Merge branch 'main' into fix/remove-pod-event-time-cache

Author: flavono123

.github/actions/install-deps/action.yaml

@@ -7,7 +7,7 @@ inputs:
 runs:
   using: "composite"
   steps:
-    - uses: actions/setup-go@0aaccfd150d50ccaeb58ebd88d36e91967a5f35b # v5.4.0
+    - uses: actions/setup-go@d35c59abb061a4a6fb18e82ac0862c26744d6ab5 # v5.5.0
       id: setup-go
       with:
         go-version-file: go.mod

.github/dependabot.yaml

@@ -61,3 +61,11 @@ updates:
       action-deps:
         patterns:
           - '*'
+  - package-ecosystem: github-actions
+    directory: .github/actions/run-bench-test
+    schedule:
+      interval: weekly
+    groups:
+      action-deps:
+        patterns:
+          - '*'

.github/workflows/kind-e2e.yaml

@@ -1,5 +1,7 @@
 name: kind-e2e
 on:
+  push:
+    branches: [main]
   workflow_dispatch:
 jobs:
   kind-e2e:
@@ -13,7 +15,7 @@ jobs:
     steps:
     - uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 # v4.2.2
     - name: Set up Python 3.10
-      uses: actions/setup-python@8d9ed9ac5c53483de85588cdf95a591a75ab9f55 # v5.5.0
+      uses: actions/setup-python@a26af69be951a213d495a4c3e4e4022e16d87065 # v5.6.0
       with:
         python-version: "3.10"
     - uses: ./.github/actions/install-deps
@@ -34,7 +36,6 @@ jobs:
     - name: install kwok and controller
       shell: bash
       run: |
-        make toolchain
         make install-kwok
         export KWOK_REPO=kind.local
         export KIND_CLUSTER_NAME=chart-testing
@@ -47,31 +48,36 @@ jobs:
         kubectl get nodepools
         kubectl get pods -A
         kubectl describe nodes
-    - uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
-      with:
-        repository: nathangeology/karpenter_evaluate
-        path: ./karpenter_eval/ # Installs to a folder in the Karpenter repo for the test
-        ref: "1130af927302e6913a4947952112f793eeafc564"
-        fetch-depth: 0
-    - name: install KPI report dependencies
-      shell: bash
-      run: |
-        pip install pandas==2.2.2
-        pip install pyarrow==16.1.0 
-        pip install tabulate==0.9.0
-        pip install prometheus-api-client==0.5.5
-        pip install ./karpenter_eval/
+    # TEMPORARILY DISABLED: Karpenter KPI Analysis Package
+    # This package analyzes key performance indicators (KPIs) for integration tests.
+    # Currently non-functional due to changes in core Karpenter metrics.
+    # Disabled due to test flakiness until underlying metric changes are addressed.
+    # Reference: https://github.com/nathangeology/karpenter_evaluate/blob/main/main.py
+    # - uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
+    #   with:
+    #     repository: nathangeology/karpenter_evaluate
+    #     path: ./karpenter_eval/ # Installs to a folder in the Karpenter repo for the test
+    #     ref: "1130af927302e6913a4947952112f793eeafc564"
+    #     fetch-depth: 0
+    # - name: install KPI report dependencies
+    #   shell: bash
+    #   run: |
+    #     pip install pandas==2.2.2
+    #     pip install pyarrow==16.1.0 
+    #     pip install tabulate==0.9.0
+    #     pip install prometheus-api-client==0.5.5
+    #     pip install ./karpenter_eval/
     - name: run test suites
       shell: bash
       run: |
         OUTPUT_DIR=$(mktemp -d)
         export OUTPUT_DIR
         echo OUTPUT_DIR="$OUTPUT_DIR" >> "$GITHUB_ENV"
         make e2etests
-    - name: run test analysis
-      shell: bash
-      run: |
-        OUTPUT_DIR=${{ env.OUTPUT_DIR }} python ./karpenter_eval/main.py
+    # - name: run test analysis
+    #   shell: bash
+    #   run: |
+    #     OUTPUT_DIR=${{ env.OUTPUT_DIR }} python ./karpenter_eval/main.py
     - name: cleanup 
       shell: bash
       run: | 

.github/workflows/release.yaml

@@ -32,7 +32,7 @@ jobs:
           prerelease: false
 
       - name: Install tejolote
-        uses: kubernetes-sigs/release-actions/setup-tejolote@a69972745f85aab4ba5d6c681e2a0e7f73eaff2b # v0.3.0
+        uses: kubernetes-sigs/release-actions/setup-tejolote@a30d93cf2aa029e1e4c8a6c79f766aebf429fddb # v0.3.1
       - name: Run tejolote
         run: |
           tejolote attest "github://kubernetes-sigs/karpenter/${{ github.run_id }}" --artifacts "github://kubernetes-sigs/karpenter/$TAG" --output karpenter.intoto.json --sign

.github/workflows/run-bench-test.yaml

@@ -0,0 +1,41 @@
+name: "Run Bench Test"
+
+on:
+  workflow_call:
+    inputs:
+      path:
+        description: Path to the performance test
+        required: true
+        type: string
+      runName:
+        description: Name of the run, for the purpose of file naming and github comments
+        required: true
+        type: string
+      githubSha:
+        description: Sha of the github commit to check out
+        required: true
+        type: string
+jobs:
+  run-test:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v4
+        with:
+          ref: ${{ inputs.githubSha }}
+      - name: Run Test
+        id: test-run
+        shell: bash
+        run: |
+          {
+            cd ${{ inputs.path }}
+            mkdir output
+            go test -tags=test_performance -run=1 -bench=. -count=1 -cpuprofile output/cpu.out -memprofile output/mem.out > output/results.txt
+            echo 'OUTPUT<<EOF'
+            cat output/results.txt
+            echo EOF
+          } >> "$GITHUB_OUTPUT"
+      - uses: actions/upload-artifact@v4
+        id: artifact-upload
+        with:
+          name: ${{ inputs.runName }}
+          path: ${{ inputs.path }}/output

.github/workflows/scheduling-benchmarking.yaml

@@ -0,0 +1,36 @@
+name: "Scheduling Benchmarking"
+
+on:
+  pull_request:
+    paths:
+      - "**.go"
+    branches:
+      - main
+
+jobs:
+  before:
+    name: Before PR
+    strategy:
+      fail-fast: false
+      matrix:
+        path: ["pkg/controllers/provisioning/scheduling"]
+    uses: ./.github/workflows/run-bench-test.yaml
+    secrets: inherit
+    with:
+      path: ${{ matrix.path }}
+      runName: before-${{ strategy.job-index }}
+      githubSha: ${{ github.event.pull_request.base.sha }}
+  after:
+    name: After PR
+    permissions:
+      pull-requests: write
+    strategy:
+      fail-fast: false
+      matrix:
+        path: ["pkg/controllers/provisioning/scheduling"]
+    uses: ./.github/workflows/run-bench-test.yaml
+    secrets: inherit
+    with:
+      path: ${{ matrix.path }}
+      runName: after-${{ strategy.job-index }}  
+      githubSha: ${{ github.sha }}

Makefile

@@ -6,7 +6,8 @@ HELM_OPTS ?= --set logLevel=debug \
 			--set controller.resources.requests.cpu=1 \
 			--set controller.resources.requests.memory=1Gi \
 			--set controller.resources.limits.cpu=1 \
-			--set controller.resources.limits.memory=1Gi
+			--set controller.resources.limits.memory=1Gi \
+			--set settings.featureGates.nodeRepair=true
 
 help: ## Display help
 	@awk 'BEGIN {FS = ":.*##"; printf "Usage:\n  make \033[36m<target>\033[0m\n"} /^[a-zA-Z_0-9-]+:.*?##/ { printf "  \033[36m%-15s\033[0m %s\n", $$1, $$2 } /^##@/ { printf "\n\033[1m%s\033[0m\n", substr($$0, 5) } ' $(MAKEFILE_LIST)
@@ -43,11 +44,12 @@ apply-with-kind: verify build-with-kind ## Deploy the kwok controller from the c
 e2etests: ## Run the e2e suite against your local cluster
 	cd test && go test \
 		-count 1 \
-		-timeout 30m \
+		-timeout 2h \
 		-v \
 		./suites/$(shell echo $(TEST_SUITE) | tr A-Z a-z)/... \
 		--ginkgo.focus="${FOCUS}" \
-		--ginkgo.timeout=30m \
+		--ginkgo.skip="${SKIP}" \
+		--ginkgo.timeout=2h \
 		--ginkgo.grace-period=5m \
 		--ginkgo.vv
 
@@ -60,6 +62,7 @@ apply: verify build ## Deploy the kwok controller from the current state of your
 		--set controller.image.repository=$(IMG_REPOSITORY) \
 		--set controller.image.tag=$(IMG_TAG) \
 		--set controller.image.digest=$(IMG_DIGEST) \
+		--set settings.preferencePolicy=Ignore \
 		--set-string controller.env[0].name=ENABLE_PROFILING \
 		--set-string controller.env[0].value=true