[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: puppet and git (README)





On Tue, Jul 8, 2008 at 11:20 PM, Mike McGrath <mmcgrath redhat com> wrote:
On Tue, 8 Jul 2008, Jesse Keating wrote:

> On Tue, 2008-07-08 at 14:24 -0500, Mike McGrath wrote:
> > Ok, I've got a git puppet mechanism up and working.  Its in a very
> > simplistic form right now but works pretty well.  I'll be sending a follow
> > up email with directions and will be updating all relevant documentation.
> >
> >     -Mike
> >
>
> What mike forgot to mention, don't make any changes in puppet without
> talking to him first.
>

This is true.  If you made a change please go back in time and don't make
it :)  Actually I didn't see any commits go through so I think we're
golden.  The puppet system is now on puppet!  Our repos are all in one git
repo (minus private, still working on that).

Here's how to use it:

Old                              | New
---------------------------------|--------------------------------------
cvs -d /cvs/puppet co manifests  | git clone /git/puppet
cvs -d /cvs/puppet co configs    | git clone /git/puppet
cvs commit                       | git commit


                                     git commit _-a_ (not the same thing, doing a git add then a git commit)


make install                     | git push

That should be enough to get you by.  I've got a syntax check on post
commit and it will let you know when you've broken something.  I'm going
to try to get this into a pre-commit but there are some speed issues.

You'll also note that make install has been replaced by git push.  This is
an instant install and is much faster.  I'm going to work on the Makefiles
but make push should still work and I am going to try to get a "make test"
back to a normal working order.

The cvs repos have been moved to /cvs/puppet.disabled/

Once this move has proved working fine I'll start working on some cleanup.
I'm also working on a way to do test runs of puppet before you push on the
actual machines in question.  This is still fuzzy in my brain but I'm
pretty sure its possible.



--
Xavier.t Lamien
--
http://fedoraproject.org/wiki/XavierLamien
GPG-Key ID: F3903DEB
Fingerprint: 0F2A 7A17 0F1B 82EE FCBF 1F51 76B7 A28D F390 3DEB
[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]