System V init is fast becoming the standard in the Linux world to control the startup of software at boot time. This is because it is easier to use and more powerful and flexible than the traditional BSD init.
The init binary is located in
/sbin and not
is important as one might try and upgrade a machine to
System V init without re-installing and reformatting. The
linux kernel looks in
/etc for its init first, so you must
make sure and delete your old init from there if any.
SYSV init also differs from BSD init in that the config
files are in a subdirectory of
/etc instead of residing
/etc. This directory is called
there you will find
rc.sysinit and the following
init.dcontains a bunch of scripts. Basically, you need one script for each service you may need to start at boot time or when entering another runlevel. Services include things like networking, nfs, sendmail, httpd, etc. Services do not include things like setserial that must only be run once and then exited. Things like that should go in
rc.local should be in
/etc/rc.d if you want one. Most systems
include one even though it doesn't do much. You can also
include an rc.serial in
/etc/rc.d if you need to do serial
port specific things at boot time.
The chain of events is as follows:
initand runs the first one it finds
rc.sysinitdoes a bunch of necessary things and then runs
rc.serial(if it exists)
initruns all the scripts for the default runlevel.
The default runlevel is decided in
/etc/inittab. You should
have a line close to the top like:
From this, you'd look in the second column and see that the
default runlevel is 3, as should be the case for most systems.
If you want to change it, you can edit
/etc/inittab by hand
and change the 3. Be very careful when you are messing with the
inittab. If you do mess up, you can get in to fix it by
rebooting and doing:
LILO boot: linux single
This should allow you to boot into single user mode so you can fix it.
Now, how does it run all the right scripts? If you do an
ls -l on
rc3.d, you might see something like:
lrwxrwxrwx 1 root root 17 13:11 S10network -> ../init.d/network lrwxrwxrwx 1 root root 16 13:11 S30syslog -> ../init.d/syslog lrwxrwxrwx 1 root root 14 13:32 S40cron -> ../init.d/cron lrwxrwxrwx 1 root root 14 13:11 S50inet -> ../init.d/inet lrwxrwxrwx 1 root root 13 13:11 S60nfs -> ../init.d/nfs lrwxrwxrwx 1 root root 15 13:11 S70nfsfs -> ../init.d/nfsfs lrwxrwxrwx 1 root root 18 13:11 S75keytable -> ../init.d/keytable lrwxrwxrwx 1 root root 23 13:11 S80sendmail -> ../init.d/sendmail.init lrwxrwxrwx 1 root root 18 13:11 S90lpd -> ../init.d/lpd.init lrwxrwxrwx 1 root root 11 13:11 S99local -> ../rc.local
What you'll notice is that there are no real ``files'' in the directory. Everything there is a link to one of the scripts in the init.d directory. The links also have an ``S'' and a number at the beginning. The ``S'' means to start this particular script and a ``K'' would mean to stop it. The number is just there for ordering purposes. Init will start all the services based on the order they appear. You can duplicate numbers, but it will only confuse you somewhat. You just need to use a two digit number only, along with an upper case ``S'' or ``K'' to start or stop the services you need to.
How does it start and stop services? Simple. Each of the scripts is written to accept an argument which can be ``start'' and ``stop''. You can execute those scripts by hand in fact with a command like:
to stop the httpd server. Init just reads the name and if it has a ``K'', it calls the script with the ``stop'' argument. If it has an ``S'' it calls the script with a ``start'' argument.
Why all these runlevels?
Some people want an easy way to setup machines to be multi-purpose. I could have a ``server'' runlevel that just runs httpd, sendmail, networking, etc. Then I could have a ``user'' runlevel that runs xdm, networking, etc.