a NFS-mounted work directory (prompted by PR 115100).
Add support for logging through multilog instead of readproctitle for
the main svscan process and document the startup script variables
(loosely based on PR 111749).
PR: 111749, 115100
Submitted by: Andrew Hammond<andrew.george.hammond@gmail.com>,
Charles Sprickman <spork@bway.net>
We have not checked for this KEYWORD for a long time now, so this
is a complete noop, and thus no PORTREVISION bump. Removing it at
this point is mostly for pedantic reasons, and partly to avoid
perpetuating this anachronism by copy and paste to future scripts.
Wihle I'm here, add a STARTUP_SCRIPT knob to specify a different name for
the etc/rc.d/svscan.sh script, e.g. allowing a name like 000.svscan.sh.
PR: 81677 [1]
Submitted by: Renato Botelho <freebsd@galle.com.br> [1]
where the config files may have been loaded already and our load_rc_config
invocation may do nothing: thus, we need to honor any previously set value
of svscan_enable;
- for some reason svscan_stop_post() does not like the IFS-like vars set by
rc.subr and localpkg; use find/xargs instead of having the shell parse its
own output;
- dequote COMMENT;
- fix the daemon(8)-ization by moving 2>&1 to its proper place, so that the
errors from svscan are actually picked up by readproctitle, as explained
in <20040630175050.GD979@straylight.m.ringlet.net>;
- move the 'Starting svscan' message to its proper place, so that the PATH
set by env(1) affects svscan, not the echo command :)
script; that is, add 'start|stop' argument processing and
the ability to actually stop the script.
Bump PORTREVISION to indicate a functionality change - or rather,
to indicate a change that actually makes this a FreeBSD port..
Final push by: Dan Langille <dan@langille.org> in a somewhat
unrelated e-mail exchange
Obtained from: a svscan SysV init script that I somehow managed
to lose the source and URL for :(
the one-second sleep between invocations was moved to after the actual
invocation. This sleep could be interrupted by a SIGCHLD in the case
of a failing service, and the service could be restarted way more than
the documented once-per-second. Move the sleep back before the fork(),
just don't sleep before the very first fork, so the service starts faster.
Silence by: Daniel J. Bernstein (no telling if he'll fix this in later
daemontools releases, so I guess we'll have to do it as part
of our port..)
Approved by: nbm (maintainer)
PR. I'll leave the creation of the directory, whatever it may be, to
the user.
PR: ports/19191
Submitted by: Mario Sergio Fujikawa Ferreira <lioux@linf.unb.br>
