feat: adding rke2

- manage rke2 repos
- add rke2 module (init, params, install, config, service)
- exclude setting ips for cilium interfaces
- 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

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,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]

hieradata/roles/infra/k8s.yaml

@@ -0,0 +1,157 @@
+---
+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
+
+# 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
+
+# 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 +1,55 @@
 ---
-profiles::pki::vault::alt_names:
+# manage rke2
-  - k8s-control.service.consul
+rke2::node_type: server
-  - k8s-control.query.consul
+rke2::config_hash:
-  - "k8s-control.service.%{facts.country}-%{facts.region}.consul"
+  advertise-address: "%{hiera('networking_loopback0_ip')}"
-
+  cluster-domain: "svc.k8s.unkin.net"
-profiles::ssh::sign::principals:
+  tls-san:
-  - k8s-control.service.consul
+    - "join-k8s.service.consul"
-  - k8s-control.query.consul
+    - "api-k8s.service.consul"
-  - "k8s-control.service.%{facts.country}-%{facts.region}.consul"
+    - "api.k8s.unkin.net"
+    - "join.k8s.unkin.net"
+  cni: cilium
 
 # 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/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/manifests/config.pp

@@ -0,0 +1,57 @@
+# 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,
+){
+
+  # 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',
+  }
+
+  # 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',
+  }
+
+}

modules/rke2/manifests/init.pp

@@ -0,0 +1,16 @@
+# manage rke2
+class rke2 (
+  Enum['server', 'agent'] $node_type = $rke2::params::node_type,
+  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,
+) inherits rke2::params {
+
+  include rke2::install
+  include rke2::config
+  include rke2::service
+
+  Class['rke2::install'] -> Class['rke2::config'] -> Class['rke2::service']
+}

modules/rke2/manifests/install.pp

@@ -0,0 +1,10 @@
+# install rke2
+class rke2::install (
+  Enum['server', 'agent'] $node_type = $rke2::node_type,
+){
+
+  package {"rke2-${node_type}":
+    ensure => installed,
+  }
+
+}

modules/rke2/manifests/params.pp

@@ -0,0 +1,9 @@
+# rke2 params
+class rke2::params (
+  Enum['server', 'agent'] $node_type = 'agent',
+  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 = '',
+) {}

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 {