues. This change makes sure that
we aren't adding leading spaces to the value.
X-Gentoo-Bug: 562354
X-Gentoo-Bug-URL: https://bugs.gentoo.org/show_bug.cgi?id=562354
The default start-stop-daemon start function expects the command
variable to be defined to point to the daemon we want to start.
If the variable is undefined, this means that there will be nothing to
start, and in this case we should complain because it is possible that
the script writer made a typo in the variable name.
The tmpfiles "d" entry will create a full path and only the last dir in
the path will have its SELinux label set correctly. This patch will
restorecon the parents as well so that the selinux labels are correct.
eg, "d /run/libvirt/lxc", then "lxc" would have the correct SELinux
label but "libvirt" would not.
Signed-off-by: Jason Zaman <jason@perfinion.com>
This change does NOT implement btrfs subvol creation. Instead, it
treats 'v' the same as 'd', which is an acceptable fallback
according to the manual.
Fixes#58
This changes the default s6 service directory to /var/svc.d, also
it changes the code to work with the individual services instead of
forcing a rescan when a service is started or stopped.
The original way of doing this allowed users to change the supervisor in
conf.d/*. This changes this so that the supervisor setup can be done in
the service script itself.
This makes binfmt processing behave like tmpfiles processing which
follows the same specification as systemd.
This fixes#48.
X-Gentoo-Bug: 545162
X-Gentoo-Bug-URL: https://bugs.gentoo.org/show_bug.cgi?id=545162
Tmpfiles.d processing had /run overriding /usr/lib and /etc, but this is
not correct. The correct order, from lowest to highest, for tmpfiles
processing is:
* /usr/lib/tmpfiles.d/*.conf
* /run/tmpfiles.d/*.conf
* /etc/tmpfiles.d
This means /run/tmpfiles.d/*.conf can override /etc/tmpfiles.d/*.conf,
but /etc/tmpfiles.d/*.conf can override both of them.
This fixes#49.
This adds support for a chroot variable which will be passed to the
start-stop-daemon --chroot switch to runscript.sh when starting a
daemon. This also needs to be saved so it can be used in locating the
pid file when stopping the daemon.
X-Gentoo-Bug: 524388
X-Gentoo-Bug-URL: https://bugs.gentoo.org/show_bug.cgi?id=524388
Note from William Hubbs:
I spoke with Roy about this, and he pointed out that user-defined
functions may need the limits applied, so it is better to go with a
method that uses exceptions to determine which functions apply the
limits.
X-Gentoo-Bug: 522408
X-Gentoo-Bug-URL: https://bugs.gentoo.org/show_bug.cgi?id=522408
Device nodes are normally never device_t so this type does not
have many permissions. After the mknod, the device should have
its label corrected before any other operations (like chmod).
install is in /usr which causes problems if /usr is not mounted.
Instead, checkpath and "mkdir -p" can do everything required and are
both available before /usr is mounted.
Since checkpath also handles selinux labels correctly,
_restorecon after is not required.
X-Gentoo-Bug: 503408
X-Gentoo-Bug-URL: https://bugs.gentoo.org/show_bug.cgi?id=503408
Status call should not set limits as it requires root permissions,
also this is not safe, as current process may reach limitation.
Solution is to set limits and move process to service cgroup only
on start.
X-GENTOO-BUG: 500364
X-GENTOO-BUG-URL: https://bugs.gentoo.org/show_bug.cgi?id=500364
