Deploy kea DHCP operator to dhcp-system (#333)

Replaces the isc-dhcpd PXE-boot VM with the kea-operator + an HA kea pair, GitOps-managed. Deploys on a new, unused anycast IP so nothing is cut over yet; the production cutover off the current dhcpd address (198.18.19.18) is a separate later task. - Add `apps/base/dhcp-system`: namespace, kea-operator RBAC + Deployment (v0.1.0), VPA, and the 4 kea.unkin.net CRDs pulled from the operator repo at tag v0.1.0. - Add CRs translating the legacy dhcpd config (source: puppet `roles/infra/dhcp/server.yaml`): KeaCluster `kea` (2 replicas, hot-standby HA, main.unkin.net, 1200/86400 leases, AU ntp pool); five KeaSubnets 198.18.13-17.0/24 with .200-.220 pools, gateways .254 except .17->.1, next-server 198.18.19.19; Legacy/UEFI-64 PXE client classes; KeaAPI. - DHCP-advertised DNS points at the in-cluster bind-resolvers cluster (PureLB 198.18.200.7), not the legacy 198.18.19.15 forwarder. - Pin the DHCP LoadBalancer Service to the free common-pool IP 198.18.200.10 via PureLB. - KeaAPI bearer token is operator-generated (no plain Secret committed). - Commit generated kea.unkin.net JSON schemas for kubeconform; register dhcp-system in the platform ApplicationSet + AppProject. Client-class object names are lowercased (`legacy`/`uefi-64`) to satisfy RFC1123 since the operator renders the kea class name from metadata.name. https://claude.ai/code/session_01JUoARVdmhxKQHyyyp1pxeT --------- Co-authored-by: Ben Vincent <neotheo@gmail.com> Reviewed-on: #333 Co-authored-by: Ben Vincent <ben@unkin.net> Co-committed-by: Ben Vincent <ben@unkin.net>
2026-08-03 00:18:05 +10:00
parent f296d0549a
commit f31552e192
16 changed files with 2406 additions and 0 deletions
@@ -0,0 +1,25 @@
+---
+# Terraform-friendly REST API for KeaSubnet/KeaClientClass CRUD. The bearer
+# token Secret is generated by the operator when absent (no plain Secret is
+# committed here); it can later be pre-seeded from Vault under the same name.
+apiVersion: kea.unkin.net/v1alpha1
+kind: KeaAPI
+metadata:
+  name: kea-api
+  namespace: dhcp-system
+  annotations:
+    argocd.argoproj.io/sync-wave: "1"
+spec:
+  replicas: 1
+  image: git.unkin.net/unkin/kea-api:v0.1.0
+  tokenSecretName: kea-api-token
+  service:
+    type: ClusterIP
+    port: 8080
+  resources:
+    requests:
+      cpu: 100m
+      memory: 64Mi
+    limits:
+      cpu: "1"
+      memory: 256Mi
@@ -0,0 +1,27 @@
+# PXE boot classes matching client architecture (option 93), replacing the
+# legacy dhcpd "Legacy" and "UEFI-64" classes. Object names are lowercased to
+# satisfy RFC1123 (the operator renders the kea class name from metadata.name).
+---
+apiVersion: kea.unkin.net/v1alpha1
+kind: KeaClientClass
+metadata:
+  name: legacy
+  namespace: dhcp-system
+  annotations:
+    argocd.argoproj.io/sync-wave: "1"
+spec:
+  clusterRef: kea
+  archHex: ["0x0000"]
+  bootFileName: /undionly.kpxe
+---
+apiVersion: kea.unkin.net/v1alpha1
+kind: KeaClientClass
+metadata:
+  name: uefi-64
+  namespace: dhcp-system
+  annotations:
+    argocd.argoproj.io/sync-wave: "1"
+spec:
+  clusterRef: kea
+  archHex: ["0x0007", "0x0009"]
+  bootFileName: /ipxe.efi
@@ -0,0 +1,37 @@
+---
+# HA pair fronted by a PureLB anycast Service on a NEW, unused common-pool IP
+# (198.18.200.10). This is intentionally NOT the current isc-dhcpd anycast
+# address (198.18.19.18) -- the production cutover is a separate later task.
+apiVersion: kea.unkin.net/v1alpha1
+kind: KeaCluster
+metadata:
+  name: kea
+  namespace: dhcp-system
+  annotations:
+    argocd.argoproj.io/sync-wave: "1"
+spec:
+  replicas: 2
+  image: git.unkin.net/unkin/kea:v0.1.0
+  domainName: main.unkin.net
+  defaultLeaseTime: 1200
+  maxLeaseTime: 86400
+  ntpServers:
+    - 0.au.pool.ntp.org
+    - 1.au.pool.ntp.org
+    - 2.au.pool.ntp.org
+    - 3.au.pool.ntp.org
+  ha:
+    mode: hot-standby
+  service:
+    type: LoadBalancer
+    ipAddressPool: common
+    loadBalancerIP: 198.18.200.10
+    annotations:
+      purelb.io/addresses: 198.18.200.10
+  resources:
+    requests:
+      cpu: 100m
+      memory: 128Mi
+    limits:
+      cpu: "1"
+      memory: 512Mi
@@ -0,0 +1,90 @@
+# Translation of the legacy ISC dhcpd pools (puppet
+# roles/infra/dhcp/server.yaml): 198.18.13-17.0/24, each a .200-.220 pool,
+# next-server 198.18.19.19. Gateways per the original config:
+# .13/.14/.15/.16 -> .254, .17 -> .1. DNS points at the in-cluster
+# bind-resolvers PureLB IP (198.18.200.7), not the legacy 198.18.19.15.
+---
+apiVersion: kea.unkin.net/v1alpha1
+kind: KeaSubnet
+metadata:
+  name: net-198-18-13
+  namespace: dhcp-system
+  annotations:
+    argocd.argoproj.io/sync-wave: "1"
+spec:
+  clusterRef: kea
+  subnet: 198.18.13.0/24
+  pools:
+    - 198.18.13.200 - 198.18.13.220
+  routers: [198.18.13.254]
+  dnsServers: [198.18.200.7]
+  domainName: main.unkin.net
+  nextServer: 198.18.19.19
+---
+apiVersion: kea.unkin.net/v1alpha1
+kind: KeaSubnet
+metadata:
+  name: net-198-18-14
+  namespace: dhcp-system
+  annotations:
+    argocd.argoproj.io/sync-wave: "1"
+spec:
+  clusterRef: kea
+  subnet: 198.18.14.0/24
+  pools:
+    - 198.18.14.200 - 198.18.14.220
+  routers: [198.18.14.254]
+  dnsServers: [198.18.200.7]
+  domainName: main.unkin.net
+  nextServer: 198.18.19.19
+---
+apiVersion: kea.unkin.net/v1alpha1
+kind: KeaSubnet
+metadata:
+  name: net-198-18-15
+  namespace: dhcp-system
+  annotations:
+    argocd.argoproj.io/sync-wave: "1"
+spec:
+  clusterRef: kea
+  subnet: 198.18.15.0/24
+  pools:
+    - 198.18.15.200 - 198.18.15.220
+  routers: [198.18.15.254]
+  dnsServers: [198.18.200.7]
+  domainName: main.unkin.net
+  nextServer: 198.18.19.19
+---
+apiVersion: kea.unkin.net/v1alpha1
+kind: KeaSubnet
+metadata:
+  name: net-198-18-16
+  namespace: dhcp-system
+  annotations:
+    argocd.argoproj.io/sync-wave: "1"
+spec:
+  clusterRef: kea
+  subnet: 198.18.16.0/24
+  pools:
+    - 198.18.16.200 - 198.18.16.220
+  routers: [198.18.16.254]
+  dnsServers: [198.18.200.7]
+  domainName: main.unkin.net
+  nextServer: 198.18.19.19
+---
+apiVersion: kea.unkin.net/v1alpha1
+kind: KeaSubnet
+metadata:
+  name: net-198-18-17
+  namespace: dhcp-system
+  annotations:
+    argocd.argoproj.io/sync-wave: "1"
+spec:
+  clusterRef: kea
+  subnet: 198.18.17.0/24
+  pools:
+    - 198.18.17.200 - 198.18.17.220
+  routers: [198.18.17.1]
+  dnsServers: [198.18.200.7]
+  domainName: main.unkin.net
+  nextServer: 198.18.19.19
@@ -0,0 +1,56 @@
+---
+apiVersion: apps/v1
+kind: Deployment
+metadata:
+  name: kea-operator
+  namespace: dhcp-system
+  labels:
+    app.kubernetes.io/name: kea-operator
+spec:
+  replicas: 1
+  selector:
+    matchLabels:
+      app.kubernetes.io/name: kea-operator
+  template:
+    metadata:
+      labels:
+        app.kubernetes.io/name: kea-operator
+    spec:
+      serviceAccountName: kea-operator
+      securityContext:
+        runAsNonRoot: true
+      containers:
+        - name: operator
+          image: git.unkin.net/unkin/kea-operator:v0.1.0
+          args:
+            - --metrics-bind-address=:8080
+            - --health-probe-bind-address=:8081
+          ports:
+            - containerPort: 8080
+              name: metrics
+            - containerPort: 8081
+              name: health
+          readinessProbe:
+            httpGet:
+              path: /readyz
+              port: 8081
+            initialDelaySeconds: 5
+            periodSeconds: 10
+          livenessProbe:
+            httpGet:
+              path: /healthz
+              port: 8081
+            initialDelaySeconds: 15
+            periodSeconds: 20
+          securityContext:
+            allowPrivilegeEscalation: false
+            readOnlyRootFilesystem: true
+            capabilities:
+              drop: ["ALL"]
+          resources:
+            requests:
+              cpu: 50m
+              memory: 64Mi
+            limits:
+              cpu: 500m
+              memory: 256Mi
@@ -0,0 +1,17 @@
+---
+apiVersion: kustomize.config.k8s.io/v1beta1
+kind: Kustomization
+
+resources:
+  - namespace.yaml
+  # CRDs are pulled from the kea-operator repo at the matching tag rather than
+  # vendored here, so they never drift from the operator.
+  - https://git.unkin.net/unkin/kea-operator/raw/tag/v0.1.0/config/crd/install.yaml
+  - rbac.yaml
+  - deployment.yaml
+  - vpa.yaml
+  # CRs (sync-wave 1) reconcile after the operator + CRDs are established.
+  - cr/keacluster.yaml
+  - cr/keasubnets.yaml
+  - cr/keaclientclasses.yaml
+  - cr/keaapi.yaml
@@ -0,0 +1,5 @@
+---
+apiVersion: v1
+kind: Namespace
+metadata:
+  name: dhcp-system
@@ -0,0 +1,46 @@
+---
+apiVersion: v1
+kind: ServiceAccount
+metadata:
+  name: kea-operator
+  namespace: dhcp-system
+---
+# Sourced from the kea-operator repo config/rbac/role.yaml (v0.1.0). Leader
+# election is disabled so no coordination.k8s.io/leases grant is needed.
+apiVersion: rbac.authorization.k8s.io/v1
+kind: ClusterRole
+metadata:
+  name: kea-operator
+rules:
+  - apiGroups: [""]
+    resources: ["configmaps", "secrets", "serviceaccounts", "services"]
+    verbs: ["create", "delete", "get", "list", "patch", "update", "watch"]
+  - apiGroups: [""]
+    resources: ["pods"]
+    verbs: ["get", "list", "watch"]
+  - apiGroups: ["apps"]
+    resources: ["deployments", "statefulsets"]
+    verbs: ["create", "delete", "get", "list", "patch", "update", "watch"]
+  - apiGroups: ["kea.unkin.net"]
+    resources: ["keaapis", "keaclientclasses", "keaclusters", "keasubnets"]
+    verbs: ["create", "delete", "get", "list", "patch", "update", "watch"]
+  - apiGroups: ["kea.unkin.net"]
+    resources:
+      ["keaapis/status", "keaclientclasses/status", "keaclusters/status", "keasubnets/status"]
+    verbs: ["get", "patch", "update"]
+  - apiGroups: ["rbac.authorization.k8s.io"]
+    resources: ["rolebindings", "roles"]
+    verbs: ["create", "delete", "get", "list", "patch", "update", "watch"]
+---
+apiVersion: rbac.authorization.k8s.io/v1
+kind: ClusterRoleBinding
+metadata:
+  name: kea-operator
+subjects:
+  - kind: ServiceAccount
+    name: kea-operator
+    namespace: dhcp-system
+roleRef:
+  apiGroup: rbac.authorization.k8s.io
+  kind: ClusterRole
+  name: kea-operator
@@ -0,0 +1,13 @@
+---
+apiVersion: autoscaling.k8s.io/v1
+kind: VerticalPodAutoscaler
+metadata:
+  name: kea-operator-vpa
+  namespace: dhcp-system
+spec:
+  targetRef:
+    apiVersion: apps/v1
+    kind: Deployment
+    name: kea-operator
+  updatePolicy:
+    updateMode: "Off"
@@ -0,0 +1,6 @@
+---
+apiVersion: kustomize.config.k8s.io/v1beta1
+kind: Kustomization
+
+resources:
+  - ../../../base/dhcp-system