Proof of concept for a high-availability/load-balancing Puppet infrastructure.
Ruby Puppet
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Failed to load latest commit information.

Puppet HA: Proof-of-Concept

Simply vagrant up and wait. Puppet clients agent* will be provisioned last. If all goes well, then they will run puppet agent -t once against any master via DNS round-robin and then once against each available master, individually. All three runs should succeed.

Now you can start shutting down instances of gluster, puppet or puppetdb. Reinstallation of instances should also work for agent, puppet and puppetdb.

If you don't have the time or a powerful enough development machine, have a look at the file output.txt. It contains the output of the shell provisioner for a successful run of vagrant up | egrep 'Machine booted and ready|Running provisioner|^[+]'.

What's Missing

We still need to work on a way to add/remove/reinstall gluster nodes and a high-availability/load-balancing setup for PostgreSQL (master plus hot-standby). The latter task is tracked in issue #1.


  • Vagrant (tested with 1.3.1)
  • VirtualBox (tested with 4.2)