Extract most of MESSAGE to README.pkgsrc.
Remove vestigial Red Hat chkconfig stanza from main control script (we
have pkgtools/rc.d-boot now).
Use CONF_FILES where applicable.
rc.d script defaults:
- Tag log entries like qmail-run's; for instance, "nbtinydns" and
"nbdnscache" become "nbdjbdns/tiny" and "nbdjbdns/cache".
- Have tinydns listen on the network.
- Auto-rebuild CDBs as needed on service start.
Bump version.
- Collapse redundant code for invoking service-specific rc.d scripts.
- Don't try to run a service's rc.d script if it isn't enabled in rc.conf.
- Prefix "nb" to procnames.
Bump version.
- Wrap long command lines for readability
- Document where we set procname=${name} for rc.d
- Detach long-running processes from controlling terminal
- Configurable path to tcpserver
- Configurable user and group names:
DJBDNS_AXFR_USER?= axfrdns
DJBDNS_CACHE_USER?= dnscache
DJBDNS_RBL_USER?= rbldns
DJBDNS_TINY_USER?= tinydns
DJBDNS_DJBDNS_GROUP?= djbdns
Bump version.
The find-prefix infrastructure was required in a pkgviews world where
packages installed from pkgsrc could have different installation
prefixes, and this was a way for a dependency prefix to be determined.
Now that pkgviews has been removed there is no longer any need for the
overhead of this infrastructure. Instead we use BUILDLINK_PREFIX.pkg
for dependencies pulled in via buildlink, or LOCALBASE/PREFIX where the
dependency is coming from pkgsrc.
Provides a reasonable performance win due to the reduction of `pkg_info
-qp` calls, some of which were redundant anyway as they were duplicating
the same information provided by BUILDLINK_PREFIX.pkg.
seed file on first startup after all. Primary rationale: it is not
desirable that the predictability of the first boot (whatever that
may happen to be) force an upper bound on the unpredictability of
random seeds for all future dnscache instances. Instead, as before,
let dnscache read 128 new bytes on every startup. Just give it
/dev/urandom instead of /dev/random.
While here, abolish references to LOCALBASE. Bump version.
respond to queries and wrote nothing to the logs. This would have
been more fun to debug if my production system hadn't been relying
on it for name resolution. The problem was a blocking read from
/dev/random.
I'm the one who put that there, over 10 years ago. Do we need it?
From my reading of <http://cr.yp.to/djbdns/dnscache-conf.html> and
dnscache-conf.c, the seed needn't consist of top-notch randomness.
A sysadmin following DJB to the letter (rather than using this
djbdns-run package) would run dnscache-conf once to configure a
dnscache instance. dnscache-conf would cook 128 bytes of randomness
-- without direct assistance from any kernel device -- and write
it to a file. dnscache would read the same file on every startup.
Therefore, we...
1. Generate a random seed iff it doesn't exist (not on every startup)
2. Generate it from /dev/urandom instead of /dev/random
...so that the configuration we generate more nearly matches that of
DJB's documented procedure, and the kind of problem I encountered
is much more difficult to trigger.
While here, add LICENSE (2-clause-bsd) and otherwise placate pkglint.
Bump version.
rcorder: Circular dependency on provision `named' in file `/etc/rc.d/ntpdate'.
Fix it by aping -r1.22 of src/etc/rc.d/named, which had replaced
"SERVERS" with "NETWORKING mountcritremote syslogd". Problem and
fix reported by Nathan Arthur.
Bump version by exactly nine years.
INSTALL/DEINSTALL script creation within pkgsrc.
If an INSTALL or DEINSTALL script is found in the package directory,
it is automatically used as a template for the pkginstall-generated
scripts. If instead, they should be used simply as the full scripts,
then the package Makefile should set INSTALL_SRC or DEINSTALL_SRC
explicitly, e.g.:
INSTALL_SRC= ${PKGDIR}/INSTALL
DEINSTALL_SRC= # emtpy
As part of the restructuring of the pkginstall framework internals,
we now *always* generate temporary INSTALL or DEINSTALL scripts. By
comparing these temporary scripts with minimal INSTALL/DEINSTALL
scripts formed from only the base templates, we determine whether or
not the INSTALL/DEINSTALL scripts are actually needed by the package
(see the generate-install-scripts target in bsd.pkginstall.mk).
In addition, more variables in the framework have been made private.
The *_EXTRA_TMPL variables have been renamed to *_TEMPLATE, which are
more sensible names given the very few exported variables in this
framework. The only public variables relating to the templates are:
INSTALL_SRC INSTALL_TEMPLATE
DEINSTALL_SRC DEINSTALL_TEMPLATE
HEADER_TEMPLATE
The packages in pkgsrc have been modified to reflect the changes in
the pkginstall framework.
And always is defined as share/examples/rc.d
which was the default before.
This rc.d scripts are not automatically added to PLISTs now also.
So add to each corresponding PLIST as required.
This was discussed on tech-pkg in late January and late April.
Todo: remove the RCD_SCRIPTS_EXAMPLEDIR uses in MESSAGES and elsewhere
and remove the RCD_SCRIPTS_EXAMPLEDIR itself.
* Add "djbdns" rc.d script to control all enabled djbdns services,
modeled after the "qmail" rc.d script from mail/qmail-run.
* Set rc.conf defaults more succinctly, modeled after mail/qmail-run.
* Add "reload" as a synonym for "cdb" in axfrdns, rbldns, and tinydns.
* Add "foo_postenv" for additional environment customization, for
instance to run dnscache as a pure forwarder.
* Add "foo_log" booleans to enable or disable logging (enabled by default).
* Always run loggers as the "dnslog" user. This makes it simpler
to define alternate loggers.
section above the inclusion of rc.subr, and express the default
assignments more succinctly (from Todd Vierling). Remove Jonathan
de Boyne Pollard's dnscache-showctl and tinydns-showctl, since they
don't apply to services run with these rc.d scripts. Bump version.
under share/examples/rc.d. The variable name already was named
RCD_SCRIPTS_EXAMPLEDIR.
This is from ideas from Greg Woods and others.
Also bumped PKGREVISION for all packages using RCD_SCRIPTS mechanism
(as requested by wiz).
are useful only for services configured in the standard way (with
{dnscache,tinydns}-conf(8)).
On second thought, don't use the G prefix for the account names.
It makes the dnscache account longer than 8 characters, which in
turn makes noise in the nightly report. (Also, it looks a little
funny.) Since this is already a heavily customized way of running
djbdns services, it's just a little more customized now.
Bump version.
rc.d scripts (inspired by Bennett Todd's Linux init.d scripts) to
run djbdns services. It also includes Jonathan de Boyne Pollard's
dnscache-showctl and tinydns-showctl scripts.
