Forge Home


Manage PF with Puppet


1,791 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

  • 1.7.0 (latest)
  • 1.6.1
  • 1.4.1
  • 1.4.0
  • 1.1.3
  • 1.1.2
  • 1.0.0
  • 0.1.2
  • 0.1.1
  • 0.1.0
  • 0.0.2
  • 0.0.1
released May 2nd 2020
This version is compatible with:
  • Puppet Enterprise 2017.2.x, 2017.1.x, 2016.5.x, 2016.4.x
  • Puppet >= 4.1.0 < 5.0.0
  • ,

Start using this module

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

Add this module to your Puppetfile:

mod 'zleslie-pf', '1.7.0'
Learn more about managing modules with a Puppetfile

Add this module to your Bolt project:

bolt module add zleslie-pf
Learn more about using this module with an existing project

Manually install this module globally with Puppet module tool:

puppet module install zleslie-pf --version 1.7.0

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.



zleslie/pf — version 1.7.0 May 2nd 2020


Puppet Forge Build Status

A Puppet module for managing PF rules on BSD. This module is pretty basic. It only wraps the logic necessary to deploy a pf.conf file and the necessary parsing and loading of the rules deployed.


To use the PF module, you only need pass in a template.

For Puppet 3.x support, use the module version 0.1.x.

With Hiera

If you are using Hiera, the following items will take care of you.

include pf

Then set pf::template to a value that you would pass to the template() function, as you would on a file resource. For example:

pf::template: 'site/mynodepf.conf.erb'

Then for each node that uses PF, simply build a template for each node where necessary.

Fun with templates

Templates are cool for many reasons. One of them is the fact that you can include templates from inside templates. As an example, you might keep pf options, macros, and tables each in a file that is common to all your hosts. Then only use differences where needed. For example, a firewall node template might look like the following.

<%= scope.function_template(['profile/network/firewall/pf/_options.erb']) %>
<%= scope.function_template(['profile/network/firewall/pf/_macros.erb']) %>
<%= scope.function_template(['profile/network/firewall/pf/_tables.erb']) %>
<%= scope.function_template(['profile/network/firewall/pf/_nat.erb']) %>
<%= scope.function_template(['profile/network/firewall/pf/_filter.erb']) %>
<%= scope.function_template(['profile/network/firewall/pf/_filter/_siteA_ipsec.erb']) %>

# Allow sasyncd in from peer
pass in on $ext_if proto tcp from $siteA_secondary_ext to $siteA_primary_ext port {isakmp}

This allows you to put the bulk of the code in common templates that can be distributed to multiple systems, which helps reduce the number of files that need modifying to make change to a potential large number of systems. This is environment dependent.

Dynamic tables with PuppetDB

Tables in PF hold groups of addresses for speedy lookup and simplified rule sets. This combined with PuppetDB queries makes for some interesting code. You can use the pf::table defined type to specify a list of classes, who's IP addresses should be in a table.

pf::table {'ldap_servers':
    class_list => ['profile::ldap::servers'],

The above code will all a PF table entry to /etc/pf.d/ that you can simply include in your main template with a simple include "/etc/pf.d/ Now you can use the <ldap_servers> table in your rule set like you would with any other PF table.

pass in proto { tcp udp } from <local_nets> to <auth_servers> port 88
pass in proto tcp from <local_nets> to <auth_servers> port { 636 749 }

This table is populated by querying PuppetDB for all nodes who have the class profile::ldap::servers in their catalog, and returning returning the values for ipaddress and ipaddress6 from those nodes, and adding them to the table. This doesn't work for all scenarios, for example, if the IP you want to add to a table is not in either of those facts.

More flexible table data using a common class

In the case where a node's ipaddress or ipaddress6 is not the desired value to enter a table, this data can be referenced from another class. Consider the following resources

class profile::network::host ($default_address) {}

pf::table {'ldap_servers':
    class_list         => ['profile::ldap::servers'],
    common_class       => 'profile::network::host',
    common_class_param => 'default_address'

In this case, the profile::network::host::default_address can be set in hiera, on a specific node. This will cause the pf::table resource to first look for nodes that have included the profile::ldap::servers class, then from each of those nodes, look up the value of default_address on the profile::network::host class. This allows the default_address to be overridden with something more creative, like an array of facts like [networking.lo1.ip6, networking.lo1.ip4] or some such. This comes in quite handy when nodes in the environment wish to be referenced by a fact that is not the ipaddress or ipaddress6 fact.