dns: dual-write toggles + drift fact

Publish records both ways during the k8s cutover, and expose expected vs
deployed records for drift detection.

- profiles::dns::updater + ::record: manage_nsupdate and manage_export
  booleans (both default on); export keeps the legacy master flow, so
  disable it once k8s is authoritative
- dns_records fact: parses the expected records file and digs the
  authoritative server for each, reporting expected / in_sync / drift
  (plus dns_records_insync boolean); updater writes the server address
  to /var/lib/dns-updater/server for the fact
- hiera: manage_export/manage_nsupdate = true (cutover)

.woodpecker/bolt-validate.yaml

@@ -0,0 +1,18 @@
+when:
+  - event: pull_request
+
+steps:
+  - name: bolt-validate
+    image: git.unkin.net/unkin/almalinux9-puppet-pr-validator:20260317
+    commands:
+      - uvx pre-commit run --all-files --config ci/bolt-validate.yaml
+    backend_options:
+      kubernetes:
+        serviceAccountName: default
+        resources:
+          requests:
+            memory: 512Mi
+            cpu: 1
+          limits:
+            memory: 2Gi
+            cpu: 1

.woodpecker/epp-validate.yaml

@@ -0,0 +1,18 @@
+when:
+  - event: pull_request
+
+steps:
+  - name: epp-validate
+    image: git.unkin.net/unkin/almalinux9-puppet-pr-validator:20260317
+    commands:
+      - uvx pre-commit run --all-files --config ci/epp-validate.yaml
+    backend_options:
+      kubernetes:
+        serviceAccountName: default
+        resources:
+          requests:
+            memory: 512Mi
+            cpu: 1
+          limits:
+            memory: 2Gi
+            cpu: 1

.woodpecker/erb-validate.yaml

@@ -0,0 +1,18 @@
+when:
+  - event: pull_request
+
+steps:
+  - name: erb-validate
+    image: git.unkin.net/unkin/almalinux9-puppet-pr-validator:20260317
+    commands:
+      - uvx pre-commit run --all-files --config ci/erb-validate.yaml
+    backend_options:
+      kubernetes:
+        serviceAccountName: default
+        resources:
+          requests:
+            memory: 512Mi
+            cpu: 1
+          limits:
+            memory: 2Gi
+            cpu: 1

.woodpecker/pre-commit.yaml

@@ -1,10 +0,0 @@
-when:
-  - event: pull_request
-
-steps:
-  - name: pre-commit
-    image: git.unkin.net/unkin/almalinux9-base:latest
-    commands:
-      - dnf groupinstall -y "Development Tools" -y
-      - dnf install uv rubygems ruby-devel gcc make redhat-rpm-config glibc-headers glibc-devel libffi libffi-devel -y
-      - uvx pre-commit run --all-files

.woodpecker/puppet-lint.yaml

@@ -0,0 +1,18 @@
+when:
+  - event: pull_request
+
+steps:
+  - name: puppet-lint
+    image: git.unkin.net/unkin/almalinux9-puppet-pr-validator:20260317
+    commands:
+      - uvx pre-commit run --all-files --config ci/puppet-lint.yaml
+    backend_options:
+      kubernetes:
+        serviceAccountName: default
+        resources:
+          requests:
+            memory: 512Mi
+            cpu: 1
+          limits:
+            memory: 2Gi
+            cpu: 1

.woodpecker/puppet-validate.yaml

@@ -0,0 +1,18 @@
+when:
+  - event: pull_request
+
+steps:
+  - name: puppet-validate
+    image: git.unkin.net/unkin/almalinux9-puppet-pr-validator:20260317
+    commands:
+      - uvx pre-commit run --all-files --config ci/puppet-validate.yaml
+    backend_options:
+      kubernetes:
+        serviceAccountName: default
+        resources:
+          requests:
+            memory: 512Mi
+            cpu: 2
+          limits:
+            memory: 2Gi
+            cpu: 2

.woodpecker/ruby-check.yaml

@@ -0,0 +1,18 @@
+when:
+  - event: pull_request
+
+steps:
+  - name: ruby-check
+    image: git.unkin.net/unkin/almalinux9-puppet-pr-validator:20260317
+    commands:
+      - uvx pre-commit run --all-files --config ci/ruby-check.yaml
+    backend_options:
+      kubernetes:
+        serviceAccountName: default
+        resources:
+          requests:
+            memory: 512Mi
+            cpu: 1
+          limits:
+            memory: 2Gi
+            cpu: 1

.woodpecker/ruby-validate.yaml

@@ -0,0 +1,18 @@
+when:
+  - event: pull_request
+
+steps:
+  - name: ruby-validate
+    image: git.unkin.net/unkin/almalinux9-puppet-pr-validator:20260317
+    commands:
+      - uvx pre-commit run --all-files --config ci/ruby-validate.yaml
+    backend_options:
+      kubernetes:
+        serviceAccountName: default
+        resources:
+          requests:
+            memory: 512Mi
+            cpu: 1
+          limits:
+            memory: 2Gi
+            cpu: 1

.woodpecker/yamllint.yaml

@@ -0,0 +1,18 @@
+when:
+  - event: pull_request
+
+steps:
+  - name: yamllint
+    image: git.unkin.net/unkin/almalinux9-base:20260606
+    commands:
+      - uvx pre-commit run --all-files --config ci/yamllint.yaml
+    backend_options:
+      kubernetes:
+        serviceAccountName: default
+        resources:
+          requests:
+            memory: 512Mi
+            cpu: 1
+          limits:
+            memory: 2Gi
+            cpu: 1

Puppetfile

@@ -53,7 +53,6 @@ mod 'saz-ssh', '13.1.0'
 mod 'saz-limits', '5.0.0'
 mod 'ghoneycutt-timezone', '4.0.0'
 mod 'ghoneycutt-puppet', '3.3.0'
-mod 'dalen-puppetdbquery', '3.0.1'
 mod 'markt-galera', '3.1.0'
 mod 'kogitoapp-minio', '1.1.4'
 mod 'broadinstitute-certs', '3.0.1'

ci/bolt-validate.yaml

@@ -0,0 +1,5 @@
+repos:
+  - repo: 'https://github.com/chriskuehl/puppet-pre-commit-hooks.git'
+    rev: v2.2.0
+    hooks:
+      - id: bolt-validate

ci/epp-validate.yaml

@@ -0,0 +1,5 @@
+repos:
+  - repo: 'https://github.com/chriskuehl/puppet-pre-commit-hooks.git'
+    rev: v2.2.0
+    hooks:
+      - id: epp-validate

ci/erb-validate.yaml

@@ -0,0 +1,5 @@
+repos:
+  - repo: 'https://github.com/chriskuehl/puppet-pre-commit-hooks.git'
+    rev: v2.2.0
+    hooks:
+      - id: erb-validate

ci/puppet-lint.yaml

@@ -0,0 +1,10 @@
+repos:
+  - repo: 'https://github.com/chriskuehl/puppet-pre-commit-hooks.git'
+    rev: v2.2.0
+    hooks:
+      - id: puppet-lint
+        args:
+          - --no-80chars-check
+          - --no-documentation-check
+          - --no-puppet_url_without_modules-check
+          - --fail-on-warnings

ci/puppet-validate.yaml

@@ -0,0 +1,5 @@
+repos:
+  - repo: 'https://github.com/chriskuehl/puppet-pre-commit-hooks.git'
+    rev: v2.2.0
+    hooks:
+      - id: puppet-validate

ci/ruby-check.yaml

@@ -0,0 +1,10 @@
+repos:
+  - repo: 'https://github.com/chriskuehl/puppet-pre-commit-hooks.git'
+    rev: v2.2.0
+    hooks:
+      - id: ruby-validate
+  - repo: 'https://github.com/jumanjihouse/pre-commit-hooks'
+    rev: 3.0.0
+    hooks:
+      - id: reek
+      - id: rubocop

ci/ruby-validate.yaml

@@ -0,0 +1,5 @@
+repos:
+  - repo: 'https://github.com/chriskuehl/puppet-pre-commit-hooks.git'
+    rev: v2.2.0
+    hooks:
+      - id: ruby-validate

ci/yamllint.yaml

@@ -0,0 +1,10 @@
+repos:
+  - repo: 'https://github.com/adrienverge/yamllint'
+    rev: v1.32.0
+    hooks:
+      - id: 'yamllint'
+        args:
+          [
+            "-d {extends: relaxed, rules: {line-length: disable}, ignore: chart}",
+            "-s",
+          ]

hiera.yaml

@@ -30,6 +30,7 @@ hierarchy:
       - "roles/%{::enc_role_tier1}.eyaml"
       - "roles/%{::enc_role_tier1}.yaml"
       - "virtual/%{facts.virtual}.yaml"
+      - "os/%{facts.os.name}/%{facts.os.name}%{facts.os.release.major}.%{facts.os.release.minor}.yaml"
       - "os/%{facts.os.name}/%{facts.os.name}%{facts.os.release.major}.yaml"
       - "os/%{facts.os.name}/all_releases.yaml"
       - "common.eyaml"

hieradata/common.yaml

@@ -208,6 +208,20 @@ vault::disable_mlock: false
 profiles::dns::base::nameservers:
   - 198.18.19.16
 profiles::dns::master::basedir: '/var/named/sources'
+
+# dns record publishing. During the k8s cutover both methods run; set
+# manage_export false once k8s is authoritative.
+#   - export:   legacy exported-resources -> puppet DNS master
+#   - nsupdate: RFC2136 to the k8s bind-authoritative write endpoint (.9),
+#               inert until the TSIG key is set in eyaml:
+#                 profiles::dns::updater::key_secret: ENC[...]
+#               (must match the key the bind-authoritative zones allow-update
+#               with; algorithm hmac-sha256)
+profiles::dns::updater::manage_export: true
+profiles::dns::updater::manage_nsupdate: true
+profiles::dns::updater::server: '198.18.200.9'
+profiles::dns::updater::key_name: 'client-update'
+profiles::dns::updater::key_algorithm: 'hmac-sha256'
 #profiles::dns::base::ns_role: 'roles::infra::dns::resolver'
 #profiles::dns::base::use_ns: 'region'
 profiles::consul::server::members_role: roles::infra::storage::consul

hieradata/country/au/region/syd1/infra/halb/haproxy2.yaml

@@ -1,4 +1,7 @@
 ---
+haproxy_server_k8s_syd1_traefik_internal: 'k8s-traefik-internal 198.18.200.4:443 ssl verify none check inter 2s rise 3 fall 2'
+haproxy_server_k8s_syd1_traefik_external: 'k8s-traefik-external 198.18.199.0:443 ssl verify none check inter 2s rise 3 fall 2'
+
 profiles::haproxy::dns::ipaddr: "%{hiera('anycast_ip')}"
 profiles::haproxy::dns::vrrp_cnames:
   - sonarr.main.unkin.net
@@ -16,6 +19,7 @@ profiles::haproxy::dns::vrrp_cnames:
   - mail.main.unkin.net
   - autoconfig.main.unkin.net
   - autodiscover.main.unkin.net
+  - auth.unkin.net
 
 profiles::haproxy::mappings:
   fe_http:
@@ -37,6 +41,7 @@ profiles::haproxy::mappings:
       - 'mail-webadmin.main.unkin.net be_stalwart_webadmin'
       - 'autoconfig.main.unkin.net be_stalwart_webadmin'
       - 'autodiscovery.main.unkin.net be_stalwart_webadmin'
+      - 'auth.unkin.net be_k8s_kanidm'
   fe_https:
     ensure: present
     mappings:
@@ -56,6 +61,7 @@ profiles::haproxy::mappings:
       - 'mail-webadmin.main.unkin.net be_stalwart_webadmin'
       - 'autoconfig.main.unkin.net be_stalwart_webadmin'
       - 'autodiscovery.main.unkin.net be_stalwart_webadmin'
+      - 'auth.unkin.net be_k8s_kanidm'
 
 profiles::haproxy::frontends:
   fe_http:
@@ -80,6 +86,7 @@ profiles::haproxy::frontends:
         - 'acl_stalwart_webadmin req.hdr(host) -i mail-webadmin.main.unkin.net'
         - 'acl_stalwart_webadmin req.hdr(host) -i autoconfig.main.unkin.net'
         - 'acl_stalwart_webadmin req.hdr(host) -i autodiscovery.main.unkin.net'
+        - 'acl_kanidm req.hdr(host) -i auth.unkin.net'
         - 'acl_internalsubnets src 198.18.0.0/16 10.10.12.0/24'
       use_backend:
         - "%[req.hdr(host),lower,map(/etc/haproxy/fe_https.map,be_default)]"
@@ -99,6 +106,7 @@ profiles::haproxy::frontends:
         - 'set-header X-Frame-Options DENY if acl_grafana'
         - 'set-header X-Frame-Options DENY if acl_ceph_dashboard'
         - 'set-header X-Frame-Options DENY if acl_stalwart_webadmin'
+        - 'set-header X-Frame-Options DENY if acl_kanidm'
         - 'set-header X-Content-Type-Options nosniff'
         - 'set-header X-XSS-Protection 1;mode=block'
 
@@ -320,6 +328,26 @@ profiles::haproxy::backends:
         - add-header X-Forwarded-Proto https if { dst_port 9443 }
       redirect: 'scheme https if !{ ssl_fc }'
       stick-table: 'type ip size 200k expire 30m'
+  be_k8s_kanidm:
+    description: Backend for Kanidm (auth.unkin.net via Kubernetes internal Traefik)
+    collect_exported: false
+    options:
+      balance: roundrobin
+      option:
+        - httpchk
+        - forwardfor
+        - http-keep-alive
+        - prefer-last-server
+      http-check:
+        - 'connect ssl sni auth.unkin.net'
+        - 'send meth GET uri /status ver HTTP/1.1 hdr Host auth.unkin.net'
+        - 'expect status 200'
+      http-reuse: always
+      http-request:
+        - set-header X-Forwarded-Port %[dst_port]
+        - add-header X-Forwarded-Proto https if { dst_port 443 }
+      redirect: 'scheme https if !{ ssl_fc }'
+      server: "%{lookup('haproxy_server_k8s_syd1_traefik_internal')} sni str(auth.unkin.net)"
   be_stalwart_imap:
     description: Backend for Stalwart IMAP (STARTTLS)
     collect_exported: false
@@ -393,6 +421,7 @@ profiles::haproxy::certlist::certificates:
   - /etc/pki/tls/letsencrypt/git.unkin.net/fullchain_combined.pem
   - /etc/pki/tls/letsencrypt/grafana.unkin.net/fullchain_combined.pem
   - /etc/pki/tls/letsencrypt/dashboard.ceph.unkin.net/fullchain_combined.pem
+  - /etc/pki/tls/letsencrypt/auth.unkin.net/fullchain_combined.pem
   - /etc/pki/tls/vault/certificate.pem
 
 # additional altnames
@@ -422,3 +451,4 @@ certbot::client::domains:
   - git.unkin.net
   - grafana.unkin.net
   - dashboard.ceph.unkin.net
+  - auth.unkin.net

hieradata/os/AlmaLinux/AlmaLinux8.yaml

@@ -1,7 +1,7 @@
 # hieradata/os/AlmaLinux/AlmaLinux8.yaml
 ---
 crypto_policies::policy: 'DEFAULT'
-
+almalinux-base-repo: almalinux
 profiles::packages::include:
   network-scripts: {}
 

hieradata/os/AlmaLinux/AlmaLinux9.7.yaml

@@ -0,0 +1,2 @@
+---
+almalinux-base-repo: almalinux-vault

hieradata/os/AlmaLinux/AlmaLinux9.yaml

@@ -1,7 +1,7 @@
 # hieradata/os/AlmaLinux/AlmaLinux9.yaml
 ---
 crypto_policies::policy: 'DEFAULT:SHA1'
-
+almalinux-base-repo: almalinux
 profiles::yum::global::repos:
   crb:
     ensure: present

hieradata/os/AlmaLinux/all_releases.yaml

@@ -23,45 +23,45 @@ profiles::yum::global::repos:
     name: baseos
     descr: baseos repository
     target: /etc/yum.repos.d/baseos.repo
-    baseurl: https://artifactapi.k8s.syd1.au.unkin.net/api/v1/remote/almalinux/%{facts.os.release.full}/BaseOS/%{facts.os.architecture}/os/
+    baseurl: https://artifactapi.k8s.syd1.au.unkin.net/api/v1/remote/%{lookup('almalinux-base-repo')}/%{facts.os.release.full}/BaseOS/%{facts.os.architecture}/os/
-    gpgkey: https://artifactapi.k8s.syd1.au.unkin.net/api/v1/remote/almalinux/%{facts.os.release.full}/BaseOS/%{facts.os.architecture}/os/RPM-GPG-KEY-AlmaLinux-%{facts.os.release.major}
+    gpgkey: https://artifactapi.k8s.syd1.au.unkin.net/api/v1/remote/%{lookup('almalinux-base-repo')}/%{facts.os.release.full}/BaseOS/%{facts.os.architecture}/os/RPM-GPG-KEY-AlmaLinux-%{facts.os.release.major}
     mirrorlist: absent
   extras:
     name: extras
     descr: extras repository
     target: /etc/yum.repos.d/extras.repo
-    baseurl: https://artifactapi.k8s.syd1.au.unkin.net/api/v1/remote/almalinux/%{facts.os.release.full}/extras/%{facts.os.architecture}/os/
+    baseurl: https://artifactapi.k8s.syd1.au.unkin.net/api/v1/remote/%{lookup('almalinux-base-repo')}/%{facts.os.release.full}/extras/%{facts.os.architecture}/os/
-    gpgkey: https://artifactapi.k8s.syd1.au.unkin.net/api/v1/remote/almalinux/%{facts.os.release.full}/extras/%{facts.os.architecture}/os/RPM-GPG-KEY-AlmaLinux-%{facts.os.release.major}
+    gpgkey: https://artifactapi.k8s.syd1.au.unkin.net/api/v1/remote/%{lookup('almalinux-base-repo')}/%{facts.os.release.full}/extras/%{facts.os.architecture}/os/RPM-GPG-KEY-AlmaLinux-%{facts.os.release.major}
     mirrorlist: absent
   appstream:
     name: appstream
     descr: appstream repository
     target: /etc/yum.repos.d/appstream.repo
-    baseurl: https://artifactapi.k8s.syd1.au.unkin.net/api/v1/remote/almalinux/%{facts.os.release.full}/AppStream/%{facts.os.architecture}/os/
+    baseurl: https://artifactapi.k8s.syd1.au.unkin.net/api/v1/remote/%{lookup('almalinux-base-repo')}/%{facts.os.release.full}/AppStream/%{facts.os.architecture}/os/
-    gpgkey: https://artifactapi.k8s.syd1.au.unkin.net/api/v1/remote/almalinux/%{facts.os.release.full}/AppStream/%{facts.os.architecture}/os/RPM-GPG-KEY-AlmaLinux-%{facts.os.release.major}
+    gpgkey: https://artifactapi.k8s.syd1.au.unkin.net/api/v1/remote/%{lookup('almalinux-base-repo')}/%{facts.os.release.full}/AppStream/%{facts.os.architecture}/os/RPM-GPG-KEY-AlmaLinux-%{facts.os.release.major}
     mirrorlist: absent
   highavailability:
     name: highavailability
     descr: highavailability repository
     target: /etc/yum.repos.d/highavailability.repo
-    baseurl: https://artifactapi.k8s.syd1.au.unkin.net/api/v1/remote/almalinux/%{facts.os.release.full}/HighAvailability/%{facts.os.architecture}/os/
+    baseurl: https://artifactapi.k8s.syd1.au.unkin.net/api/v1/remote/%{lookup('almalinux-base-repo')}/%{facts.os.release.full}/HighAvailability/%{facts.os.architecture}/os/
-    gpgkey: https://artifactapi.k8s.syd1.au.unkin.net/api/v1/remote/almalinux/%{facts.os.release.full}/HighAvailability/%{facts.os.architecture}/os/RPM-GPG-KEY-AlmaLinux-%{facts.os.release.major}
+    gpgkey: https://artifactapi.k8s.syd1.au.unkin.net/api/v1/remote/%{lookup('almalinux-base-repo')}/%{facts.os.release.full}/HighAvailability/%{facts.os.architecture}/os/RPM-GPG-KEY-AlmaLinux-%{facts.os.release.major}
     mirrorlist: absent
   crb:
     ensure: absent
     name: crb
     descr: crb repository
     target: /etc/yum.repos.d/crb.repo
-    baseurl: https://artifactapi.k8s.syd1.au.unkin.net/api/v1/remote/almalinux/%{facts.os.release.full}/CRB/%{facts.os.architecture}/os/
+    baseurl: https://artifactapi.k8s.syd1.au.unkin.net/api/v1/remote/%{lookup('almalinux-base-repo')}/%{facts.os.release.full}/CRB/%{facts.os.architecture}/os/
-    gpgkey: https://artifactapi.k8s.syd1.au.unkin.net/api/v1/remote/almalinux/%{facts.os.release.full}/CRB/%{facts.os.architecture}/os/RPM-GPG-KEY-AlmaLinux-%{facts.os.release.major}
+    gpgkey: https://artifactapi.k8s.syd1.au.unkin.net/api/v1/remote/%{lookup('almalinux-base-repo')}/%{facts.os.release.full}/CRB/%{facts.os.architecture}/os/RPM-GPG-KEY-AlmaLinux-%{facts.os.release.major}
     mirrorlist: absent
   powertools:
     ensure: absent
     name: powertools
     descr: powertools repository
     target: /etc/yum.repos.d/powertools.repo
-    baseurl: https://artifactapi.k8s.syd1.au.unkin.net/api/v1/remote/almalinux/%{facts.os.release.full}/PowerTools/%{facts.os.architecture}/os/
+    baseurl: https://artifactapi.k8s.syd1.au.unkin.net/api/v1/remote/%{lookup('almalinux-base-repo')}/%{facts.os.release.full}/PowerTools/%{facts.os.architecture}/os/
-    gpgkey: https://artifactapi.k8s.syd1.au.unkin.net/api/v1/remote/almalinux/%{facts.os.release.full}/PowerTools/%{facts.os.architecture}/os/RPM-GPG-KEY-AlmaLinux-%{facts.os.release.major}
+    gpgkey: https://artifactapi.k8s.syd1.au.unkin.net/api/v1/remote/%{lookup('almalinux-base-repo')}/%{facts.os.release.full}/PowerTools/%{facts.os.architecture}/os/RPM-GPG-KEY-AlmaLinux-%{facts.os.release.major}
     mirrorlist: absent
   epel:
     name: epel

hieradata/roles/infra/dns/resolver.yaml

@@ -6,8 +6,10 @@ hiera_include:
 profiles::dns::resolver::acls:
   acl-main.unkin.net:
     addresses:
-      - 10.10.8.1/32
+      - 198.18.1.10/32
+      - 198.18.2.160/27
       - 198.18.21.160/27
+      - 198.18.2.192/27
       - 198.18.21.192/27
       - 198.18.13.0/24
       - 198.18.14.0/24

hieradata/roles/infra/git/server.yaml

@@ -82,8 +82,14 @@ profiles::sql::postgresdb::dbname: gitea
 profiles::sql::postgresdb::dbuser: gitea
 
 # deploy gitea
-gitea::ensure: '1.22.4'
+gitea::base_url: 'https://artifactapi.k8s.syd1.au.unkin.net/api/v1/remote/gitea-dl/gitea'
-gitea::checksum: 'd549104f55067e6fb156e7ba060c9af488f36e12d5e747db7563fcc99eaf8532'
+gitea::install::checksums:
+  1.26.2:
+    linux:
+      amd64: 5b37937b625de69196748f7293eee1a7363f8637ae6e3da3c359bb380bd61a6a
+
+gitea::ensure: '1.26.2'
+gitea::checksum: '5b37937b625de69196748f7293eee1a7363f8637ae6e3da3c359bb380bd61a6a'
 gitea::manage_user: false
 gitea::manage_group: false
 gitea::manage_home: false

hieradata/roles/infra/incus/node.yaml

@@ -5,6 +5,10 @@ hiera_include:
   - incus
   - zfs
   - profiles::ceph::node
+  - profiles::ceph::mon
+  - profiles::ceph::mgr
+  - profiles::ceph::mds
+  - profiles::ceph::osd
   - profiles::ceph::client
   - profiles::ceph::dashboard
   - profiles::storage::cephfsvols
@@ -99,7 +103,7 @@ profiles::yum::global::repos:
 profiles::dns::base::primary_interface: loopback0
 
 # dashboard/haproxy
-profiles::ceph::dashboard::ipaddress: "%{hiera('networking_loopback0_ip')}"
+profiles::ceph::dashboard::ipaddress: "%{hiera('networking_loopback2_ip')}"
 
 # networking
 systemd::manage_networkd: true

hieradata/roles/infra/k8s.yaml

@@ -2,6 +2,7 @@
 hiera_include:
   - profiles::selinux::setenforce
   - profiles::ceph::node
+  - profiles::ceph::osd
   - profiles::ceph::client
   - exporters::frr_exporter
   - frrouting
@@ -10,6 +11,62 @@ hiera_include:
 # manage rke2
 rke2::bootstrap_node: prodnxsr0001.main.unkin.net
 rke2::join_url: https://join-k8s.service.consul:9345
+rke2::manage_registries: true
+rke2::registries:
+  docker.io:
+    endpoint:
+      - "https://artifactapi.k8s.syd1.au.unkin.net"
+    rewrite:
+      "^(.*)$": "dockerhub/$1"
+    disable-default-registry-endpoint: true
+  ghcr.io:
+    endpoint:
+      - "https://artifactapi.k8s.syd1.au.unkin.net"
+    rewrite:
+      "^(.*)$": "ghcr/$1"
+    disable-default-registry-endpoint: true
+  quay.io:
+    endpoint:
+      - "https://artifactapi.k8s.syd1.au.unkin.net"
+    rewrite:
+      "^(.*)$": "quay/$1"
+    disable-default-registry-endpoint: true
+  registry.k8s.io:
+    endpoint:
+      - "https://artifactapi.k8s.syd1.au.unkin.net"
+    rewrite:
+      "^(.*)$": "k8s-registry/$1"
+    disable-default-registry-endpoint: true
+  registry.gitlab.com:
+    endpoint:
+      - "https://artifactapi.k8s.syd1.au.unkin.net"
+    rewrite:
+      "^(.*)$": "gitlab/$1"
+    disable-default-registry-endpoint: true
+  docker.elastic.co:
+    endpoint:
+      - "https://artifactapi.k8s.syd1.au.unkin.net"
+    rewrite:
+      "^(.*)$": "elastic/$1"
+    disable-default-registry-endpoint: true
+  gcr.io:
+    endpoint:
+      - "https://artifactapi.k8s.syd1.au.unkin.net"
+    rewrite:
+      "^(.*)$": "gcr/$1"
+    disable-default-registry-endpoint: true
+  docker.litellm.ai:
+    endpoint:
+      - "https://artifactapi.k8s.syd1.au.unkin.net"
+    rewrite:
+      "^(.*)$": "litellm/$1"
+    disable-default-registry-endpoint: true
+  public.ecr.aws:
+    endpoint:
+      - "https://artifactapi.k8s.syd1.au.unkin.net"
+    rewrite:
+      "^(.*)$": "ecr-public/$1"
+    disable-default-registry-endpoint: true
 rke2::config_hash:
   bind-address: "%{hiera('networking_loopback0_ip')}"
   node-ip: "%{hiera('networking_loopback0_ip')}"
@@ -125,6 +182,17 @@ 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
 
+# sysctl recommendations
+sysctl::base::values:
+  net.ipv4.conf.default.rp_filter:
+    value: '0'
+  net.ipv4.conf.all.rp_filter:
+    value: '0'
+  fs.inotify.max_user_watches:
+    value: '524288'
+  fs.inotify.max_user_instances:
+    value: '512'
+
 # add loopback interfaces to ssh list
 ssh::server::options:
   ListenAddress:

hieradata/roles/infra/metrics/grafana.yaml

@@ -11,6 +11,7 @@ profiles::metrics::grafana::db_name: "%{hiera('profiles::sql::postgresdb::dbname
 profiles::metrics::grafana::db_user: "%{hiera('profiles::sql::postgresdb::dbuser')}"
 profiles::metrics::grafana::db_pass: "%{hiera('profiles::sql::postgresdb::dbpass')}"
 profiles::metrics::grafana::pgsql_backend: true
+profiles::metrics::grafana::version: '13.0.2'
 profiles::metrics::grafana::plugins:
   victoriametrics-logs-datasource:
     ensure: present

hieradata/roles/infra/pki/certbot.yaml

@@ -16,3 +16,4 @@ certbot::domains:
   - git.unkin.net
   - grafana.unkin.net
   - dashboard.ceph.unkin.net
+  - auth.unkin.net

hieradata/roles/infra/puppet/master.yaml

@@ -26,7 +26,7 @@ profiles::puppet::cobbler_enc::packages:
   - 'requests'
   - 'PyYAML'
 profiles::puppet::enc::repo: https://git.service.au-syd1.consul/unkinben/puppet-enc.git
-profiles::puppet::r10k::r10k_repo: https://git.service.au-syd1.consul/unkin/puppet-r10k.git
+profiles::puppet::r10k::r10k_repo: https://git.unkin.net/unkin/puppet-r10k.git
 profiles::puppet::g10k::bin_path: '/usr/bin/g10k'
 profiles::puppet::g10k::cfg_path: '/etc/puppetlabs/r10k/r10k.yaml'
 profiles::puppet::g10k::environments_path: '/etc/puppetlabs/code/environments'

modules/externaldns/manifests/master.pp

@@ -28,8 +28,8 @@ class externaldns::master inherits externaldns {
       dynamic         => true,
       allow_updates   => ['key externaldns-key'],
       allow_transfers => empty($slave_ips) ? {
-        true  => [],
+        true  => ['key externaldns-key'],
-        false => ['dns-slaves'],
+        false => ['key externaldns-key','dns-slaves'],
       },
       ns_notify       => !empty($slave_ips),
       also_notify     => $slave_ips,
@@ -42,4 +42,4 @@ class externaldns::master inherits externaldns {
     recursion => false,
     zones     => $externaldns::k8s_zones,
   }
-}
+}

modules/incus/manifests/cluster.pp

@@ -22,7 +22,12 @@ class incus::cluster (
     }
 
     # if it is, find hosts, sort them so they dont cause changes every run
-    $servers_array = sort(query_nodes("enc_role='${members_role}' and region='${facts['region']}'", 'networking.fqdn'))
+    $servers_array = sort(puppetdb_query(
+      "facts[certname] {
+        name = 'enc_role' and value = '${members_role}' and
+        certname in facts[certname] { name = 'region' and value = '${facts['region']}' }
+      }"
+    ).map |$fact| { $fact['certname'] })
 
   # else use provided array from params
   }else{

modules/libs/lib/facter/ceph_services.rb

@@ -0,0 +1,56 @@
+# frozen_string_literal: true
+
+require 'facter'
+
+# Detects active ceph service instances via systemctl and exposes facts
+# for use in ceph service management profiles.
+# rubocop:disable Style/ClassAndModuleChildren
+module Unkin
+  module Ceph
+    # Detects active ceph service instances via systemctl and exposes Facter facts.
+    module Utils
+      TYPES = %w[mon mgr mds osd].freeze
+
+      def self.services
+        output = Facter::Core::Execution.execute(
+          'systemctl list-units "ceph*" --no-legend --plain --all 2>/dev/null',
+          on_fail: ''
+        )
+        parse_units(output)
+      end
+
+      def self.parse_units(output)
+        result = TYPES.each_with_object({}) { |type, hash| hash[type] = [] }
+        output.each_line do |line|
+          unit = line.split.first
+          next unless unit
+
+          match_unit(result, unit)
+        end
+        result
+      end
+
+      def self.match_unit(result, unit)
+        TYPES.each do |type|
+          match = unit.match(/\Aceph-#{type}@(.+)\.service\z/)
+          result[type] << "ceph-#{type}@#{match[1]}" if match
+        end
+      end
+
+      TYPES.each do |type|
+        define_singleton_method(:"#{type}?") { !services[type].empty? }
+      end
+    end
+  end
+end
+# rubocop:enable Style/ClassAndModuleChildren
+
+Facter.add('ceph_services') do
+  setcode { Unkin::Ceph::Utils.services }
+end
+
+Unkin::Ceph::Utils::TYPES.each do |type|
+  Facter.add("is_ceph_#{type}") do
+    setcode { Unkin::Ceph::Utils.public_send(:"#{type}?") }
+  end
+end

modules/libs/lib/facter/dns_records.rb

@@ -0,0 +1,88 @@
+# frozen_string_literal: true
+
+# lib/facter/dns_records.rb
+#
+# Reports this host's expected DNS records (assembled by profiles::dns::updater
+# into its records file) versus what is currently deployed on the authoritative
+# server, so puppet can detect drift and re-apply.
+#
+# Structured value:
+#   { server, count, expected => [{zone,fqdn,type,ttl,value}], in_sync,
+#     drift => [{...,deployed => [...]}] }
+
+# Helpers for the dns_records fact.
+module DnsRecordsFact
+  RECORDS_FILE = '/var/lib/dns-updater/records'
+  SERVER_FILE  = '/var/lib/dns-updater/server'
+
+  module_function
+
+  # normalise a value for comparison: strip, drop trailing dot, downcase
+  def norm(value)
+    value.to_s.strip.chomp('.').downcase
+  end
+
+  def server
+    File.exist?(SERVER_FILE) ? File.read(SERVER_FILE).strip : nil
+  end
+
+  # a name relative to a zone (or @) as a fully-qualified name
+  def to_fqdn(name, zone)
+    return "#{zone}." if name.to_s.empty? || name == '@'
+
+    "#{name}.#{zone}."
+  end
+
+  # parse one "zone|name|type|ttl|value" line into a record hash (nil to skip)
+  def parse_line(line)
+    line = line.strip
+    return nil if line.empty? || line.start_with?('#')
+
+    zone, name, type, ttl, value = line.split('|', 5)
+    return nil unless zone && type && value
+
+    { 'zone' => zone, 'fqdn' => to_fqdn(name, zone), 'type' => type, 'ttl' => ttl, 'value' => value }
+  end
+
+  # parse the records file into record hashes
+  def expected
+    return [] unless File.exist?(RECORDS_FILE)
+
+    File.readlines(RECORDS_FILE).filter_map { |line| parse_line(line) }
+  end
+
+  # the values currently deployed for a record, per the authoritative server
+  def deployed(record, srv)
+    cmd = ['dig', '+short', '+time=2', '+tries=1']
+    cmd << "@#{srv}" if srv && !srv.empty?
+    cmd += [record['fqdn'], record['type']]
+    out = Facter::Core::Execution.execute(cmd.join(' '), on_fail: '')
+    out.to_s.split("\n").map { |line| norm(line) }.reject(&:empty?)
+  end
+
+  def report
+    srv = server
+    exp = expected
+    drift = exp.filter_map do |record|
+      dep = deployed(record, srv)
+      record.merge('deployed' => dep) unless dep.include?(norm(record['value']))
+    end
+    { 'server' => srv, 'count' => exp.length, 'expected' => exp, 'in_sync' => drift.empty?, 'drift' => drift }
+  end
+end
+
+Facter.add(:dns_records) do
+  confine kernel: 'Linux'
+  setcode do
+    File.exist?(DnsRecordsFact::RECORDS_FILE) ? DnsRecordsFact.report : nil
+  end
+end
+
+# Convenience boolean for `if $facts['dns_records_insync']` guards.
+Facter.add(:dns_records_insync) do
+  confine kernel: 'Linux'
+  setcode do
+    v = Facter.value(:dns_records)
+    v.nil? ? nil : v['in_sync']
+  end
+end

modules/redisha/manifests/redis.pp

@@ -20,7 +20,12 @@ class redisha::redis (
     }
 
     # if it is, find hosts, sort them so they dont cause changes every run
-    $servers_array = sort(query_nodes("enc_role='${redisha_members_role}' and region='${facts['region']}'", 'networking.fqdn'))
+    $servers_array = sort(puppetdb_query(
+      "facts[certname] {
+        name = 'enc_role' and value = '${redisha_members_role}' and
+        certname in facts[certname] { name = 'region' and value = '${facts['region']}' }
+      }"
+    ).map |$fact| { $fact['certname'] })
 
   # else use provided array from params
   }else{

modules/redisha/manifests/sentinel.pp

@@ -23,7 +23,12 @@ class redisha::sentinel (
     }
 
     # if it is, find hosts, sort them so they dont cause changes every run
-    $servers_array = sort(query_nodes("enc_role='${redisha_members_role}' and region='${facts['region']}'", 'networking.fqdn'))
+    $servers_array = sort(puppetdb_query(
+      "facts[certname] {
+        name = 'enc_role' and value = '${redisha_members_role}' and
+        certname in facts[certname] { name = 'region' and value = '${facts['region']}' }
+      }"
+    ).map |$fact| { $fact['certname'] })
 
   # else use provided array from params
   }else{

modules/rke2/manifests/config.pp

@@ -7,6 +7,8 @@ class rke2::config (
   Stdlib::Fqdn            $bootstrap_node = $rke2::bootstrap_node,
   String                  $node_token = $rke2::node_token,
   Array[String[1]]        $extra_config_files = $rke2::extra_config_files,
+  Boolean                 $manage_registries = $rke2::manage_registries,
+  Hash                    $registries = $rke2::registries,
 ){
 
   # if its not the bootstrap node, add join path to config
@@ -28,6 +30,24 @@ class rke2::config (
     $config = $config_hash
   }
 
+  if $manage_registries {
+    file { '/etc/rancher/rke2/registries.yaml':
+      ensure  => file,
+      owner   => 'root',
+      group   => 'root',
+      mode    => '0644',
+      content => epp('rke2/registries.yaml.epp', { registries => $registries }),
+      require => Package["rke2-${node_type}"],
+      notify  => Service["rke2-${node_type}"],
+    }
+  }else{
+    file { '/etc/rancher/rke2/registries.yaml':
+      ensure  => absent,
+      require => Package["rke2-${node_type}"],
+      notify  => Service["rke2-${node_type}"],
+    }
+  }
+
   # create the config file
   file { $config_file:
     ensure  => file,

modules/rke2/manifests/init.pp

@@ -12,6 +12,8 @@ class rke2 (
   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,
+  Boolean                 $manage_registries = $rke2::params::manage_registries,
+  Hash                    $registries = $rke2::params::registries,
 ) inherits rke2::params {
 
   include rke2::install

modules/rke2/manifests/params.pp

@@ -12,4 +12,6 @@ class rke2::params (
   Hash                    $helm_repos = {},
   Array[String[1]]        $extra_config_files = [],
   Stdlib::HTTPUrl         $container_archive_source = 'https://github.com/rancher/rke2/releases/download',
+  Boolean                 $manage_registries = false,
+  Hash                    $registries = {},
 ) {}

modules/rke2/templates/registries.yaml.epp

@@ -0,0 +1,20 @@
+<%- | Hash $registries | -%>
+---
+# DO NOT MODIFY - MANAGED BY PUPPET
+mirrors:
+<%- $registries.each |$registry, $config| { -%>
+  <%= $registry %>:
+    endpoint:
+<%- $config['endpoint'].each |$ep| { -%>
+      - "<%= $ep %>"
+<%- } -%>
+<%- if $config['rewrite'] { -%>
+    rewrite:
+<%- $config['rewrite'].each |$pattern, $replacement| { -%>
+      "<%= $pattern %>": "<%= $replacement %>"
+<%- } -%>
+<%- } -%>
+<%- if $config['disable-default-registry-endpoint'] { -%>
+    disable-default-registry-endpoint: true
+<%- } -%>
+<%- } -%>

modules/stalwart/manifests/init.pp

@@ -167,7 +167,13 @@ class stalwart (
 
   # Query cluster members for validation
   $cluster_query = "enc_role='${cluster_role}' and country='${facts['country']}' and region='${facts['region']}'"
-  $cluster_members_raw = query_nodes($cluster_query, 'networking.fqdn')
+  $cluster_members_raw = puppetdb_query(
+    "facts[certname] {
+      name = 'enc_role' and value = '${cluster_role}' and
+      certname in facts[certname] { name = 'country' and value = '${facts['country']}' } and
+      certname in facts[certname] { name = 'region' and value = '${facts['region']}' }
+    }"
+  ).map |$fact| { $fact['certname'] }
   $cluster_members = $cluster_members_raw ? {
     undef   => [],
     default => $cluster_members_raw,
@@ -180,7 +186,20 @@ class stalwart (
 
   # Query HAProxy nodes for proxy trusted networks
   $haproxy_query = "enc_role='${haproxy_role}' and country='${facts['country']}' and region='${facts['region']}'"
-  $haproxy_members_raw = query_nodes($haproxy_query, 'networking.ip')
+  $haproxy_members_raw = puppetdb_query(
+    "facts[certname,value] {
+      name = 'networking' and
+      certname in facts[certname] {
+        name = 'enc_role' and value = '${haproxy_role}'
+      } and
+      certname in facts[certname] {
+        name = 'country' and value = '${facts['country']}'
+      } and
+      certname in facts[certname] {
+        name = 'region' and value = '${facts['region']}'
+      }
+    }"
+  ).map |$fact| { $fact['value']['ip'] }
   $haproxy_ips = $haproxy_members_raw ? {
     undef   => [],
     default => sort($haproxy_members_raw),

site/profiles/manifests/ceph/mds.pp

@@ -0,0 +1,13 @@
+class profiles::ceph::mds (
+  Boolean $ensure_running = true,
+) {
+
+  if $ensure_running and $facts['is_ceph_mds'] {
+    $facts['ceph_services']['mds'].each |String $svc| {
+      service { $svc:
+        ensure => running,
+        enable => true,
+      }
+    }
+  }
+}

site/profiles/manifests/ceph/mgr.pp

@@ -0,0 +1,13 @@
+class profiles::ceph::mgr (
+  Boolean $ensure_running = true,
+) {
+
+  if $ensure_running and $facts['is_ceph_mgr'] {
+    $facts['ceph_services']['mgr'].each |String $svc| {
+      service { $svc:
+        ensure => running,
+        enable => true,
+      }
+    }
+  }
+}

site/profiles/manifests/ceph/mon.pp

@@ -0,0 +1,13 @@
+class profiles::ceph::mon (
+  Boolean $ensure_running = true,
+) {
+
+  if $ensure_running and $facts['is_ceph_mon'] {
+    $facts['ceph_services']['mon'].each |String $svc| {
+      service { $svc:
+        ensure => running,
+        enable => true,
+      }
+    }
+  }
+}

site/profiles/manifests/ceph/osd.pp

@@ -0,0 +1,13 @@
+class profiles::ceph::osd (
+  Boolean $ensure_running = true,
+) {
+
+  if $ensure_running and $facts['is_ceph_osd'] {
+    $facts['ceph_services']['osd'].each |String $svc| {
+      service { $svc:
+        ensure => running,
+        enable => true,
+      }
+    }
+  }
+}

site/profiles/manifests/consul/client.pp

@@ -28,7 +28,12 @@ class profiles::consul::client (
       }
 
       # if it is, find hosts, sort them so they dont cause changes every run
-      $servers_array = sort(query_nodes("enc_role='${members_role}' and region='${::facts['region']}'", 'networking.fqdn'))
+      $servers_array = sort(puppetdb_query(
+        "facts[certname] {
+          name = 'enc_role' and value = '${members_role}' and
+          certname in facts[certname] { name = 'region' and value = '${::facts['region']}' }
+        }"
+      ).map |$fact| { $fact['certname'] })
 
     # else use provided array from params
     }else{

site/profiles/manifests/consul/server.pp

@@ -65,12 +65,22 @@ class profiles::consul::server (
       }
 
       # if it is, find hosts, sort them so they dont cause changes every run
-      $servers_array = sort(query_nodes("enc_role='${members_role}' and region='${::facts['region']}'", 'networking.fqdn'))
+      $servers_array = sort(puppetdb_query(
+        "facts[certname] {
+          name = 'enc_role' and value = '${members_role}' and
+          certname in facts[certname] { name = 'region' and value = '${::facts['region']}' }
+        }"
+      ).map |$fact| { $fact['certname'] })
 
       if $join_remote_regions {
         # get all nodes in the members_role for each other region
         $region_to_servers = $remote_regions.reduce({}) |$memo, $region| {
-          $servers = sort(query_nodes("enc_role='${members_role}' and region='${region}'", 'networking.fqdn'))
+          $servers = sort(puppetdb_query(
+            "facts[certname] {
+              name = 'enc_role' and value = '${members_role}' and
+              certname in facts[certname] { name = 'region' and value = '${region}' }
+            }"
+          ).map |$fact| { $fact['certname'] })
           $memo + { $region => $servers }
         }
 

site/profiles/manifests/dns/base.pp

@@ -11,16 +11,38 @@ class profiles::dns::base (
   Optional[String] $ns_role = undef,
 ){
 
-  # install bind_utils
+  # install bind_utils (provides nsupdate)
   include bind::updater
 
+  # assemble the host's DNS records and nsupdate them to the authoritative server
+  include profiles::dns::updater
+
   # if ns_role is set, find all hosts matching that enc_role
   $nameserver_array = $ns_role ? {
     undef   => $nameservers,
     default => $use_ns ? {
-      'all'     => query_nodes("enc_role='${ns_role}'", 'networking.ip'),
+      'all'     => puppetdb_query(
-      'region'  => query_nodes("enc_role='${ns_role}' and region=${facts['region']}", 'networking.ip'),
+        "facts[certname,value] {
-      'country' => query_nodes("enc_role='${ns_role}' and country=${facts['country']}", 'networking.ip'),
+          name = 'networking' and
+          certname in nodes[certname] { facts.enc_role = '${ns_role}' }
+        }"
+      ).map |$fact| { $fact['value']['ip'] },
+      'region'  => puppetdb_query(
+        "facts[certname,value] {
+          name = 'networking' and
+          certname in nodes[certname] {
+            facts.enc_role = '${ns_role}' and facts.region = '${facts['region']}'
+          }
+        }"
+      ).map |$fact| { $fact['value']['ip'] },
+      'country' => puppetdb_query(
+        "facts[certname,value] {
+          name = 'networking' and
+          certname in nodes[certname] {
+            facts.enc_role = '${ns_role}' and facts.country = '${facts['country']}'
+          }
+        }"
+      ).map |$fact| { $fact['value']['ip'] },
     }
   }
 

site/profiles/manifests/dns/master.pp

@@ -20,9 +20,21 @@ class profiles::dns::master (
   $nameservers_array = $ns_role ? {
     undef   => [$facts['networking']['fqdn']],
     default => $use_ns ? {
-      'all'     => sort(query_nodes("enc_role='${ns_role}'", 'networking.fqdn')),
+      'all'     => sort(puppetdb_query(
-      'region'  => sort(query_nodes("enc_role='${ns_role}' and region=${facts['region']}", 'networking.fqdn')),
+        "facts[certname] { name = 'enc_role' and value = '${ns_role}' }"
-      'country' => sort(query_nodes("enc_role='${ns_role}' and country=${facts['country']}", 'networking.fqdn')),
+      ).map |$fact| { $fact['certname'] }),
+      'region'  => sort(puppetdb_query(
+        "facts[certname] {
+          name = 'enc_role' and value = '${ns_role}' and
+          certname in facts[certname] { name = 'region' and value = '${facts['region']}' }
+        }"
+      ).map |$fact| { $fact['certname'] }),
+      'country' => sort(puppetdb_query(
+        "facts[certname] {
+          name = 'enc_role' and value = '${ns_role}' and
+          certname in facts[certname] { name = 'country' and value = '${facts['country']}' }
+        }"
+      ).map |$fact| { $fact['certname'] }),
     }
   }
 
@@ -32,7 +44,9 @@ class profiles::dns::master (
       $facts['networking']['fqdn'] => $facts['networking']['ip']
     },
     default => $nameservers_array.reduce({}) |$acc, $fqdn| {
-      $result = query_nodes("networking.fqdn='${fqdn}'", 'networking.ip')
+      $result = puppetdb_query(
+        "facts[certname,value] { name = 'networking' and certname = '${fqdn}' }"
+      ).map |$fact| { $fact['value']['ip'] }
       $ip = $result[0]
       $acc + { "${fqdn}." => $ip }
     }

site/profiles/manifests/dns/record.pp

@@ -1,4 +1,10 @@
-# defines the base record that will be exported
+# profiles::dns::record
+#
+# Declares a DNS record for this host. Publishes it via either or both methods,
+# controlled by profiles::dns::updater's toggles (both on during cutover):
+#   - nsupdate: a local concat fragment consumed by profiles::dns::updater,
+#     which nsupdates it to the authoritative server.
+#   - export: the legacy @@concat::fragment exported to the puppet DNS master.
 define profiles::dns::record (
   String  $record,
   Enum[
@@ -13,11 +19,26 @@ define profiles::dns::record (
   String  $value,
   String  $zone,
   Integer $order,
-  Stdlib::AbsolutePath $basedir = lookup('profiles::dns::master::basedir'),
+  Integer $ttl = 300,
 ) {
-  @@concat::fragment { "${zone}_${name}":
+  include profiles::dns::updater
-    target  => "${basedir}/${zone}.conf",
+
-    content => "${record} IN ${type} ${value}\n",
+  # new: local records file consumed by the nsupdate service
-    order   => $order,
+  if $profiles::dns::updater::manage_nsupdate {
+    # zone|name|type|ttl|value  (parsed by the dns-update script)
+    concat::fragment { "dns-record-${name}":
+      target  => $profiles::dns::updater::records_file,
+      content => "${zone}|${record}|${type}|${ttl}|${value}\n",
+      order   => sprintf('%03d', $order),
+    }
+  }
+
+  # legacy: export the fragment to the puppet DNS master
+  if $profiles::dns::updater::manage_export {
+    @@concat::fragment { "${zone}_${name}":
+      target  => "${profiles::dns::updater::master_basedir}/${zone}.conf",
+      content => "${record} IN ${type} ${value}\n",
+      order   => $order,
+    }
   }
 }

site/profiles/manifests/dns/updater.pp

@@ -0,0 +1,127 @@
+# profiles::dns::updater
+#
+# Publishes this host's DNS records. Two methods, independently toggled so both
+# can run during the k8s cutover (profiles::dns::record honours the same flags):
+#
+#   - nsupdate ($manage_nsupdate): assemble the records into a local file and
+#     nsupdate them to the k8s authoritative write endpoint via a systemd .path
+#     unit that watches the file. Inert until $key_secret (TSIG) is set.
+#   - export ($manage_export): the legacy exported-resources flow to the puppet
+#     DNS master. Kept during cutover; disable once k8s is authoritative.
+#
+# nsupdate comes from bind-utils (installed via bind::updater in
+# profiles::dns::base).
+class profiles::dns::updater (
+  Boolean                     $manage_nsupdate = true,
+  Boolean                     $manage_export   = true,
+  String                      $server          = '198.18.200.9',
+  String                      $key_name        = 'client-update',
+  String                      $key_algorithm   = 'hmac-sha256',
+  Optional[Sensitive[String]] $key_secret      = undef,
+  Integer                     $default_ttl     = 300,
+  Stdlib::AbsolutePath        $records_file    = '/var/lib/dns-updater/records',
+  Stdlib::AbsolutePath        $state_dir       = '/var/lib/dns-updater',
+  Stdlib::AbsolutePath        $config_dir      = '/etc/dns-updater',
+  Stdlib::AbsolutePath        $master_basedir  = lookup('profiles::dns::master::basedir'),
+) {
+
+  $state_file  = "${state_dir}/applied"
+  $server_file = "${state_dir}/server"
+  $key_file    = "${config_dir}/key"
+
+  if $manage_nsupdate {
+
+    file { $state_dir:
+      ensure => directory,
+      owner  => 'root',
+      group  => 'root',
+      mode   => '0755',
+    }
+
+    # Server address, read by the dns_records fact for drift detection.
+    file { $server_file:
+      ensure  => file,
+      owner   => 'root',
+      group   => 'root',
+      mode    => '0644',
+      content => "${server}\n",
+      require => File[$state_dir],
+    }
+
+    # Records file, assembled from profiles::dns::record fragments.
+    concat { $records_file:
+      ensure         => present,
+      owner          => 'root',
+      group          => 'root',
+      mode           => '0644',
+      ensure_newline => true,
+      warn           => false,
+      require        => File[$state_dir],
+    }
+
+    concat::fragment { 'dns-update-header':
+      target  => $records_file,
+      content => "# Managed by puppet (profiles::dns::record): zone|name|type|ttl|value\n",
+      order   => '00',
+    }
+
+    if $key_secret =~ Undef {
+      notify { 'dns-updater-inert':
+        message  => 'profiles::dns::updater: key_secret unset; records assembled but not applied.',
+        loglevel => 'info',
+      }
+    } else {
+      file { $config_dir:
+        ensure => directory,
+        owner  => 'root',
+        group  => 'root',
+        mode   => '0700',
+      }
+
+      file { $key_file:
+        ensure    => file,
+        owner     => 'root',
+        group     => 'root',
+        mode      => '0600',
+        show_diff => false,
+        content   => Sensitive(epp('profiles/dns/tsig-key.epp', {
+            'name'      => $key_name,
+            'algorithm' => $key_algorithm,
+            'secret'    => $key_secret.unwrap,
+        })),
+      }
+
+      file { '/usr/local/bin/dns-update':
+        ensure  => file,
+        owner   => 'root',
+        group   => 'root',
+        mode    => '0755',
+        content => epp('profiles/dns/dns-update.sh.epp', {
+            'server'       => $server,
+            'key_file'     => $key_file,
+            'records_file' => $records_file,
+            'state_file'   => $state_file,
+        }),
+      }
+
+      systemd::unit_file { 'dns-update.service':
+        content => epp('profiles/dns/dns-update.service.epp', { 'script' => '/usr/local/bin/dns-update' }),
+      }
+
+      # The .path unit watches the records file and triggers the service.
+      systemd::unit_file { 'dns-update.path':
+        content => epp('profiles/dns/dns-update.path.epp', { 'records_file' => $records_file }),
+        active  => true,
+        enable  => true,
+      }
+
+      # Also apply within the puppet run whenever the records change.
+      exec { 'dns-update-apply':
+        command     => '/usr/local/bin/dns-update',
+        refreshonly => true,
+        subscribe   => Concat[$records_file],
+        require     => [File['/usr/local/bin/dns-update'], File[$key_file]],
+      }
+    }
+  }
+}

site/profiles/manifests/etcd/node.pp

@@ -18,7 +18,12 @@ class profiles::etcd::node (
     }
 
     # if it is, find hosts, sort them so they dont cause changes every run
-    $servers_array = sort(query_nodes("enc_role='${members_role}' and region='${facts['region']}'", 'networking.fqdn'))
+    $servers_array = sort(puppetdb_query(
+      "facts[certname] {
+        name = 'enc_role' and value = '${members_role}' and
+        certname in facts[certname] { name = 'region' and value = '${facts['region']}' }
+      }"
+    ).map |$fact| { $fact['certname'] })
 
   # else use provided array from params
   }else{
@@ -31,7 +36,7 @@ class profiles::etcd::node (
     $initial_cluster = $servers_array.map |$fqdn| {
 
       # lookup the ip address for the current fqdn
-      $ip = query_nodes("networking.fqdn='${fqdn}'", 'networking.ip')[0]
+      $ip = puppetdb_query("facts[certname,value] { name = 'networking' and certname = '${fqdn}' }").map |$fact| { $fact['value']['ip'] }[0]
 
       # construct the string for this server
       "${fqdn}=https://${ip}:${peer_port}"

site/profiles/manifests/haproxy/dns.pp

@@ -30,13 +30,14 @@ class profiles::haproxy::dns (
   }
 
   # if it is, find hosts, sort them so they dont cause changes every run
-  $servers_array = sort(query_nodes(
+  $servers_array = sort(puppetdb_query(
-    "enc_role='${facts['enc_role']}' and
+    "facts[certname] {
-      country='${facts['country']}' and
+      name = 'enc_role' and value = '${facts['enc_role']}' and
-      region='${facts['region']}' and
+      certname in facts[certname] { name = 'country' and value = '${facts['country']}' } and
-      environment='${facts['environment']}'",
+      certname in facts[certname] { name = 'region' and value = '${facts['region']}' } and
-    'networking.fqdn'
+      certname in facts[certname] { name = 'environment' and value = '${facts['environment']}' }
-  ))
+    }"
+  ).map |$fact| { $fact['certname'] })
 
   # give enough time for a few hosts to be provisioned
   if length($servers_array) >= 3 {

site/profiles/manifests/metrics/grafana.pp

@@ -1,6 +1,7 @@
 # profiles::metrics::grafana
 class profiles::metrics::grafana (
   String    $ldap_bind_pass,
+  String    $version        = 'installed',
   Stdlib::Port $http_port   = 8080,
   String    $app_mode       = 'production',
   Boolean   $allow_sign_up  = false,
@@ -107,6 +108,7 @@ class profiles::metrics::grafana (
 
   # deploy grafana
   class { 'grafana':
+    version  => $version,
     cfg      => $cfg,
     ldap_cfg => $ldap_cfg,
     plugins  => $plugins,

site/profiles/manifests/minio/server.pp

@@ -98,8 +98,15 @@ class profiles::minio::server (
     }
 
     # if it is, find hosts, sort them so they dont cause changes every run
-    #$servers_array = sort(query_nodes("enc_role='${minio_members_role}'", 'networking.fqdn'))
+    #$servers_array = sort(puppetdb_query(
-    $servers_array = sort(query_nodes("enc_role='${minio_members_role}' and minio_region='${minio_region}'", 'networking.fqdn'))
+    #  "facts[certname] { name = 'enc_role' and value = '${minio_members_role}' }"
+    #).map |$fact| { $fact['certname'] })
+    $servers_array = sort(puppetdb_query(
+      "facts[certname] {
+        name = 'enc_role' and value = '${minio_members_role}' and
+        certname in facts[certname] { name = 'minio_region' and value = '${minio_region}' }
+      }"
+    ).map |$fact| { $fact['certname'] })
 
   # else use provided array from params
   }else{

site/profiles/manifests/ntp/client.pp

@@ -26,9 +26,21 @@ class profiles::ntp::client (
     $ntpserver_array = $ntp_role ? {
       undef   => $peers,
       default => $use_ntp ? {
-        'all'     => query_nodes("enc_role='${ntp_role}'", 'networking.fqdn'),
+        'all'     => puppetdb_query(
-        'region'  => query_nodes("enc_role='${ntp_role}' and region=${facts['region']}", 'networking.fqdn'),
+          "facts[certname] { name = 'enc_role' and value = '${ntp_role}' }"
-        'country' => query_nodes("enc_role='${ntp_role}' and country=${facts['country']}", 'networking.fqdn'),
+        ).map |$fact| { $fact['certname'] },
+        'region'  => puppetdb_query(
+          "facts[certname] {
+            name = 'enc_role' and value = '${ntp_role}' and
+            certname in facts[certname] { name = 'region' and value = '${facts['region']}' }
+          }"
+        ).map |$fact| { $fact['certname'] },
+        'country' => puppetdb_query(
+          "facts[certname] {
+            name = 'enc_role' and value = '${ntp_role}' and
+            certname in facts[certname] { name = 'country' and value = '${facts['country']}' }
+          }"
+        ).map |$fact| { $fact['certname'] },
       }
     }
 

site/profiles/manifests/proxmox/clusterinit.pp

@@ -24,10 +24,13 @@ class profiles::proxmox::clusterinit {
     }
   }
 
-  $servers_array = sort(query_nodes(
+  $servers_array = sort(puppetdb_query(
-    "enc_role='${membersrole}' and country='${facts['country']}' and region='${facts['region']}'",
+    "facts[certname] {
-    'networking.fqdn'
+      name = 'enc_role' and value = '${membersrole}' and
-  ))
+      certname in facts[certname] { name = 'country' and value = '${facts['country']}' } and
+      certname in facts[certname] { name = 'region' and value = '${facts['region']}' }
+    }"
+  ).map |$fact| { $fact['certname'] })
 
   if ! $profiles::proxmox::params::pve_clusterinit_master {
     if !empty($servers_array) {

site/profiles/manifests/proxmox/clusterjoin.pp

@@ -11,13 +11,14 @@ class profiles::proxmox::clusterjoin {
   $root_password      = $profiles::proxmox::params::root_password
 
   # query puppetdb for list of cluster members
-  $members_array = sort(query_nodes(
+  $members_array = sort(puppetdb_query(
-    "enc_role='${membersrole}' and \
+    "facts[certname] {
-    country='${facts['country']}' and \
+      name = 'enc_role' and value = '${membersrole}' and
-    region='${facts['region']}' and \
+      certname in facts[certname] { name = 'country' and value = '${facts['country']}' } and
-    pve_cluster.cluster_name='${clustername}'",
+      certname in facts[certname] { name = 'region' and value = '${facts['region']}' } and
-    'networking.fqdn'
+      certname in facts[certname] { name = 'pve_cluster' and value.cluster_name = '${clustername}' }
-  ))
+    }"
+  ).map |$fact| { $fact['certname'] })
 
   # check if the pve kernerl is running
   if $facts['kernelrelease'] == $profiles::proxmox::params::pve_kernel_release {

site/profiles/manifests/sql/galera_member.pp

@@ -48,7 +48,12 @@ class profiles::sql::galera_member (
     }
 
     # if it is, find hosts, sort them so they dont cause changes every run
-    $servers_array = sort(query_nodes("enc_role='${galera_members_role}' and region='${facts['region']}'", 'networking.fqdn'))
+    $servers_array = sort(puppetdb_query(
+      "facts[certname] {
+        name = 'enc_role' and value = '${galera_members_role}' and
+        certname in facts[certname] { name = 'region' and value = '${facts['region']}' }
+      }"
+    ).map |$fact| { $fact['certname'] })
 
   # else use provided array from params
   }else{

site/profiles/manifests/sql/postgresdb.pp

@@ -18,7 +18,12 @@ class profiles::sql::postgresdb (
     }
 
     # if it is, find hosts, sort them so they dont cause changes every run
-    $servers_array = sort(query_nodes("enc_role='${members_role}' and region='${facts['region']}'", 'networking.fqdn'))
+    $servers_array = sort(puppetdb_query(
+      "facts[certname] {
+        name = 'enc_role' and value = '${members_role}' and
+        certname in facts[certname] { name = 'region' and value = '${facts['region']}' }
+      }"
+    ).map |$fact| { $fact['certname'] })
 
   # else use provided array from params
   }else{

site/profiles/manifests/vault/server.pp

@@ -29,7 +29,12 @@ class profiles::vault::server (
   if $members_lookup and $members_role != undef {
 
     # if it is, find hosts, sort them so they dont cause changes every run
-    $servers_array = sort(query_nodes("enc_role='${members_role}' and region='${::facts['region']}'", 'networking.fqdn'))
+    $servers_array = sort(puppetdb_query(
+      "facts[certname] {
+        name = 'enc_role' and value = '${members_role}' and
+        certname in facts[certname] { name = 'region' and value = '${::facts['region']}' }
+      }"
+    ).map |$fact| { $fact['certname'] })
 
   # else use provided array from params
   }else{

site/profiles/templates/dns/dns-update.path.epp

@@ -0,0 +1,10 @@
+<%- | String $records_file | -%>
+[Unit]
+Description=Watch the DNS records file and apply changes
+
+[Path]
+PathModified=<%= $records_file %>
+Unit=dns-update.service
+
+[Install]
+WantedBy=multi-user.target

site/profiles/templates/dns/dns-update.service.epp

@@ -0,0 +1,9 @@
+<%- | String $script | -%>
+[Unit]
+Description=Apply host DNS records via nsupdate
+After=network-online.target
+Wants=network-online.target
+
+[Service]
+Type=oneshot
+ExecStart=<%= $script %>

site/profiles/templates/dns/dns-update.sh.epp

@@ -0,0 +1,56 @@
+<%- | String $server, String $key_file, String $records_file, String $state_file | -%>
+#!/bin/bash
+# Managed by puppet (profiles::dns::updater). Applies this host's records to the
+# authoritative DNS server via TSIG nsupdate. Only the delta since the last
+# successful run is sent; removed records are deleted.
+set -euo pipefail
+
+SERVER="<%= $server %>"
+KEYFILE="<%= $key_file %>"
+RECORDS="<%= $records_file %>"
+STATE="<%= $state_file %>"
+
+[ -f "$RECORDS" ] || exit 0
+touch "$STATE"
+
+# Format per line: zone|name|type|ttl|value  (name is relative to zone, or @).
+desired="$(grep -vE '^[[:space:]]*(#|$)' "$RECORDS" | sort -u || true)"
+applied="$(grep -vE '^[[:space:]]*(#|$)' "$STATE" 2>/dev/null | sort -u || true)"
+
+[ "$desired" = "$applied" ] && exit 0
+
+fqdn() { # name zone
+  if [ -z "$1" ] || [ "$1" = "@" ]; then printf '%s.' "$2"; else printf '%s.%s.' "$1" "$2"; fi
+}
+
+msg="$(mktemp)"
+trap 'rm -f "$msg"' EXIT
+printf 'server %s\n' "$SERVER" >> "$msg"
+
+# Process per zone so each UPDATE message targets a single zone.
+zones="$(printf '%s\n%s\n' "$desired" "$applied" | cut -d'|' -f1 | sort -u | grep -v '^$' || true)"
+for zone in $zones; do
+  printf 'zone %s.\n' "$zone" >> "$msg"
+  # Additions/updates: replace the RRset for every desired record in this zone.
+  printf '%s\n' "$desired" | awk -F'|' -v z="$zone" 'NF>=5 && $1==z' | \
+    while IFS='|' read -r z name type ttl value; do
+      f="$(fqdn "$name" "$z")"
+      printf 'update delete %s %s\n' "$f" "$type" >> "$msg"
+      printf 'update add %s %s %s %s\n' "$f" "$ttl" "$type" "$value" >> "$msg"
+    done
+  # Deletions: records present last run but gone now.
+  comm -23 <(printf '%s\n' "$applied") <(printf '%s\n' "$desired") | \
+    awk -F'|' -v z="$zone" 'NF>=5 && $1==z' | \
+    while IFS='|' read -r z name type ttl value; do
+      f="$(fqdn "$name" "$z")"
+      printf 'update delete %s %s %s\n' "$f" "$type" "$value" >> "$msg"
+    done
+  printf 'send\n' >> "$msg"
+done
+
+if nsupdate -k "$KEYFILE" "$msg"; then
+  printf '%s\n' "$desired" > "$STATE"
+else
+  echo "dns-update: nsupdate to ${SERVER} failed" >&2
+  exit 1
+fi

site/profiles/templates/dns/tsig-key.epp

@@ -0,0 +1,5 @@
+<%- | String $name, String $algorithm, String $secret | -%>
+key "<%= $name %>" {
+    algorithm <%= $algorithm %>;
+    secret "<%= $secret %>";
+};