This will be pushed out to the clients as long as: pluginsync = true

For me the next step is to create a manifests/init.pp in the new module directory. This is important to me because I want to sync out a /usr/local/etc/pkg.conf to all my machines so that they point to my internal poudriere repos. So I end up with something like this:

file { "/usr/local/etc/pkg.conf":
        mode => 755,
        owner => root,
        content => "packagesite: http://pkg/91-web/
",
}

Once that is done it is easy to use pkgng packages via:

package { "www/apache22":
        ensure => installed,
        provider => pkgng,
        require => File['/usr/local/etc/pkg.conf'],
}

Tools:

Collectd: https://collectd.org/

Graphite: http://graphite.wikidot.com/
Nagios: http://www.nagios.org/

Poudriere: http://fossil.etoilebsd.net/poudriere

Config Management:
Salt Stack: http://saltstack.com/
Chef: http://www.opscode.com/chef/
Puppet: http://puppetlabs.com/

Subversion: http://subversion.apache.org/

LogStash: http://logstash.net/
Audit: http://www.freebsd.org/handbook/audit.html

CARP: http://www.freebsd.org/handbook/carp.html

OATH: http://www.openauthentication.org/

Serial Console: http://www.freebsd.org/handbook/serialconsole-setup.html

Generic Resources:
FreeBSD Handbook: http://freebsd.org/handbook
Everything Sysadmin Blog: http://everythingsysadmin.com/resources.html

Over the years, I have done serial console in many ways. I began with a single null modem cable between the back of two servers. Next, I utilized a RocketPort multi-port serial card with 8 serial ports on it. These days, I have moved on to employing big serial console servers such as those made by OpenGear, providing up to 48 ports. They also have ancillary features such as providing a Nagios platform and Environmental monitoring.

No matter your physical connectivity, I recommend using Conserver. This helps by logging what is happening on the console, which can be very handy if you need to see what happened in the past whether it be a function of the system, or to see who did what. It also provides multi-user access, so you can watch while someone else is working and both of you can collaborate on fixing a problem.

In order for the previous technologies to be useful, the servers require configuration as well. The first step is to configure the BIOS for serial console redirection. Once this has been performed, the OS will need to be configured to present a console login via the serial port. The FreeBSD Handbook explains how to do this Here.

http://freebsd.so14k.com/freebsd9_scripted_install.shtml

It is pretty basic, here are some of the highlights:

1. GPT disk layout
2. ZFS Only Install (which could be easily converted to UFS)
3. The Script is nice and short!

http://freebsd.so14k.com/freebsd9_pxe.shtml

In FreeBSD 9, a few things have changed. If you have an old PXE environment from FreeBSD 8, you will want to make note of the following:

• No more mfsroot.
• Which means, no more changes to /boot/loader.conf, it should be empty infact.
• You need the new pxeboot binary from 9, do not try using an old one.

Anoncvs I have not gotten around to setting up yet, hopefully soon.

The commit is here: https://github.com/puppetlabs/puppet/pull/338

I opened a ticket in the Puppet Bug tracker, 11318. Then I found out that someone by the nick of tdb had already incorporated our changes into another pull request that adds more functionality and some unit tests. So hopefully this will be committed soon and we can have this support upstream.

I just wanted to thank tdb for taking this work and running with it!

I have setup the hardware and I am prepared to ship the server out. Just need to confirm the new IP info and ship it out. Hopefully I will get this done this week prior to heading out to LISA.