On Tue, Aug 17, 2004 at 07:58:26AM +0200, Jeff Fisher wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > On Tue, Aug 17, 2004 at 01:10:09AM +0200, Oliver Eikemeier wrote: > > > > There is no compelling reason *not* to let ports script participate in > > rcorder(8). > > > > I'm back on 4.x, so I don't have this man page... However, why not use > S###name.sh, and let the shell wildcard order them for you? It's simple, > effective, and matches what almost everybody else does, which makes it easier > to manage. It's also hard to maintain by virtue of attempting to may a simple but correction model (dependencies) into an even simpler, but wrong one (linear ordering). Please spend some time learning about the system before bitching. There are plenty of resource to learn about the new system including this writeup: http://www.daemonnews.org/200108/rcdsystem.html and of course lukem's origional paper: http://www.cs.rmit.edu.au/~lukem/papers/rc.d.pdf For that matter you can read any freebsd manpage with a simple query like: http://www.freebsd.org/cgi/man.cgi?query=rcorder > Dependencies are handled by having a standard on what number to choose; i.e. > S1xx = Adding libraries to ldpath or essential system daemons, S2xx = > Non-essential daemons with no dependencies, S3xx = Non-essential daemons with > dependencies, etc.... Someone would be the maintaner of the numbers, and give > everybody their unique number. It's not perfect, but is relatively easy to > manage. <puke> Real dependencies are much easier to manage (speaking as someone who admin'd Solaris for many years.) -- Brooks -- Any statement of the form "X is the one, true Y" is FALSE. PGP fingerprint 655D 519C 26A7 82E7 2529 9BF0 5D8E 8BE9 F238 1AD4
This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:38:06 UTC