Stop using the obsolete test(1) -o operator, use sh || instead.
Prefer sh ! over test's ! (though that one is not completely obsolete).

Avoid unnecessary use of basename(1). From Jukka Salmi a long, long time
ago on a tech-kern far, far away.

Switch device database to cdb(5). Rework ttyname(3) and ttyname_r(3) to
depend on new devname_r(3) as heart. Add /dev/pts magic directly to
devname(3). While it can lead to returning non-existing paths, the
behavior is more consistent that way. Drop caching layer in devname(3),
it doesn't buy anything for the common case of having access to the
database. Teach devname(3) proper fallback behavior of scanning /dev.
Create both old-style and new-style database for now in /etc/rc.d/sysdb.

Avoid rebuilding services database every boot time.
The database has been renamed since the default was changed to CDB format.

branches: 1.22.2; 1.22.4;
Terminate line with a '.', not the word 'done', for consistency with other
startup scripts.

now that pwd_mkdb does not change the version, warn if we still have version
0 databases.

Let an rc variable, sysdb=YES|NO, control the building of the system
databases. Defaults to YES.

check if the source exists, not that it is a file. Reported by
Blair Sadewitz

don't attempt to build the db file if the src file does not exist.

Move .db files for services and netgroup to /var/db per gimpy's request.

- don't trash utmp{,x} needlessly
- also support netgroup services and password
- pretty print

Fix for /bin/ksh, from Jukka Salmi in PR 27232.

Add an _rc_subr_loaded variable, set to ":" by rc.subr. Scripts can use this
for a speedup by doing:
$_rc_subr_loaded . /etc/rc.subr

branches: 1.13.4;
Remove kvm_mkdb, it's not used anymore.

Execute the actual command as it's set instead of calling the sh
routine directly. This allows one to to override a specific command
via the use of /etc/rc.conf.d/sysdb.

Reorganize the sysdb script so that lkm{1,2,3} can request that the
dev database be rebuilt after loading modules. Loading modules might
create new nodes in /dev, but the dev database is being initialized
too soon for lkm2 and lkm3. This means that, eg, /dev/xf86 (typically
marked AFTERMOUNT and loaded by lkm2) will show up properly. Reviewed
by lukem.

create a utmpx with the same permissions as utmp
Solves pr 18282, although very differently.

Correct the "direction" of the barrier dependencies (DAEMON, LOGIN,
NETWORKING, and SERVERS) by specifying that certain things should
come BEFORE a given barrier, rather than having the barrier REQUIRE
a service. This allows scripts to be removed without having to
edit the barrier dependencies.

As discussed on tech-userlevel, and approved by Luke.

wrap kvm_mkdb arg in "s. from adam laurie in pr 12664

Add missing -n option of sysctl(8) booted_kernel name retrival.

- avoid -a or -o logical operator of test command as possible. They would
introduce syntaxical ambiguousity, and having concatenation with && or
|| costs nothing because our sh(1) has test(1) builtin.
- use elif construct to avoid dungling else-ifs.
- while-read construct does not need enclosed by a sub-shell.
- variable detection could be done in eval args=\$ifconfig_$int
- smart variable substitution occationally saves lines and extra command
invocations.

- only perform the checkyesno on the variable named in $rcvar (rather than
implicitly using $name if $rcvar isn't set), and always perform this check,
even when using start_cmd (et al).
this check is performed before the pidcmd
is run, speeding up scripts that weren't going to be run anyway.
this should speed up booting slow systems.

- take advantage of the above and remove
start_precmd="checkyesno foo"
in scripts that use start_cmd.

- explicitly set rcvar=foo in the rc.d/foo scripts which have an equivalent
rc.conf entry

- fix `rcvar' and `restart' when $rcvar isn't set.
these above changes fix PR [bin/11027].

- when doing `force*', ignore the return value of *_precmd.
this fixes PR [bin/10781].

- rename what sysdb provides from `databases' to `sysdb', to reflect
the name of the script.

- improve the comments in rc.subr

move recreating /var/run/utmp from mountcritlocal (where /var/run is
purged but /usr/bin/install is not available) to sysdb.
problem noted by Matthias Drochner.

* add new dummy dependancy `NETWORK' to be REQUIREd by services which need
networking to be operational before starting, and use as appropriate.
NETWORK depends upon network and dhclient.
* move the guts of systemfs into mountcritlocal
* replace the dependancy on systemfs with mountcritremote, and remove the
former.
* SERVERS now also depends upon ppp

Notes:
* dhclient (and others) needs /var to be a $critical_filesystem_beforenet
* dhclient now starts before syslogd (because the latter needs /usr, and
/usr might need dhclient to be mounted)

Should fix PRs:
[install/9853] [bin/10002] [misc/10349] [port-i386/10633] [misc/10641]

branches: 1.2.4;
Use load_rc_config() (from rc.subr) instead of sourcing /etc/rc.conf.
This allows us or a user to change the configuration file method in
one place - rc.subr - without having to edit all of the rc.d/* files.

branches: 1.1.1;
Initial revision