Update CI and tests for latest fleet versions

Replace deprecated fleet test with fleet apply→target→deploy
workflow.This requires a Kubernetes cluster.

Add k3d test infrastructure with Fleet CRDs and generated expected
output baselines for all 10 examples.

The old fleet test -l env=dev/test/prod generated three files even for
single-cluster examples. All three were identical (no env-based
customizations exist in single-cluster examples).
The new pipeline produces only local-output.yaml, which is the actual
cluster being targeted.

Author: thardeck

.github/workflows/ci-0.5.yml

@@ -14,28 +14,57 @@ jobs:
       fail-fast: false
       matrix:
         fleet_version:
-          - "v0.8.1"
-          - "v0.9.0"
+          - "v0.13.8"
+          - "latest"
 
     steps:
       -
-        uses: actions/checkout@v3
+        uses: actions/checkout@v4
         with:
           fetch-depth: 0
       -
-        uses: actions/cache@v3
+        name: Resolve Fleet Version
+        id: resolve-version
+        run: |
+          VERSION="${{ matrix.fleet_version }}"
+          if [ "$VERSION" = "latest" ]; then
+            VERSION=$(curl -s https://api.github.com/repos/rancher/fleet/releases/latest | jq -r '.tag_name')
+          fi
+          echo "version=${VERSION}" >> $GITHUB_OUTPUT
+      -
+        uses: actions/cache@v4
         id: fleet-cli-cache
         with:
           path: /home/runner/.local/bin
-          key: ${{ runner.os }}-fleet-cli-${{ matrix.fleet_version }}
+          key: ${{ runner.os }}-fleet-cli-${{ steps.resolve-version.outputs.version }}
       -
         name: Download CLI
         if: steps.fleet-cli-cache.outputs.cache-hit != 'true'
         run: |
           mkdir -p /home/runner/.local/bin
-          wget -nv "https://github.com/rancher/fleet/releases/download/${{ matrix.fleet_version }}/fleet-linux-amd64"
+          wget -nv "https://github.com/rancher/fleet/releases/download/${{ steps.resolve-version.outputs.version }}/fleet-linux-amd64"
           mv fleet-linux-amd64 /home/runner/.local/bin/fleet
           chmod +x /home/runner/.local/bin/fleet
+      -
+        name: Set up k3d cluster
+        run: |
+          # Install k3d
+          curl --silent --fail https://raw.githubusercontent.com/k3d-io/k3d/main/install.sh | TAG=v5.8.3 bash
+          # Create a minimal cluster (no load balancer needed)
+          k3d cluster create fleet-test --no-lb --wait --timeout 120s
+          k3d kubeconfig merge fleet-test --kubeconfig-merge-default
+
+          # Install Fleet CRDs for the exact CLI version under test so that
+          # `kubectl apply` and `fleet target` can use the Cluster CRD.
+          FLEET_VER="${{ steps.resolve-version.outputs.version }}"
+          curl -sL "https://github.com/rancher/fleet/releases/download/${FLEET_VER}/fleet-crd-${FLEET_VER#v}.tgz" \
+            | tar -xz -O fleet-crd/templates/crds.yaml \
+            | kubectl apply -f -
+
+          # Create namespaces and minimal cluster objects for targeting
+          kubectl create namespace fleet-local
+          kubectl create namespace fleet-default
+          kubectl apply -f tests/setup-cluster.yaml
       -
         name: Test
         run: |

.github/workflows/ci-0.6.yml

@@ -36,3 +36,72 @@ and configure the app differently for each target.
 
 Using downstream clusters with Windows nodes?
 Check out the [windows-helm](multi-cluster/windows-helm/) multi-cluster example above.
+
+## Running Tests Locally
+
+The test suite renders every example using the Fleet CLI and compares the output
+against committed expected files.
+
+### Prerequisites
+
+- [k3d](https://k3d.io/) v5.8.3 or newer
+- `kubectl`
+- Fleet CLI — download a release binary from
+  [fleet releases](https://github.com/rancher/fleet/releases) and put it on
+  your `PATH`, e.g.:
+  ```bash
+  FLEET_VERSION=v0.14.3
+  curl -sL "https://github.com/rancher/fleet/releases/download/${FLEET_VERSION}/fleet-linux-amd64" \
+    -o ~/.local/bin/fleet && chmod +x ~/.local/bin/fleet
+  ```
+
+### One-time cluster setup
+
+Create a minimal k3d cluster, install Fleet CRDs, and populate the cluster
+objects the tests rely on:
+
+```bash
+FLEET_VERSION=v0.14.3
+
+# Create cluster
+k3d cluster create fleet-test --no-lb --wait --timeout 120s
+k3d kubeconfig merge fleet-test --kubeconfig-merge-default
+
+# Install Fleet CRDs (use the same version as your fleet CLI)
+curl -sL "https://github.com/rancher/fleet/releases/download/${FLEET_VERSION}/fleet-crd-${FLEET_VERSION#v}.tgz" \
+  | tar -xz -O fleet-crd/templates/crds.yaml \
+  | kubectl apply -f -
+
+# Create namespaces and cluster objects
+kubectl create namespace fleet-local
+kubectl create namespace fleet-default
+kubectl apply -f tests/setup-cluster.yaml
+```
+
+The cluster can be reused across test runs and deleted afterwards with
+`k3d cluster delete fleet-test`.
+
+### Running the tests
+
+```bash
+tests/test.sh
+```
+
+The script runs `fleet apply`, `fleet target`, and `fleet deploy -d` for each
+example and compares the rendered output against the files in `tests/expected/`.
+
+### Updating expected files
+
+When an example or the Fleet CLI changes the rendered output, regenerate the
+expected files by running the tests once until they fail on the `diff` step,
+then copy the fresh output into `expected/`:
+
+```bash
+# Run once — will fail at the diff step but generate the new output
+tests/test.sh || true
+
+# Accept the new output as the expected baseline
+cp -r tests/output/* tests/expected/
+# or selectively: cp tests/output/multi-cluster/kustomize/dev-output.yaml \
+#                    tests/expected/multi-cluster/kustomize/dev-output.yaml
+```

.github/workflows/ci.yml

@@ -6,10 +6,8 @@ metadata:
   name: test
   namespace: fleet-default
 spec:
-  correctDrift: {}
   helm:
     chart: https://github.com/rancher/fleet-examples/releases/download/example/guestbook-0.0.1.tgz
-  ignore: {}
   namespace: fleet-mc-helm-external-example
   resources:
   - content: |
@@ -206,65 +204,29 @@ spec:
               env: prod
       ```
     name: README.md
-  - content: |
-      namespace: fleet-mc-helm-external-example
-      helm:
-        chart: https://github.com/rancher/fleet-examples/releases/download/example/guestbook-0.0.1.tgz
-      targetCustomizations:
-      - name: dev
-        helm:
-          values:
-            replication: false
-        clusterSelector:
-          matchLabels:
-            env: dev
-
-      - name: test
-        helm:
-          values:
-            replicas: 3
-        clusterSelector:
-          matchLabels:
-            env: test
-
-      - name: prod
-        helm:
-          values:
-            serviceType: LoadBalancer
-            replicas: 3
-        clusterSelector:
-          matchLabels:
-            env: prod
-    name: fleet.yaml
   targets:
   - clusterSelector:
       matchLabels:
         env: dev
-    correctDrift: {}
     helm:
       chart: https://github.com/rancher/fleet-examples/releases/download/example/guestbook-0.0.1.tgz
       values:
         replication: false
-    ignore: {}
     name: dev
   - clusterSelector:
       matchLabels:
         env: test
-    correctDrift: {}
     helm:
       chart: https://github.com/rancher/fleet-examples/releases/download/example/guestbook-0.0.1.tgz
       values:
         replicas: 3
-    ignore: {}
     name: test
   - clusterSelector:
       matchLabels:
         env: prod
-    correctDrift: {}
     helm:
       chart: https://github.com/rancher/fleet-examples/releases/download/example/guestbook-0.0.1.tgz
       values:
         replicas: 3
         serviceType: LoadBalancer
-    ignore: {}
     name: prod