The default init scheme in Dragora is known as Runit.
Runit is a cross-platform Unix init scheme with service supervision.
Runit is a replacement for the traditional sysvinit, and other init schemes.
- Runit was implemented in the series of Dragora 2.
Each service is associated with a service directory, and each service daemon runs as a child process of a supervising runsv process running in this directory. The runsv program provides a reliable interface for signalling the service daemon and controlling the service and supervisor. Normally the sv program is used to send commands through this interface, and to query status informations about the service.
The runsv program supervises the corresponding service daemon. By default a service is defined to be up, that means, if the service daemon dies, it will be restarted. Of course, you can tell runsv otherwise.
This reliable interface to control daemons and supervisors obsoletes pid-guessing programs, such as pidof, killall, which, due to guessing, are prone to failures by design. It also obsoletes so called pid-files, no need for each and every service daemon to include code to daemonize, to write the new process id into a file, and to take care that the file is removed properly on shutdown, which might be very difficult in case of a crash.
Clean process state
Runit guarantees each service a clean process state, no matter if the service is activated for the first time or automatically at boot time, reactivated, or simply restarted. This means that the service always is started with the same environment, resource limits, open file descriptors, and controlling terminals.
You don't necessarily have that with sysv init scripts for example. It requires a carefully written init script that reliably cleans up and sets the process state before starting the service daemon. This adds even more complexity to the init script in comparison with a run script used by runit. Many of today's init scripts don't provide a clean process state, here is an example on what could happen:
# /etc/rc.d/foo-daemon start Starting foo daemon: food.
Fine. Everything works, nothing to worry about. After rebooting the system this shows up on the screen:
Starting foo daemon: food: command not found failed. ...
The foo program is installed in /opt/foo/bin. When starting the service for the first time using the init script, the PATH environment variable contained /opt/foo/bin. After reboot init started the service using the init script, but with a different value for the PATH variable, not containing /opt/foo/bin. Of course the init script should have set PATH before starting the daemon; the problem is that it worked in the first place, and that the error didn't show up until system reboot.
With bad init scripts miraculous things could also happen when just doing at the command line:
# /etc/rc.d/foo-daemon restart
The clean process state includes open file descriptors, obsoleting the widely used hack in many service daemons to force-close all file descriptors that might be open, up to the limit of available file descriptors for the daemon process (often results in 1024 unnecessary close() system calls in a great number of service daemon implementations).
Reliable logging facility
The runsv program provides a reliable logging facility for the service daemon. If configured, runsv creates a pipe, starts and supervises an additional log service, redirects the log daemon's standard input to read from the pipe, and redirects the service daemon's standard output to write to the pipe. Restarting the service does not require restarting the log service, and vice versa.
The service daemon and the log daemon can run with different process states, and under different user id's. Runit supports easy and reliable logging for service daemons running chroot'ed.
If runsv is told to shutdown a service, e.g. at system shutdown, it ensures that the log service stays up as long as the corresponding service daemon is running and possibly writing to the log.
Fast system boot up and shutdown
After the system's one time tasks (Stage 1) are done, the system services are started up in parallel. The operating system's process scheduler takes care of having the services available as soon as possible.
On system shutdown, Stage 3 uses runsv's control interface to wait until each service daemon is terminated and all logs are written. Again, services are taken down in parallel. As soon as all services are down, system halt or system reboot is initiated.
Small code size
One of the runit project's principles is to keep the code size small. As of version 1.0.0 of runit, the runit.c source contains 330 lines of code; the runsvdir.c source is 274 lines of code, the runsv.c source 509. This minimizes the possibility of bugs introduced by programmer's fault, and makes it more easy for security related people to proofread the source code.
The runit core programs have a very small memory footprint and do not allocate memory dynamically.
The program runit is intended to run as Unix process no 1, it is automatically started by runit-init(8) through /sbin/init.
runit performs the system's booting, running and shutting down in three stages:
- Stage 1. runit starts /etc/runit/1 and waits for it to terminate. The system's one.
Time initialization tasks are done here. /etc/runit/1 has full control over /dev/console to be able to start an emergency shell in case the one time initialization tasks fail.
- Stage 2. runit starts /etc/runit/2 which should not return until the system is going to halt or reboot; if it crashes, it will be restarted. Normally, /etc/runit/2 runs runsvdir.
In Stage 2, runit optionally handles the INT signal (CTRL-ALT-DEL keyboard request).
- Stage 3. If runit is told to halt or reboot the system, or Stage 2 returns without errors, it terminates Stage 2 if it's running, and runs /etc/runit/3. The systems tasks to shutdown and halt or reboot are done here.
A description of the scheme configured on Dragora about the directories and files of Runit:
Service directory. Normally, is a symbolic link to the current runlevel in which you are actually running.
Service directory with service daemons for runit.
Contains the three stages for runit (files: 1 2 3) and the ctrlaltdel file to handle the keyboard request. Also, contains the runsvdir directory. See below.
This directory includes a sub-directory associated to a runlevel. For example: /etc/runit/runsvdir/single provides some services from /etc/sv for the single-user mode. Two symbolic links (current and previous) are provided to know in which runlevel you are, and which one was the previous level.
The /etc/rc.d directory was designated for background services and the use of other commands. Since runit, want the daemons to run in "foreground".
|runit||Performs system's booting, running, and shutdown in three stages.|
|runit-init||The first process the Kernel starts. If runit-init is started as process no 1, it runs and replaces itself with runit(8).|
|runsv||Starts and monitors a service and optionally an appendant log service.|
|runsvdir||Starts and monitors a collection of runsv(8) processes.|
|runsvchdir||Change services directory of runsvdir(8).|
|sv||Control and manage services monitored by runsv(8).|
|chpst||Runs a program with a changed process state.|
|utmpset||Logout a line from utmp and wtmp file.|
Examples to manage services with sv(8).
Query the status of a service
# sv status dhclient or
# sv check dhclient
- You can check the status of a service to see if it's running or not.
Query the status of all services
# sv status /service/* or
# sv check /service/*
- Note: If "service" doesn't start with a dot or slash and doesn't end with a slash, it is searched in the default services directory /service/, otherwise relative to the current directory.
Stopping a service
# sv stop sshd
Starting a service
# sv start sshd
Restarting a service
# sv restart sshd
Send signals to a service
# sv hup wicd
To detach a service, just remove the symbolic link under /service/ pointing to the service directory (/etc/sv/); runit recognizes the removed service within the next five seconds, then stops the service, the optional log service, and finally the supervisor process. e.g.:
# rm -v /service/xinetd
Attaching the removed service (again)
# ln -s /etc/sv/xinetd /service
Let's see some examples that are provided in the distribution:
$ cat /etc/sv/cupsd/run #!/bin/sh # # cupsd: Common unix printing system daemon. # if [ -x /usr/sbin/cupsd ]; then exec /usr/sbin/cupsd -f fi
As you can see, the first line is the shell (/bin/sh) followed by a comment. On the left side of the commentary, is the name of the service, and on the right side: a brief description. Then, a condition checks that the service is an executable. The option -f is passed to cupsd to run cupsd in "foreground".
What about a service depend on another service?
Make sure in the ./run script of the dependant service that the service it depends on is available before the service daemon starts. The sv program can be used for that. e.g. the klogd service wants the syslogd system logging service to be available before starting the klog service daemon. (Check the line 7).
$ cat -n /etc/sv/klogd/run 1 #!/bin/sh 2 # 3 # klogd: Kernel log daemon. 4 # 5 6 if [ -x /sbin/klogd ]; then 7 sv up syslogd || exit 1; 8 exec /sbin/klogd -n -c 4 -x 9 fi 10
Two runlevels are provided by default, they are called: "default" and "single". The "default" runlevel contains all the daemons included on /etc/sv. The "single" runlevel, includes one service of tty for the single-user mode. So, you can create more runlevels attaching services -- according to your needs. For example:
# mkdir /etc/runit/runsvdir/myrunlevel # ln -sv /etc/sv/getty-tty1 /etc/runit/runsvdir/myrunlevel # ln -sv /etc/sv/getty-tty2 /etc/runit/runsvdir/myrunlevel # ln -sv /etc/sv/sshd /etc/runit/runsvdir/myrunlevel # ln -sv /etc/sv/gpm /etc/runit/runsvdir/myrunlevel # ln -sv /etc/sv/klogd /etc/runit/runsvdir/myrunlevel # ln -sv /etc/sv/syslogd /etc/runit/runsvdir/myrunlevel
Switching runlevels with runit is done by switching the directory the runsvdir program is running in. This is done by the runsvchdir program, e.g. to switch to the single user runlevel, do:
# runsvchdir myrunlevel
To switch back to the default runlevel:
# runsvchdir default
To learn more about runit, see the manual pages. Also check these links: