diff --git a/packages/core/platform/templates/bundles/system.yaml b/packages/core/platform/templates/bundles/system.yaml index 43ea7266..6c587c82 100644 --- a/packages/core/platform/templates/bundles/system.yaml +++ b/packages/core/platform/templates/bundles/system.yaml @@ -105,9 +105,6 @@ {{- /* cozystack-api DaemonSet */ -}} {{- $apiValues := dict "cozystackAPI" (dict "nodeSelector" $genericNodeSelector) -}} {{- $_ := set $cozystackEngineComponents "cozystack-api" (dict "values" $apiValues) -}} -{{- /* lineage-controller-webhook DaemonSet */ -}} -{{- $lineageValues := dict "lineageControllerWebhook" (dict "nodeSelector" $genericNodeSelector) -}} -{{- $_ := set $cozystackEngineComponents "lineage-controller-webhook" (dict "values" $lineageValues) -}} {{- end -}} {{- if .Values.authentication.oidc.enabled }} {{include "cozystack.platform.package" (list "cozystack.cozystack-engine" "oidc" $ $cozystackEngineComponents) }} diff --git a/packages/system/lineage-controller-webhook/Makefile b/packages/system/lineage-controller-webhook/Makefile index d5bada31..cdc1cf01 100644 --- a/packages/system/lineage-controller-webhook/Makefile +++ b/packages/system/lineage-controller-webhook/Makefile @@ -4,8 +4,13 @@ NAMESPACE=cozy-system include ../../../hack/common-envs.mk include ../../../hack/package.mk +.PHONY: test + image: image-lineage-controller-webhook +test: + helm unittest . + image-lineage-controller-webhook: docker buildx build -f images/lineage-controller-webhook/Dockerfile ../../.. \ --tag $(REGISTRY)/lineage-controller-webhook:$(call settag,$(TAG)) \ diff --git a/packages/system/lineage-controller-webhook/README.md b/packages/system/lineage-controller-webhook/README.md new file mode 100644 index 00000000..1e43c9c8 --- /dev/null +++ b/packages/system/lineage-controller-webhook/README.md @@ -0,0 +1,67 @@ +# lineage-controller-webhook + +Cozystack system package for the **lineage controller webhook** — a mutating +admission webhook that stamps "lineage" labels onto tenant workloads, linking +each resource back to the Cozystack `Application` that ultimately owns it. + +The webhook intercepts CREATE and UPDATE on `pods`, `secrets`, `services`, +`persistentvolumeclaims`, `ingresses.networking.k8s.io`, and +`workloadmonitors.cozystack.io` outside system namespaces. For each request it +walks the ownership graph upward (Kubernetes `ownerReferences`, then the +`HelmRelease` Flux installed the resource with, then the Cozystack +`Application`-derived CRD that produced the HelmRelease) and writes the +discovered application's group, kind and name as labels +(`apps.cozystack.io/application.{group,kind,name}`) on the incoming object. +Those labels let the aggregated API server, the Cozystack dashboard, and other +lineage-aware consumers reason about which application a resource belongs to. + +The webhook serves TLS on port 9443 with a cert-manager issued certificate, and +runs with `failurePolicy: Fail` — every CREATE/UPDATE on the resources above +is gated on the webhook being reachable. + +## Topology + +The chart ships a single shape, modelled on `cozystack-api`: + +- **Deployment** with two replicas (override via `replicas`). +- **Soft `nodeAffinity`** preferring `node-role.kubernetes.io/control-plane` + (`Exists`, so it matches both Talos's empty value and k3s/kubeadm's `"true"`). + Soft means: the pod lands on a control-plane node when one is reachable, and + on any worker otherwise — no override needed for managed Kubernetes, + Cozy-in-Cozy tenants, or any other cluster where control-plane nodes aren't + visible. +- **Permissive `tolerations`** (`[{operator: Exists}]`) so the pod can land on + tainted control-plane nodes when the soft affinity is satisfiable. +- **Soft `podAntiAffinity`** on `kubernetes.io/hostname` so replicas spread + across nodes when possible (best-effort). +- **`PodDisruptionBudget`** with `maxUnavailable: 1`, unconditional. At + `replicas: 1` it's a useful no-op (allows full drain); at `replicas >= 2` + it caps disruption to one pod. +- **`Service` with `spec.trafficDistribution: PreferClose`** so the apiserver + prefers a webhook endpoint on its own node when one exists, and transparently + falls over to a remote endpoint otherwise. Requires Kubernetes ≥ 1.31; on + older clusters the field is silently ignored and traffic uses default + cluster-wide distribution. + +## Parameters + +All keys live under the top-level `lineageControllerWebhook:` map. + +| Name | Description | Value | +| ----------------------------- | --------------------------------------------------------- | -------------------------------------------------------------------------------- | +| `image` | Container image (digest-pinned) | `ghcr.io/cozystack/cozystack/lineage-controller-webhook:v1.3.0@sha256:e898…6fb0` | +| `debug` | Enable `--zap-log-level=debug` instead of `info` | `false` | +| `replicas` | Deployment replica count | `2` | +| `localK8sAPIEndpoint.enabled` | **Deprecated.** See note below. | `false` | + +### `localK8sAPIEndpoint.enabled` (deprecated) + +When enabled, this injects `KUBERNETES_SERVICE_HOST=status.hostIP` and +`KUBERNETES_SERVICE_PORT=6443` so the webhook talks to the kube-apiserver on +its own node. It was originally added to avoid latency on the +webhook-to-apiserver path, but it is only valid when the pod is actually +scheduled on an apiserver-bearing node — which the chart's soft control-plane +affinity no longer guarantees. With this flag enabled and the pod scheduled +off a control-plane node, the controller will crash-loop dialing a non- +apiserver hostIP. Slated for removal once the latency motivation is addressed +in the webhook itself; **leave disabled**. diff --git a/packages/system/lineage-controller-webhook/templates/pdb.yaml b/packages/system/lineage-controller-webhook/templates/pdb.yaml new file mode 100644 index 00000000..0786c8ff --- /dev/null +++ b/packages/system/lineage-controller-webhook/templates/pdb.yaml @@ -0,0 +1,9 @@ +apiVersion: policy/v1 +kind: PodDisruptionBudget +metadata: + name: lineage-controller-webhook +spec: + maxUnavailable: 1 + selector: + matchLabels: + app: lineage-controller-webhook diff --git a/packages/system/lineage-controller-webhook/templates/service.yaml b/packages/system/lineage-controller-webhook/templates/service.yaml index c5df90fb..fbb3ad54 100644 --- a/packages/system/lineage-controller-webhook/templates/service.yaml +++ b/packages/system/lineage-controller-webhook/templates/service.yaml @@ -5,7 +5,7 @@ metadata: labels: app: lineage-controller-webhook spec: - internalTrafficPolicy: Local + trafficDistribution: PreferClose type: ClusterIP ports: - port: 443 diff --git a/packages/system/lineage-controller-webhook/templates/daemonset.yaml b/packages/system/lineage-controller-webhook/templates/workload.yaml similarity index 67% rename from packages/system/lineage-controller-webhook/templates/daemonset.yaml rename to packages/system/lineage-controller-webhook/templates/workload.yaml index 860aee6e..097919c6 100644 --- a/packages/system/lineage-controller-webhook/templates/daemonset.yaml +++ b/packages/system/lineage-controller-webhook/templates/workload.yaml @@ -1,10 +1,11 @@ apiVersion: apps/v1 -kind: DaemonSet +kind: Deployment metadata: name: lineage-controller-webhook labels: app: lineage-controller-webhook spec: + replicas: {{ .Values.lineageControllerWebhook.replicas }} selector: matchLabels: app: lineage-controller-webhook @@ -13,13 +14,25 @@ spec: labels: app: lineage-controller-webhook spec: - {{- with .Values.lineageControllerWebhook.nodeSelector }} - nodeSelector: - {{- toYaml . | nindent 8 }} - {{- end }} + serviceAccountName: lineage-controller-webhook tolerations: - operator: Exists - serviceAccountName: lineage-controller-webhook + affinity: + nodeAffinity: + preferredDuringSchedulingIgnoredDuringExecution: + - weight: 100 + preference: + matchExpressions: + - key: node-role.kubernetes.io/control-plane + operator: Exists + podAntiAffinity: + preferredDuringSchedulingIgnoredDuringExecution: + - weight: 100 + podAffinityTerm: + topologyKey: kubernetes.io/hostname + labelSelector: + matchLabels: + app: lineage-controller-webhook containers: - name: lineage-controller-webhook image: "{{ .Values.lineageControllerWebhook.image }}" diff --git a/packages/system/lineage-controller-webhook/tests/service_test.yaml b/packages/system/lineage-controller-webhook/tests/service_test.yaml new file mode 100644 index 00000000..3cdbddd8 --- /dev/null +++ b/packages/system/lineage-controller-webhook/tests/service_test.yaml @@ -0,0 +1,30 @@ +suite: lineage-controller-webhook service +templates: + - templates/service.yaml + +release: + name: lineage-controller-webhook + namespace: cozy-system + +tests: + - it: renders a ClusterIP service on 443 -> 9443 with PreferClose traffic distribution + asserts: + - isKind: + of: Service + - equal: + path: spec.type + value: ClusterIP + - equal: + path: spec.trafficDistribution + value: PreferClose + - notExists: + path: spec.internalTrafficPolicy + - equal: + path: spec.ports[0].port + value: 443 + - equal: + path: spec.ports[0].targetPort + value: 9443 + - equal: + path: spec.selector.app + value: lineage-controller-webhook diff --git a/packages/system/lineage-controller-webhook/tests/workload_test.yaml b/packages/system/lineage-controller-webhook/tests/workload_test.yaml new file mode 100644 index 00000000..bcb24014 --- /dev/null +++ b/packages/system/lineage-controller-webhook/tests/workload_test.yaml @@ -0,0 +1,99 @@ +suite: lineage-controller-webhook workload + PDB +templates: + - templates/workload.yaml + - templates/pdb.yaml + +release: + name: lineage-controller-webhook + namespace: cozy-system + +tests: + # ---- Default Deployment shape ---------------------------------------------- + + - it: defaults render a 2-replica Deployment with soft control-plane nodeAffinity, soft podAntiAffinity, and Exists tolerations + template: templates/workload.yaml + asserts: + - isKind: + of: Deployment + - equal: + path: spec.replicas + value: 2 + - equal: + path: spec.template.spec.tolerations + value: + - operator: Exists + - equal: + path: spec.template.spec.affinity.nodeAffinity.preferredDuringSchedulingIgnoredDuringExecution + value: + - weight: 100 + preference: + matchExpressions: + - key: node-role.kubernetes.io/control-plane + operator: Exists + - notExists: + path: spec.template.spec.affinity.nodeAffinity.requiredDuringSchedulingIgnoredDuringExecution + - equal: + path: spec.template.spec.affinity.podAntiAffinity.preferredDuringSchedulingIgnoredDuringExecution[0].podAffinityTerm.topologyKey + value: kubernetes.io/hostname + + - it: defaults inject no localK8sAPIEndpoint env vars (knob is opt-in) + template: templates/workload.yaml + asserts: + - notExists: + path: spec.template.spec.containers[0].env + + # ---- PDB -------------------------------------------------------------------- + + - it: PDB renders unconditionally with maxUnavailable=1 at the default replica count + template: templates/pdb.yaml + asserts: + - isKind: + of: PodDisruptionBudget + - equal: + path: spec.maxUnavailable + value: 1 + - notExists: + path: spec.minAvailable + + - it: PDB still renders at replicas=1 (no-op, but consistent shape) + set: + lineageControllerWebhook.replicas: 1 + template: templates/pdb.yaml + asserts: + - isKind: + of: PodDisruptionBudget + - equal: + path: spec.maxUnavailable + value: 1 + + # ---- Replica override ------------------------------------------------------- + + - it: replicas value drives spec.replicas + set: + lineageControllerWebhook.replicas: 5 + template: templates/workload.yaml + asserts: + - equal: + path: spec.replicas + value: 5 + + # ---- Deprecated localK8sAPIEndpoint knob ----------------------------------- + + - it: localK8sAPIEndpoint.enabled=true injects KUBERNETES_SERVICE_HOST and PORT env vars + set: + lineageControllerWebhook.localK8sAPIEndpoint.enabled: true + template: templates/workload.yaml + asserts: + - contains: + path: spec.template.spec.containers[0].env + content: + name: KUBERNETES_SERVICE_HOST + valueFrom: + fieldRef: + apiVersion: v1 + fieldPath: status.hostIP + - contains: + path: spec.template.spec.containers[0].env + content: + name: KUBERNETES_SERVICE_PORT + value: "6443" diff --git a/packages/system/lineage-controller-webhook/values.yaml b/packages/system/lineage-controller-webhook/values.yaml index c5671548..e56d332d 100644 --- a/packages/system/lineage-controller-webhook/values.yaml +++ b/packages/system/lineage-controller-webhook/values.yaml @@ -1,10 +1,13 @@ lineageControllerWebhook: image: ghcr.io/cozystack/cozystack/lineage-controller-webhook:v1.3.0@sha256:e8984709686a5eaf19b89da378d7b8c688ea5607e0783a88d9c9e4ccfca96fb0 debug: false + replicas: 2 + # DEPRECATED. Injects KUBERNETES_SERVICE_HOST=status.hostIP and + # KUBERNETES_SERVICE_PORT=6443 so the webhook talks to the apiserver on its + # own node — only valid when the pod is actually scheduled on an apiserver- + # bearing node. Now that the chart's nodeAffinity to control-plane is soft + # (preferred, not required), the pod can land off-control-plane and crash- + # loop dialing a non-apiserver hostIP. Slated for removal once webhook + # performance work obviates the locality motivation. Leave disabled. localK8sAPIEndpoint: - enabled: true - # nodeSelector for the DaemonSet - # Talos uses empty value: "node-role.kubernetes.io/control-plane": "" - # Generic k8s (k3s, kubeadm) uses: "node-role.kubernetes.io/control-plane": "true" - nodeSelector: - node-role.kubernetes.io/control-plane: "" + enabled: false