Forge Home


A module for monitoring and applying package updates


17,482 latest version

4.3 quality score

We run a couple of automated
scans to help you access a
module's quality. Each module is
given a score based on how well
the author has formatted their
code and documentation and
modules are also checked for
malware using VirusTotal.

Please note, the information below
is for guidance only and neither of
these methods should be considered
an endorsement by Puppet.

Version information

  • 1.0.0 (latest)
  • 0.1.0
  • 0.0.1
released Nov 2nd 2016
This version is compatible with:
  • ,

Start using this module

  • r10k or Code Manager
  • Bolt
  • Manual installation
  • Direct download

Add this module to your Puppetfile:

mod 'puppetlabs-package_updates', '1.0.0'
Learn more about managing modules with a Puppetfile

Add this module to your Bolt project:

bolt module add puppetlabs-package_updates
Learn more about using this module with an existing project

Manually install this module globally with Puppet module tool:

puppet module install puppetlabs-package_updates --version 1.0.0

Direct download is not typically how you would use a Puppet module to manage your infrastructure, but you may want to download the module in order to inspect the code.



puppetlabs/package_updates — version 1.0.0 Nov 2nd 2016


Table of Contents

  1. Module Description
  2. Setup
  3. Usage
  4. Reference
  5. Limitations - OS and Puppet compatibility

Module Description

This is an experimental module that aims to enable continuous delivery of all package updates within an infrastructure across any package manager that has a Puppet provider that supports the upgradeable feature. Package information is stored in PuppetDB is inventory information and package update versions are specified in Hiera as part of a r10k change management process.

The module provides a puppet package updates subcommand to query available package updates from all package providers available on the system. The subcommand is able to query from almost all package managers out of the box and more can be added by downloading modules from the Forge that include additional package providers, such as the chocolatey/chocolatey module for Windows.

The provided package_updates class manages a cron job to scan the system for available package updates on a regular schedule. The cron job takes the output from the included puppet package updates plugin and generates a Facter fact so the package update status is always up to date in PuppetDB. Keeping the data in PuppetDB provides an easy interface to query for available updates and generate custom reports.

This module also includes a catalog terminus that searches for package update information in Hiera, and injects that information into a normally compiled catalog. This way, packages can be managed as usual as package resources in Puppet code, while the updates to those packages, and all system packages with updates NOT managed by Puppet code, can be managed as Puppet resources. Updates are continuously enforced each Puppet run, show up in the Puppet reports, and are fully auditable.


What package_updates affects

  • A cron job in the root user's crontab
  • A custom Facter fact with package update information
  • A custom catalog compiler

Setup requirements

  • Add the package_updates class to all node groups you want to monitor updates on

Beginning with package_updates

To have nodes scan for updates on a regular cadence and report the result as a custom fact, declare the package_updates class to any node or node group you'd like to monitor for updates.

To deploy package updates, declare the package_updates::pe_master class to the PE Master group in the Puppet Enterprise Console. See Patch deployment for how to define and deploy patches.


Setting a scan schedule

The module contains a single class: package_updates. This class sets up a cron job to run the puppet face and caches the result in an external fact. By default, the cron job runs every day at 3:00am. You can change that with the available class parameters.

  • minute - The minute at which to run the scan. Default: undef
  • hour - The hour at which to run the scan. Default: 3
  • month - The month of the year. Default: undef
  • monthday - The day of the month to run the scan. Default: undef
  • weekday - The day of th week to run the scan. Default: undef

Using the Puppet Command Line Interface

After installing the module on the Puppet master, each Puppet agent will pluginsync the libraries to their local file systems. Once the sync happens, you can use the following command to get a list of all the packages that have updates available.

$ puppet package updates

You can also request the output be in JSON serialized format

$ puppet package updates --render-as json

Using the custom fact

The available package updates on the system can be retrieved as a structured custom fact. Since it can take several seconds to scan the system for updates, it's preferable to scan for updates at a regular cadence and cache the results for Facter to retrieve.

The package_updates class provides a way to set a schedule for the system to scan for package updates and caches the results for Facter.

Querying infrastructure patch state

You can use the puppet query command to query the patch state for different parts of the infrastructure. For example, the following command will return all package updates for the production environment.

puppet query 'facts { name = "package_updates" and environment = "production" }'

To learn more about using PQL, go here.

Patch deployment

This module provides a catalog terminus called package_updates. The catalog terminus injects patch information into a node's commpiled catalog. To set the terminus, set the catalog_terminus setting in the master section of the /etc/puppetlabs/puppet/puppet.conf file to the value of package_updates by running the folllowing comand. Restart the puppetserver service once complete.

puppet config set catalog_terminus package_updates

For Puppet Enterprise installations, simply declare the package_updates::pe_master class in the PE Master node group in the Puppet Enterprise Console.

To apply patches to systems, a hash of package versions to be applied must be generated and added to your r10k control repository. By specifying patch information in the control repo, patches can be defined, tested, and promoted through the delivery process you already use for all other code.

The hash follows the following example yaml format:

        version: '2.9.3.el7'
        provider: 'yum'
    version: '4.8.5-4.el7'
    provider: 'yum'

The classes key in the package_updates hash contains a hash where each key is the name of a Puppet class that should have the patches specified applied to any system with that class. Any packages specified outside the classes key are assumed global and will apply to any system at all.

Using Hiera

The default terminus for retrieving patches is from Hiera. Hiera enables users to break the package_updates hash into hierarchies such as patch information for Red Hat systems vs Ubuntu or specifying patches assigned to geographical location.

The Puppet::Node::Patches indirector finds all instances of the package_updates hash in any hierarchy that applies to the node, merging all found instances of package_updates.

Report generation

Since the PuppetDB query outputs standard JSON, existing tools can be used to generate spreadsheet reports or custom interfaces can be built that renders the serialized data.

Suggested tools:

Limitations and support

This module is compatible with Puppet 4.x+ only. It makes use of the Puppet 4 parameter data type validation which is incompatible with Puppet 3.x

Setting a schedule to scan for updates on a regular schedule currently only works with non-Windows systems. Once the package_updates interface can handle both cron and scheduled_task resources, Windows support for package management systems like Chocolatey can easily be added.