feat: adding rke2 (#394)

- manage rke2 repos
- add rke2 module (init, params, install, config, service)
- split roles::infra::k8s::node -> control/compute roles
- moved common k8s config into k8s.yaml
- add bootstrap_node, manage server and token fields in rke2 config
- manage install of helm
- manage node attributes (from puppet facts)
- manage frr exclusions for service/cluster network

Reviewed-on: #394

hieradata/common.yaml

@@ -155,6 +155,9 @@ lookup_options:
   zfs::datasets:
     merge:
       strategy: deep
+  rke2::config_hash:
+    merge:
+      strategy: deep
 
 facts_path: '/opt/puppetlabs/facter/facts.d'
 

hieradata/roles/infra/k8s.eyaml

@@ -0,0 +1 @@
+rke2::node_token: ENC[PKCS7,MIIB2gYJKoZIhvcNAQcDoIIByzCCAccCAQAxggEhMIIBHQIBADAFMAACAQEwDQYJKoZIhvcNAQEBBQAEggEACQ4IIE9qT7x+KBTtvksUwvko6U/wnRn4PZEHUc91u9StIcjd0N6uH3hrJ4ddJmwikXf9Sz/5t+kSPvcVewFcjzvMHzJXqaE57KX0hqjFpqSGMlqWy4hdrZTHm/W65PkB73j21HLn9gH7m4rmqLomSbIe21P7sFM5+1pbnQejx9Us43Mu7PNy+8rQCcdEeBnRmzLaVWR8c35aXMHANqefLaLPCWjTbDY4jhcHPkNse5thpyUW89Bnq6caTJjNNQRI9U2FX2qHqwFh/rcjtGwDk9JUJvpF1qb9mdY68txMfaQ2KQh0gH67nxVf0cDppUxJaQ86Uu71BQX54l/om9AFrDCBnAYJKoZIhvcNAQcBMB0GCWCGSAFlAwQBKgQQFfKTlWnJq2zqP6jsLAzw94BwDu32IVvfvYvsuNcOJHc4ipI0PV/6AcatIDaXjK+K8BMLdWxpOVsVf4vL1zPatB5UTwc8rm1UvxWN8SbtqF8accmAA0GbIexQezsJYaAI3NT9gItGj/aJjTitl+ed7QfNCd36HVH9FEmfDaGwN7f/yw==]

hieradata/roles/infra/k8s.yaml

@@ -0,0 +1,169 @@
+---
+hiera_include:
+  - profiles::selinux::setenforce
+  - profiles::ceph::node
+  - profiles::ceph::client
+  - exporters::frr_exporter
+  - frrouting
+  - rke2
+
+# manage rke2
+rke2::bootstrap_node: prodnxsr0001.main.unkin.net
+rke2::join_url: https://join-k8s.service.consul:9345
+rke2::config_hash:
+  bind-address: "%{hiera('networking_loopback0_ip')}"
+  node-ip: "%{hiera('networking_loopback0_ip')}"
+  node-external-ip: "%{hiera('networking_loopback0_ip')}"
+  write-kubeconfig-mode: 644
+  kubelet-arg:
+    - '--node-status-update-frequency=4s'
+    - '--max-pods=100'
+  node-label:
+    - "region=%{facts.region}"
+    - "country=%{facts.country}"
+    - "asset=%{facts.dmi.product.serial_number}"
+    - "zone=%{zone}"
+    - "environment=%{environment}"
+
+# FIXME: puppet-python wants to try manage python-dev, which is required by the ceph package
+python::manage_dev_package: false
+
+profiles::packages::include:
+  bridge-utils: {}
+  cephadm: {}
+
+profiles::selinux::setenforce::mode: disabled
+
+profiles::ceph::client::manage_ceph_conf: false
+profiles::ceph::client::manage_ceph_package: false
+profiles::ceph::client::manage_ceph_paths: false
+profiles::ceph::client::fsid: 'de96a98f-3d23-465a-a899-86d3d67edab8'
+profiles::ceph::client::mons:
+  - 198.18.23.9
+  - 198.18.23.10
+  - 198.18.23.11
+  - 198.18.23.12
+  - 198.18.23.13
+
+# additional repos
+profiles::yum::global::repos:
+  ceph:
+    name: ceph
+    descr: ceph repository
+    target: /etc/yum.repos.d/ceph.repo
+    baseurl: https://edgecache.query.consul/ceph/yum/el%{facts.os.release.major}/%{facts.os.architecture}
+    gpgkey: https://download.ceph.com/keys/release.asc
+    mirrorlist: absent
+  ceph-noarch:
+    name: ceph-noarch
+    descr: ceph-noarch repository
+    target: /etc/yum.repos.d/ceph-noarch.repo
+    baseurl: https://edgecache.query.consul/ceph/yum/el%{facts.os.release.major}/noarch
+    gpgkey: https://download.ceph.com/keys/release.asc
+    mirrorlist: absent
+  frr-extras:
+    name: frr-extras
+    descr: frr-extras repository
+    target: /etc/yum.repos.d/frr-extras.repo
+    baseurl: https://packagerepo.service.consul/frr/el9/extras-daily/%{facts.os.architecture}/os
+    gpgkey: https://packagerepo.service.consul/frr/el9/extras-daily/%{facts.os.architecture}/os/RPM-GPG-KEY-FRR
+    mirrorlist: absent
+  frr-stable:
+    name: frr-stable
+    descr: frr-stable repository
+    target: /etc/yum.repos.d/frr-stable.repo
+    baseurl: https://packagerepo.service.consul/frr/el9/stable-daily/%{facts.os.architecture}/os
+    gpgkey: https://packagerepo.service.consul/frr/el9/stable-daily/%{facts.os.architecture}/os/RPM-GPG-KEY-FRR
+    mirrorlist: absent
+  rancher-rke2-common-latest:
+    name: rancher-rke2-common-latest
+    descr: rancher-rke2-common-latest
+    target: /etc/yum.repos.d/rke2-common.repo
+    baseurl: https://rpm.rancher.io/rke2/latest/common/centos/%{facts.os.release.major}/noarch
+    gpgkey: https://rpm.rancher.io/public.key
+    mirrorlist: absent
+  rancher-rke2-1-33-latest:
+    name: rancher-rke2-1-33-latest
+    descr: rancher-rke2-1-33-latest
+    target: /etc/yum.repos.d/rke2-1-33.repo
+    baseurl: https://rpm.rancher.io/rke2/latest/1.33/centos/%{facts.os.release.major}/x86_64
+    gpgkey: https://rpm.rancher.io/public.key
+    mirrorlist: absent
+
+# dns
+profiles::dns::base::primary_interface: loopback0
+
+# networking
+systemd::manage_networkd: true
+systemd::manage_all_network_files: true
+networking::interfaces:
+  "%{hiera('networking_1000_iface')}":
+    type: physical
+    ipaddress: "%{hiera('networking_1000_ip')}"
+    gateway: 198.18.15.254
+    txqueuelen: 10000
+    forwarding: true
+  "%{hiera('networking_2500_iface')}":
+    type: physical
+    ipaddress: "%{hiera('networking_2500_ip')}"
+    mtu: 1500
+    txqueuelen: 10000
+    forwarding: true
+  loopback0:
+    type: dummy
+    ipaddress: "%{hiera('networking_loopback0_ip')}"
+    netmask: 255.255.255.255
+    mtu: 1500
+  loopback1:
+    type: dummy
+    ipaddress: "%{hiera('networking_loopback1_ip')}"
+    netmask: 255.255.255.255
+    mtu: 1500
+  loopback2:
+    type: dummy
+    ipaddress: "%{hiera('networking_loopback2_ip')}"
+    netmask: 255.255.255.255
+    mtu: 1500
+
+# configure consul service
+profiles::consul::client::host_addr: "%{hiera('networking_loopback0_ip')}"
+profiles::consul::client::node_rules:
+  - resource: service
+    segment: frr_exporter
+    disposition: write
+
+# frrouting
+exporters::frr_exporter::enable: true
+frrouting::ospfd_router_id: "%{hiera('networking_loopback0_ip')}"
+frrouting::ospf_preferred_source_enable: true
+frrouting::ospf_preferred_source: "%{hiera('networking_loopback0_ip')}"
+frrouting::ospfd_redistribute:
+  - connected
+frrouting::ospfd_interfaces:
+  "%{hiera('networking_1000_iface')}":
+    area: 0.0.0.0
+  "%{hiera('networking_2500_iface')}":
+    area: 0.0.0.0
+  loopback0:
+    area: 0.0.0.0
+  loopback1:
+    area: 0.0.0.0
+  loopback2:
+    area: 0.0.0.0
+frrouting::daemons:
+  ospfd: true
+frrouting::ospf_exclude_k8s_enable: true
+frrouting::k8s_cluster_cidr: '10.42.0.0/16'  # RKE2 cluster-cidr (pods)
+frrouting::k8s_service_cidr: '10.43.0.0/16'  # RKE2 service-cidr
+
+# add loopback interfaces to ssh list
+ssh::server::options:
+  ListenAddress:
+    - "%{hiera('networking_loopback0_ip')}"
+    - "%{hiera('networking_1000_ip')}"
+    - "%{hiera('networking_2500_ip')}"
+
+profiles::ssh::sign::principals:
+  - "%{hiera('networking_loopback0_ip')}"
+  - "%{hiera('networking_1000_ip')}"
+  - "%{hiera('networking_2500_ip')}"

hieradata/roles/infra/k8s/compute.yaml

@@ -1,10 +1,3 @@
 ---
-# networking
+# manage rke2
-systemd::manage_networkd: true
+rke2::node_type: agent
-systemd::manage_all_network_files: true
-networking::interfaces:
-  eth0:
-    type: physical
-    forwarding: true
-    dhcp: true
-    mtu: 1500

hieradata/roles/infra/k8s/control.yaml

@@ -1,42 +1,73 @@
 ---
-profiles::pki::vault::alt_names:
+# manage rke2
-  - k8s-control.service.consul
+rke2::node_type: server
-  - k8s-control.query.consul
+rke2::helm_install: true
-  - "k8s-control.service.%{facts.country}-%{facts.region}.consul"
+rke2::helm_repos:
-
+  metallb: https://metallb.github.io/metallb
-profiles::ssh::sign::principals:
+  rancher-stable: https://releases.rancher.com/server-charts/stable
-  - k8s-control.service.consul
+rke2::extra_config_files:
-  - k8s-control.query.consul
+  - rke2-canal-config
-  - "k8s-control.service.%{facts.country}-%{facts.region}.consul"
+rke2::config_hash:
+  advertise-address: "%{hiera('networking_loopback0_ip')}"
+  cluster-domain: "svc.k8s.unkin.net"
+  tls-san:
+    - "join-k8s.service.consul"
+    - "api-k8s.service.consul"
+    - "api.k8s.unkin.net"
+    - "join.k8s.unkin.net"
+  cni: canal
+  cluster-cidr: 10.42.0.0/16
+  service-cidr: 10.43.0.0/16
+  cluster-dns: 10.43.0.10
+  etcd-arg: "--quota-backend-bytes 2048000000"
+  etcd-snapshot-schedule-cron: "0 3 * * *"
+  etcd-snapshot-retention: 10
+  kube-apiserver-arg:
+    - '--default-not-ready-toleration-seconds=30'
+    - '--default-unreachable-toleration-seconds=30'
+  kube-controller-manager-arg:
+    - '--node-monitor-period=4s'
+  protect-kernel-defaults: true
 
 # configure consul service
 consul::services:
-  k8s-control:
+  api-k8s:
-    service_name: 'k8s-control'
+    service_name: 'api-k8s'
-    tags:
-      - 'k8s'
-      - 'container'
     address: "%{facts.networking.fqdn}"
     port: 6443
     checks:
-      - id: 'k8s-control_https_check'
+      - id: 'api-k8s_livez_check'
-        name: 'k8s-control HTTPS Check'
+        name: 'api-k8s livez Check'
-        http: "https://%{facts.networking.fqdn}:6443"
+        args:
-        method: 'GET'
+          - sudo
-        tls_skip_verify: true
+          - /usr/local/bin/check_k8s_api.sh
+        interval: '10s'
+        timeout: '1s'
+  join-k8s:
+    service_name: 'join-k8s'
+    address: "%{facts.networking.fqdn}"
+    port: 9345
+    checks:
+      - id: 'rke2_tcp_check_9345'
+        name: 'rke2 TCP Check 9345'
+        tcp: "%{hiera('networking_loopback0_ip')}:9345"
         interval: '10s'
         timeout: '1s'
 profiles::consul::client::node_rules:
   - resource: service
-    segment: k8s-control
+    segment: api-k8s
+    disposition: write
+  - resource: service
+    segment: join-k8s
     disposition: write
 
-# networking
+profiles::pki::vault::alt_names:
-systemd::manage_networkd: true
+  - api-k8s.service.consul
-systemd::manage_all_network_files: true
+  - api-k8s.query.consul
-networking::interfaces:
+  - "api-k8s.service.%{facts.country}-%{facts.region}.consul"
-  eth0:
+
-    type: physical
+sudo::configs:
-    forwarding: true
+  consul-checks:
-    dhcp: true
+    priority: 20
-    mtu: 1500
+    content: |
+      consul ALL=(ALL) NOPASSWD: /usr/local/bin/check_k8s_api.sh

hieradata/roles/infra/k8s/node.yaml

@@ -5,6 +5,24 @@ hiera_include:
   - profiles::ceph::node
   - profiles::ceph::client
   - exporters::frr_exporter
+  - profiles::rke2::node
+
+# manage rke2
+profiles::rke2::node::servers:
+  - prodnxsr0001.main.unkin.net
+  - prodnxsr0002.main.unkin.net
+  - prodnxsr0003.main.unkin.net
+
+rke2::config_hash:
+  bind-address: "%{hiera('networking_loopback0_ip')}"
+  advertise-address: "%{hiera('networking_loopback0_ip')}"
+  node-ip: "%{hiera('networking_loopback0_ip')}"
+  node-external-ip: "%{hiera('networking_loopback0_ip')}"
+  cluster-domain: "svc.k8s.unkin.net"
+  tls-san:
+    - "api.k8s.unkin.net"
+    - "join.k8s.unkin.net"
+  cni: cilium
 
 # FIXME: puppet-python wants to try manage python-dev, which is required by the ceph package
 python::manage_dev_package: false
@@ -25,6 +43,7 @@ profiles::ceph::client::mons:
   - 198.18.23.11
   - 198.18.23.12
   - 198.18.23.13
+
 # additional repos
 profiles::yum::global::repos:
   ceph:
@@ -55,6 +74,20 @@ profiles::yum::global::repos:
     baseurl: https://packagerepo.service.consul/frr/el9/stable-daily/%{facts.os.architecture}/os
     gpgkey: https://packagerepo.service.consul/frr/el9/stable-daily/%{facts.os.architecture}/os/RPM-GPG-KEY-FRR
     mirrorlist: absent
+  rancher-rke2-common-latest:
+    name: rancher-rke2-common-latest
+    descr: rancher-rke2-common-latest
+    target: /etc/yum.repos.d/rke2-common.repo
+    baseurl: https://rpm.rancher.io/rke2/latest/common/centos/%{facts.os.release.major}/noarch
+    gpgkey: https://rpm.rancher.io/public.key
+    mirrorlist: absent
+  rancher-rke2-1-33-latest:
+    name: rancher-rke2-1-33-latest
+    descr: rancher-rke2-1-33-latest
+    target: /etc/yum.repos.d/rke2-1-33.repo
+    baseurl: https://rpm.rancher.io/rke2/latest/1.33/centos/%{facts.os.release.major}/x86_64
+    gpgkey: https://rpm.rancher.io/public.key
+    mirrorlist: absent
 
 # dns
 profiles::dns::base::primary_interface: loopback0
@@ -91,9 +124,38 @@ networking::interfaces:
     netmask: 255.255.255.255
     mtu: 1500
 
-# consul
+# configure consul service
 profiles::consul::client::host_addr: "%{hiera('networking_loopback0_ip')}"
+consul::services:
+  api-k8s:
+    service_name: 'api-k8s'
+    address: "%{facts.networking.fqdn}"
+    port: 6443
+    checks:
+      - id: 'api-k8s_https_check'
+        name: 'api-k8s HTTPS Check'
+        http: "https://%{facts.networking.fqdn}:6443"
+        method: 'GET'
+        tls_skip_verify: true
+        interval: '10s'
+        timeout: '1s'
+  join-k8s:
+    service_name: 'join-k8s'
+    address: "%{facts.networking.fqdn}"
+    port: 9345
+    checks:
+      - id: 'etcd_tcp_check_9345'
+        name: 'ETCD TCP Check 9345'
+        tcp: "%{facts.networking.fqdn}:9345"
+        interval: '10s'
+        timeout: '1s'
 profiles::consul::client::node_rules:
+  - resource: service
+    segment: api-k8s
+    disposition: write
+  - resource: service
+    segment: join-k8s
+    disposition: write
   - resource: service
     segment: frr_exporter
     disposition: write
@@ -130,3 +192,8 @@ profiles::ssh::sign::principals:
   - "%{hiera('networking_loopback0_ip')}"
   - "%{hiera('networking_1000_ip')}"
   - "%{hiera('networking_2500_ip')}"
+
+profiles::pki::vault::alt_names:
+  - api-k8s.service.consul
+  - api-k8s.query.consul
+  - "api-k8s.service.%{facts.country}-%{facts.region}.consul"

modules/frrouting/manifests/init.pp

@@ -16,8 +16,16 @@ class frrouting (
   Array[String] $mpls_ldp_interfaces         = [],
   Boolean $ospf_preferred_source_enable      = false,
   Optional[Stdlib::IP::Address] $ospf_preferred_source = undef,
+  Boolean $ospf_exclude_k8s_enable           = false,
+  Optional[Stdlib::IP::Address::V4::CIDR] $k8s_cluster_cidr = undef, # pod/cluster CIDR (e.g. 10.42.0.0/16)
+  Optional[Stdlib::IP::Address::V4::CIDR] $k8s_service_cidr = undef, # service CIDR (e.g. 10.43.0.0/16)
 ) {
 
+    # sanity check
+  if $ospf_exclude_k8s_enable and $k8s_cluster_cidr == undef and $k8s_service_cidr == undef {
+    warning('frrouting: ospf_exclude_k8s_enable is true but no k8s_*_cidr provided; nothing will be filtered.')
+  }
+
   $daemons_defaults = {
     'bgpd'    => false,
     'ospfd'   => true,

modules/frrouting/templates/frr.conf.erb

@@ -2,6 +2,7 @@
 frr defaults traditional
 hostname <%= @hostname %>
 no ipv6 forwarding
+
 <% @ospfd_interfaces.each do |iface, params| -%>
 interface <%= iface %>
 <% if params['area'] -%>
@@ -12,12 +13,31 @@ interface <%= iface %>
 <% end -%>
 exit
 <% end -%>
+
+<%# ---- K8s exclude (prefix-list + route-map) ---- -%>
+<% if @ospf_exclude_k8s_enable && (@k8s_cluster_cidr || @k8s_service_cidr) -%>
+! Do not redistribute Kubernetes Pod/Service CIDRs
+ip prefix-list K8S-BLOCK seq 5 permit <%= @k8s_cluster_cidr %> le 32<% if !@k8s_cluster_cidr %>  ! (unset)<% end %>
+<% if @k8s_service_cidr -%>
+ip prefix-list K8S-BLOCK seq 10 permit <%= @k8s_service_cidr %> le 32
+<% end -%>
+route-map CONNECTED-NON-K8S deny 5
+ match ip address prefix-list K8S-BLOCK
+exit
+route-map CONNECTED-NON-K8S permit 100
+exit
+<% end -%>
+
 router ospf
  ospf router-id <%= @ospfd_router_id %>
  log-adjacency-changes detail
 <% @ospfd_redistribute.each do |type| -%>
+<% if @ospf_exclude_k8s_enable && type == 'connected' && (@k8s_cluster_cidr || @k8s_service_cidr) -%>
+ redistribute connected route-map CONNECTED-NON-K8S
+<% else -%>
  redistribute <%= type %>
 <% end -%>
+<% end -%>
 <% @ospfd_networks.each do |network| -%>
  network <%= network %>
 <% end -%>
@@ -31,6 +51,8 @@ router ospf
  mpls-te inter-as area 0.0.0.0
 <% end -%>
 exit
+
+<%# ---- MPLS/LDP config ---- -%>
 <% if @mpls_ldp_router_id and @mpls_ldp_transport_addr and @mpls_ldp_interfaces.any? -%>
 mpls ldp
  router-id <%= @mpls_ldp_router_id %>
@@ -43,6 +65,8 @@ mpls ldp
  exit-address-family
 exit
 <% end -%>
+
+<%# ---- Preferred OSPF source ---- -%>
 <% if @ospf_preferred_source_enable && @ospf_preferred_source -%>
 ip prefix-list ANY seq 5 permit 0.0.0.0/0 le 32
 route-map OSPF-SRC permit 10

modules/rke2/files/check_k8s_api.sh

@@ -0,0 +1,2 @@
+#!/usr/bin/bash
+/var/lib/rancher/rke2/bin/kubectl --kubeconfig=/etc/rancher/rke2/rke2.yaml get --raw /livez

modules/rke2/files/rke2-canal-config.yaml

@@ -0,0 +1,10 @@
+---
+apiVersion: helm.cattle.io/v1
+kind: HelmChartConfig
+metadata:
+  name: rke2-canal
+  namespace: kube-system
+spec:
+  valuesContent: |-
+    flannel:
+      iface: "loopback0"

modules/rke2/lib/facter/helm_repos.rb

@@ -0,0 +1,15 @@
+# frozen_string_literal: true
+
+require 'facter/util/helm'
+
+Facter.add(:helm_repos) do
+  confine kernel: 'Linux'
+  confine enc_role: [
+    'roles::infra::k8s::control',
+    'roles::infra::k8s::compute'
+  ]
+
+  setcode do
+    Facter::Util::Helm.get_helm_repos('/usr/bin/helm')
+  end
+end

modules/rke2/lib/facter/util/helm.rb

@@ -0,0 +1,31 @@
+# frozen_string_literal: true
+
+require 'facter'
+require 'json'
+
+# a simple helm module
+module Facter::Util::Helm
+  def self.get_helm_repos(helm_cmd)
+    return [] unless File.executable?(helm_cmd)
+
+    output = Facter::Core::Execution.execute(
+      "#{helm_cmd} repo list --output json --repository-config /etc/helm/repositories.yaml",
+      on_fail: nil
+    )
+    return [] if output.to_s.strip.empty?
+
+    parse_helm_output(output)
+  rescue StandardError => e
+    Facter.debug("helm_repos fact error: #{e}")
+    []
+  end
+
+  def self.parse_helm_output(output)
+    JSON.parse(output).map do |repo|
+      {
+        'name' => repo['name'],
+        'url' => repo['url']
+      }
+    end
+  end
+end

modules/rke2/manifests/config.pp

@@ -0,0 +1,87 @@
+# config rke2
+class rke2::config (
+  Enum['server', 'agent'] $node_type = $rke2::node_type,
+  Stdlib::Absolutepath    $config_file = $rke2::config_file,
+  Hash                    $config_hash = $rke2::config_hash,
+  Stdlib::HTTPSUrl        $join_url = $rke2::join_url,
+  Stdlib::Fqdn            $bootstrap_node = $rke2::bootstrap_node,
+  String                  $node_token = $rke2::node_token,
+  Array[String[1]]        $extra_config_files = $rke2::extra_config_files,
+){
+
+  # if agent, add token. what other fields should i add?
+  # how can I add a tls secret using kubectl to add ephemeral certs.
+
+  # if its not the bootstrap node, add join path to config
+  if $node_type == 'server' {
+    if $trusted['certname'] != $bootstrap_node {
+      $config = merge($config_hash, {
+        server => $join_url,
+        token  => $node_token,
+      } )
+    }else{
+      $config = $config_hash
+    }
+  } elsif $node_type == 'agent' {
+      $config = merge($config_hash, {
+        server => $join_url,
+        token  => $node_token,
+      } )
+  }else{
+    $config = $config_hash
+  }
+
+  # create the config file
+  file { $config_file:
+    ensure  => file,
+    content => Sensitive($config.to_yaml),
+    owner   => 'root',
+    group   => 'root',
+    mode    => '0644',
+    require => Package["rke2-${node_type}"],
+    before  => Service["rke2-${node_type}"],
+  }
+
+  # create a script to verify k8s api is up (used by consul)
+  file {'/usr/local/bin/check_k8s_api.sh':
+    ensure => file,
+    owner  => 'root',
+    group  => 'root',
+    mode   => '0755',
+    source => 'puppet:///modules/rke2/check_k8s_api.sh',
+  }
+
+  # symlink kubectl to path
+  file {'/usr/bin/kubectl':
+    ensure  => link,
+    target  => '/var/lib/rancher/rke2/bin/kubectl',
+    require => Package["rke2-${node_type}"],
+  }
+
+  # when ProtectKernelDefaults=true
+  sysctl { 'vm.overcommit_memory':
+    value  => '1',
+    before => Service["rke2-${node_type}"],
+  }
+  sysctl { 'kernel.panic':
+    value  => '10',
+    before => Service["rke2-${node_type}"],
+  }
+
+  # on the controller nodes only
+  if $node_type == 'server' {
+
+    # manage extra config config
+    $extra_config_files.each |$file| {
+
+      file {"/var/lib/rancher/rke2/server/manifests/${file}.yaml":
+        ensure  => file,
+        owner   => 'root',
+        group   => 'root',
+        mode    => '0644',
+        source  => "puppet:///modules/rke2/${file}.yaml",
+        require => Service['rke2-server'],
+      }
+    }
+  }
+}

modules/rke2/manifests/helm.pp

@@ -0,0 +1,44 @@
+
+# manage helm
+class rke2::helm (
+  Enum['server', 'agent'] $node_type = $rke2::node_type,
+  Stdlib::Fqdn            $bootstrap_node = $rke2::bootstrap_node,
+  Boolean                 $helm_install = $rke2::helm_install,
+  Hash                    $helm_repos = $rke2::helm_repos
+){
+
+  # when installing helm, manage the repos
+  if $helm_install {
+
+    package {'helm':
+      ensure => installed,
+    }
+
+    file { '/etc/helm':
+      ensure => directory,
+      owner  => 'root',
+      group  => 'root',
+      mode   => '0755',
+    }
+
+    # on the controller nodes only
+    if $node_type == 'server' {
+
+      # check if the repo already exists
+      $helm_repos.each | String $repo, Stdlib::HTTPSUrl $url | {
+
+        # if repo isnt in repo list from helm, install it
+        if ! $facts['helm_repos'].any |$existing| { $existing['name'] == $repo } {
+
+          exec { "helm_add_repo_${repo}":
+            command     => "helm repo add ${repo} ${url} --repository-config /etc/helm/repositories.yaml",
+            path        => ['/usr/bin'],
+            environment => [
+              'KUBECONFIG=/etc/rancher/rke2/rke2.yaml',
+            ],
+          }
+        }
+      }
+    }
+  }
+}

modules/rke2/manifests/init.pp

@@ -0,0 +1,23 @@
+# manage rke2
+class rke2 (
+  Enum['server', 'agent'] $node_type = $rke2::params::node_type,
+  String                  $rke2_version = $rke2::params::rke2_version,
+  String                  $rke2_release = $rke2::params::rke2_release,
+  Stdlib::Absolutepath    $config_file = $rke2::params::config_file,
+  Hash                    $config_hash = $rke2::params::config_hash,
+  Stdlib::HTTPSUrl        $join_url = $rke2::params::join_url,
+  Stdlib::Fqdn            $bootstrap_node = $rke2::params::bootstrap_node,
+  String                  $node_token = $rke2::params::node_token,
+  Boolean                 $helm_install = $rke2::params::helm_install,
+  Hash                    $helm_repos = $rke2::params::helm_repos,
+  Array[String[1]]        $extra_config_files = $rke2::params::extra_config_files,
+  Stdlib::HTTPUrl         $container_archive_source = $rke2::params::container_archive_source,
+) inherits rke2::params {
+
+  include rke2::install
+  include rke2::config
+  include rke2::service
+  include rke2::helm
+
+  Class['rke2::install'] ->  Class['rke2::service'] -> Class['rke2::helm']
+}

modules/rke2/manifests/install.pp

@@ -0,0 +1,53 @@
+# install rke2
+class rke2::install (
+  Enum['server', 'agent'] $node_type = $rke2::node_type,
+  String                  $rke2_version = $rke2::rke2_version,
+  String                  $rke2_release = $rke2::rke2_release,
+  Stdlib::HTTPUrl         $container_archive_source = $rke2::container_archive_source,
+){
+
+  # versionlock rke2
+  yum::versionlock{"rke2-${node_type}":
+    ensure  => present,
+    version => "${rke2_version}~${rke2_release}",
+  }
+
+  # install rke2
+  package {"rke2-${node_type}":
+    ensure => "${rke2_version}~${rke2_release}",
+  }
+
+  # ensure images path exists
+  file { ['/var/lib/rancher/rke2/agent', '/var/lib/rancher/rke2/agent/images']:
+    ensure  => 'directory',
+    owner   => 'root',
+    group   => 'root',
+    mode    => '0750',
+    require => Package["rke2-${node_type}"],
+    before  => Service["rke2-${node_type}"],
+  }
+
+  # download required archive of containers
+  archive { '/var/lib/rancher/rke2/agent/images/rke2-images.linux-amd64.tar.zst':
+    ensure  => present,
+    source  => "https://github.com/rancher/rke2/releases/download/v${rke2_version}%2B${rke2_release}/rke2-images.linux-amd64.tar.zst",
+    require => [
+      Package["rke2-${node_type}"],
+      File['/var/lib/rancher/rke2/agent/images'],
+    ],
+    before  => Service["rke2-${node_type}"],
+  }
+
+  # ensure the images cache file exists
+  file {'/var/lib/rancher/rke2/agent/images/.cache.json':
+    ensure  => file,
+    owner   => 'root',
+    group   => 'root',
+    mode    => '0644',
+    require => [
+      Package["rke2-${node_type}"],
+      File['/var/lib/rancher/rke2/agent/images'],
+    ],
+    before  => Service["rke2-${node_type}"],
+  }
+}

modules/rke2/manifests/params.pp

@@ -0,0 +1,15 @@
+# rke2 params
+class rke2::params (
+  Enum['server', 'agent'] $node_type = 'agent',
+  String                  $rke2_version = '1.33.4',
+  String                  $rke2_release = 'rke2r1',
+  Stdlib::Absolutepath    $config_file = '/etc/rancher/rke2/config.yaml',
+  Hash                    $config_hash = {},
+  Stdlib::HTTPSUrl        $join_url = 'https://127.0.0.1:9345',
+  Stdlib::Fqdn            $bootstrap_node = 'localhost.localdomain',
+  String                  $node_token = '',
+  Boolean                 $helm_install = false,
+  Hash                    $helm_repos = {},
+  Array[String[1]]        $extra_config_files = [],
+  Stdlib::HTTPUrl         $container_archive_source = 'https://github.com/rancher/rke2/releases/download',
+) {}

modules/rke2/manifests/service.pp

@@ -0,0 +1,13 @@
+# manage rke2 service
+class rke2::service (
+  Enum['server', 'agent'] $node_type = $rke2::node_type,
+  Stdlib::Absolutepath    $config_file = $rke2::config_file,
+){
+
+  service {"rke2-${node_type}":
+    ensure    => true,
+    enable    => true,
+    subscribe => File[$config_file],
+  }
+
+}

site/profiles/manifests/dns/base.pp

@@ -47,7 +47,7 @@ class profiles::dns::base (
   $facts['networking']['interfaces'].each | $interface, $data | {
 
     # exclude those without ipv4 address, lo, docker0 and anycast addresses
-    if $data['ip'] and $interface != 'lo' and $interface != 'docker0' and $interface !~ /^anycast[0-9]$/ {
+    if $data['ip'] and $interface != 'lo' and $interface != 'docker0' and $interface !~ /^anycast[0-9]$/ and $interface !~ /^cilium_/  {
 
       # use defaults for the primary_interface
       if $interface == $primary_interface {