PuppetDB collects data generated by Puppet. It enables advanced Puppet features like exported resources, and can be the foundation for other applications that use Puppet’s data.
To start using PuppetDB today:
This documentation covers PuppetDB 5, which adds several new features and contains some breaking changes since PuppetDB 4.
See the release notes for information on all changes.
Together, these give you a huge inventory of metadata about every node in your infrastructure and a searchable database of every single resource being managed on any node.
Puppet itself can search a subset of this data using
exported resources, which allow nodes to manage resources on other
nodes. This is similar to the capabilities of the legacy ActiveRecord
storeconfigs interface, but much, much faster. The remaining data is available
through PuppetDB’s query APIs (see the navigation sidebar for details).
Puppet provides PuppetDB packages and a module which simplify the setup of its SSL certificates and init scripts. The packages are available for the following operating systems:
If you’re willing to do some manual configuration, PuppetDB can run on any Unix-like OS with JVM 8 or newer, including:
Your site’s Puppet masters must be running Puppet Server 5.0.0 or later. You will need to connect your Puppet masters to PuppetDB after installing it. If you wish to use PuppetDB with standalone nodes that are running puppet apply, every node must be running 5.0.0 or later.
PuppetDB requires PostgreSQL 9.6 or later. If not provided by your distribution, compatible versions of Postgres can be installed from the PGDG (PostgreSQL Global Development Group) repositories. See apt.postgresql.org or yum.postgresql.org for more information.
You can also install a compatible version of Postgres using the puppetlabs-puppetdb module.
PuppetDB will be a critical component of your Puppet deployment and should be run on a robust and reliable server.
However, it can do a lot with fairly modest hardware. In benchmarks using real-world catalogs from a customer, a single 2012 laptop (16 GB of RAM, consumer-grade SSD, and quad-core processor) running PuppetDB and PostgreSQL was able to keep up with sustained input from 8,000 simulated Puppet nodes checking in every 30 minutes. Powerful server-grade hardware will perform even better.
The actual requirements will vary wildly depending on your site’s size and characteristics. At smallish sites, you may even be able to run PuppetDB on your Puppet master server.
For more on fitting PuppetDB to your site, see our scaling recommendations.
PuppetDB is developed openly, and is released under the Apache 2.0 license. You can get the source — and contribute to it! — at the PuppetDB GitHub repo. Bug reports and feature requests are welcome at the Puppet Labs issue tracker.