Pull request Compare This branch is 125 commits behind kr:master.
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Failed to load latest commit information.


The usual way to run beanstalkd is to type its name
in a Unix shell prompt, like this:

    $ beanstalkd

This will start up the process and give you control over it.
You can control its output (by default output is printed to
the screen; you can arrange to have output go into file
b.log by typing ">b.log" at the end of the command line),
pause and restart the process (by pressing Control-Z and
typing "fg"), and kill it (by pressing Control-C).
This is most convenient while writing programs that use
beanstalkd (or when working on beanstalkd itself), since
you might want to start and stop it many times and regularly
inspect its output.

If you want beanstalkd to start when your operating system
boots, the mechanism varies.

Traditionally, you must add a command line to the shell script
in /etc/rc (which is read by init when the system boots),
using the "&" notation to run beanstalkd in the background.
This would suffice for most situations, but it isn't always

These days, many popular operating systems have a replacement
init program with its own configuration language. Example
configuration files for several of these are included in
subdirectories here, but the most common is probably "System V
init", which reads /etc/inittab for lines describing commands
to run at various times. If this file exists, you can add
a line something like

    bean:345:respawn:su nobody -c 'exec /usr/bin/beanstalkd'

and type "telinit q" to tell init to reread its configuration.
Type "man 5 inittab" for details of this notation.