cvs commit: ports/security/courier-authlib Makefile
flz at xbsd.org
Thu Jan 12 11:55:40 PST 2006
On Thursday 12 January 2006 20:38, Oliver Lehmann wrote:
> Florent Thoumie wrote:
> > On Thursday 12 January 2006 19:56, Oliver Lehmann wrote:
> > > oliver 2006-01-12 18:56:04 UTC
> > >
> > > FreeBSD ports repository
> > >
> > > Modified files:
> > > security/courier-authlib Makefile
> > > security/courier-authlib/files courier-authdaemond.sh.in
> > > Log:
> > > make courier-authdaemon work if it's build with MySQL support by
> > > changing the rcNg startscript.
> > >
> > > Revision Changes Path
> > > 1.22 +1 -1 ports/security/courier-authlib/Makefile
> > > 1.4 +1 -1
> > > ports/security/courier-authlib/files/courier-authdaemond.sh.in
> > No rc.d script should depend on ldconfig_compat, this doesn't exist in
> > HEAD.
> > Actually, ldconfig_compat script should be run right after ldconfig so
> > everything should be working fine. If that's not the case, I'll consider
> > adding ldconfig_compat in PROVIDE to /etc/rc.d/ldconfig but I don't
> > quite like it.
> It does but this is not guaranted - or? And what's the problem in having
> it in there? There is no provider for it on HEAD, but that hurts none?!
> One example:
> sh -c 'rcorder /etc/rc.d/* /usr/local/etc/rc.d/* 2>/dev/null'
> produces sth. like:
> Adding "ldconfig" to apache2.sh's REQUIRE doesn't move apache2.sh
> behind ldconfig_compat.
That does only prove that apache2 is started too soon, and should have
REQUIRE DAEMON instead of BEFORE: DAEMON.
Nice thing is that rcorder /etc/rc.d/* /usr/local/etc/rc.d/* doesn't produce
the same result as rcorder /usr/local/etc/rc.d/* /etc/rc.d/*.
flz at FreeBSD.org
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Size: 187 bytes
Desc: not available
Url : http://lists.freebsd.org/pipermail/cvs-ports/attachments/20060112/b680463f/attachment.bin
More information about the cvs-ports