Ok, so my company was much more amenable to OpenVox than I anticipated, which is great. As a result, I'm delegating a parallel deployment of OpenVox (I already have the puppet deployment fully automated, and if OpenVox is a drop-in replacement, the effort is trivial). I still need to complete the POC with OSS Puppet, but I wanted to get a bit more insight into the viability of replacing what I've built using OSS Puppet with OpenVox, from a design standpoint. Here's the design: • One HAProxy server to route traffic to the catalog compilation servers. (routing/access layer) • Three catalog compilation servers (One Primary Puppet + CA Master, Two Compile Servers) (I refer to this as the processing layer) • Two PuppetDBs for indexing and Bolt inventory purposes, as well as reducing DB load. (indexing layer) • Two PostgreSQL DB Servers (a primary and a warm standby with archives backups going to cloud storage) (data layer) From a purely Puppet perspective, I only need to account for the processing and indexing layers as far as replacing it with OpenVox, so my main questions would be: • How does Bolt work in an OpenVox implementation? Would I still use the Puppet Bolt I already have or is there an OpenVox equivalent? (This is vital to the design). • Same with PuppetDB, really. OpenVox equivalent or use current binary? (This is also vital... mostly because of Bolt)