RE: initializing types.ok (was Re: new parser)

On Wed, 12 Feb 2003, Stephen Schaefer wrote:

> My procedure is to use everything you publish in the CVS repository --
> including *your* hosts, hosts.ok, types, types.ok, filesets -- and then
> change them as needed for my environment.  This is how I came to have an
> oracle_8.1.7.3_server target name: it's in *your* types and types.ok
> file!  I don't even run oracle :-}.  But don't kill the CVS copies of
> the hosts and types files -- it's the best documentation I have, and the
> fact that it works (or at some point worked) for you makes it extremely
> valuable as a starting point.  I am only at the beginning of revising
> things to reflect my environment.  I intend to study what you're doing,
> and then revise it for my environment as necessary.  Removing activation
> of telnetd and ftpd is only the seventh or eighth step in a thousand
> mile journey.  (Although the binaries will be there: see my comments in
> the earlier discussion on the infrastructures.org list).

I've already killed the files, because they shouldn't have been published
in the first place; they're basically proprietary to my company, since
they're actual config data, and not code.  I plan on coming up with
examples, probably by going through what's in those files and removing
what I can.

> What you can't publish I'll either disable or reinvent myself.  I've
> always sought automation of system administration tasks, but this is my
> first serious use of ISconf.  I've just joined a new employer who will
> be the first major beneficiary of your work through me.  My status is: I
> need to populate isconf::filesets/common, isconf::filesets/sunos,
> isconf::filesets/linux, and isconf::filesets/domain.com.  I need to
> review the construction of /etc/services.  I need to populate the
> packages server.  All will come in time.  I'll squawk here when I think
> the problem I'm facing is something that others of us, present or
> future, may face.  I need to bring along a colleague here, so I may even
> document the nitty-gritty.

Just about everything except the configs and the actual files in the
filesets should be available.  I can tell you that my filesets aren't
actually pulling many files down.  Mostly things like resolv.conf (per
domain), ssh*_config, sudoers, a few others.  I also have a single fileset
for /usr/local/lib/perl5, so it's always consistent, and a fileset for
isconf, and stuff like that, but you should be able to figure out how
all of those were done by looking at the configs you have.

I'm certainly willing to ask questions about anything you can see, if the
progress of any of it doesn't make sense.

-- 
"Only wimps use tape backup: _real_ men just upload their important
stuff on ftp, and let the rest of the world mirror it."
                --Linus Torvalds on linux-kernel


-------------------------------------------------------
This SF.NET email is sponsored by:
SourceForge Enterprise Edition + IBM + LinuxWorld = Something 2 See!
http://www.vasoftware.com