Merge branch 'master' into simplex-tracker

Signed-off-by: Sam Liokumovich <[email protected]>

Author: samliok

.github/actionlint.yml

@@ -2,3 +2,4 @@ self-hosted-runner:
   labels:
     - custom-arm64-jammy
     - custom-arm64-noble
+    - avalanche-avalanchego-runner # Github Action Runner Controller

.github/actions/run-monitored-tmpnet-cmd/action.yml

@@ -40,6 +40,9 @@ inputs:
     default: ${{ github.run_attempt }}
   job:
     default: ${{ github.job }}
+  grafana_dashboard_id:
+    description: 'The identifier of the Grafana dashboard to use, in the format <UID>/<dashboard-name>.'
+    default: 'kBQpRdWnk/avalanche-main-dashboard'
 
 runs:
   using: composite
@@ -51,14 +54,14 @@ runs:
     - uses: cachix/install-nix-action@02a151ada4993995686f9ed4f1be7cfbb229e56f #v31
       with:
         github_access_token: ${{ inputs.github_token }}
-    - run: ${{ github.action_path }}/nix-develop.sh --command echo "dependencies installed"
+    - run: $GITHUB_ACTION_PATH/nix-develop.sh --command echo "dependencies installed"
       shell: bash
     - name: Notify of metrics availability
       if: (inputs.prometheus_username != '')
       shell: bash
-      run: ${{ github.action_path }}/notify-metrics-availability.sh
+      run: $GITHUB_ACTION_PATH/notify-metrics-availability.sh
       env:
-        GRAFANA_URL: https://grafana-poc.avax-dev.network/d/kBQpRdWnk/avalanche-main-dashboard?orgId=1&refresh=10s&var-filter=is_ephemeral_node%7C%3D%7Cfalse&var-filter=gh_repo%7C%3D%7C${{ inputs.repository_owner }}%2F${{ inputs.repository_name }}&var-filter=gh_run_id%7C%3D%7C${{ inputs.run_id }}&var-filter=gh_run_attempt%7C%3D%7C${{ inputs.run_attempt }}
+        GRAFANA_URL: https://grafana-poc.avax-dev.network/d/${{ inputs.grafana_dashboard_id }}?orgId=1&refresh=10s&var-filter=is_ephemeral_node%7C%3D%7Cfalse&var-filter=gh_repo%7C%3D%7C${{ inputs.repository_owner }}%2F${{ inputs.repository_name }}&var-filter=gh_run_id%7C%3D%7C${{ inputs.run_id }}&var-filter=gh_run_attempt%7C%3D%7C${{ inputs.run_attempt }}
         GH_JOB_ID: ${{ inputs.job }}
         FILTER_BY_OWNER: ${{ inputs.filter_by_owner }}
     - name: Warn that collection of metrics and logs will not be performed
@@ -68,7 +71,7 @@ runs:
     - name: Run command
       shell: bash
       # --impure ensures the env vars are accessible to the command
-      run: ${{ inputs.run_env }} ${{ github.action_path }}/nix-develop.sh --impure --command bash -x ${{ inputs.run }}
+      run: ${{ inputs.run_env }} $GITHUB_ACTION_PATH/nix-develop.sh --impure --command bash -x ${{ inputs.run }}
       env:
         # Always collect metrics locally even when nodes are running in kube to enable collection from the test workload
         TMPNET_START_METRICS_COLLECTOR: ${{ inputs.prometheus_username != '' }}

.github/workflows/ci.yml

@@ -140,7 +140,7 @@ jobs:
           # responsible for pushes.
           push: false
           # This version should match the version installed in the nix dev shell
-          version: 1.47.2
+          version: 1.52.1
   links-lint:
     name: Markdown Links Lint
     runs-on: ubuntu-latest
@@ -233,7 +233,7 @@ jobs:
         shell: bash
         run: nix develop --command ./scripts/run_task.sh test-bootstrap-monitor-e2e
   load:
-    name: Run load test
+    name: Run process-based load test
     runs-on: ubuntu-latest
     steps:
       - uses: actions/checkout@v4
@@ -246,15 +246,15 @@ jobs:
           prometheus_password: ${{ secrets.PROMETHEUS_PASSWORD || '' }}
           loki_username: ${{ secrets.LOKI_ID || '' }}
           loki_password: ${{ secrets.LOKI_PASSWORD || '' }}
-  load_kube:
-    name: Run load test on kube
+  load_kube_kind:
+    name: Run load test on kind cluster
     runs-on: ubuntu-latest
     steps:
       - uses: actions/checkout@v4
       - uses: ./.github/actions/setup-go-for-project
       - uses: ./.github/actions/run-monitored-tmpnet-cmd
         with:
-          run: ./scripts/run_task.sh test-load-kube
+          run: ./scripts/run_task.sh test-load-kube-kind
           artifact_prefix: load-kube
           prometheus_username: ${{ secrets.PROMETHEUS_ID || '' }}
           prometheus_password: ${{ secrets.PROMETHEUS_PASSWORD || '' }}
@@ -274,4 +274,3 @@ jobs:
             prometheus_password: ${{ secrets.PROMETHEUS_PASSWORD || '' }}
             loki_username: ${{ secrets.LOKI_ID || '' }}
             loki_password: ${{ secrets.LOKI_PASSWORD || '' }}
-

.github/workflows/self-hosted-load-tests.yaml

@@ -0,0 +1,50 @@
+name: Load test on self-hosted runners
+
+# This workflow runs load tests against our Kubernetes cluster
+on:
+  workflow_dispatch:
+    inputs:
+      avalanchego_image:
+        description: 'AvalancheGo image to test'
+        required: false
+        default: 'avaplatform/avalanchego:latest'
+        type: string
+      exclusive_scheduling:
+        description: 'Enable exclusive scheduling'
+        required: false
+        default: false
+        type: boolean
+      duration:
+        description: "Load test duration: e.g. 5m, 10m, 1h..."
+        required: false
+
+jobs:
+  load_test:
+    name: Run load test on self-hosted runners
+    runs-on: avalanche-avalanchego-runner
+    container:
+      image: ghcr.io/actions/actions-runner:2.325.0
+    steps:
+      - name: Install dependencies
+        shell: bash
+        # The xz-utils might be present on some containers
+        run: |
+          if ! command -v xz &> /dev/null; then
+            sudo apt-get update
+            sudo apt-get install -y xz-utils
+          fi
+      - uses: actions/checkout@v4
+      - uses: ./.github/actions/setup-go-for-project
+      - name: Run load test
+        uses: ./.github/actions/run-monitored-tmpnet-cmd
+        with:
+          run: >-
+            ./scripts/run_task.sh test-load-kube --
+            --kube-image ${{ inputs.avalanchego_image }}
+            ${{ inputs.exclusive_scheduling == 'true' && '--kube-use-exclusive-scheduling' || '' }}
+            ${{ inputs.duration && format('--duration {0}', inputs.duration) || '' }}
+          artifact_prefix: self-hosted-load-test${{ inputs.exclusive_scheduling == 'true' && '-exclusive' || '' }}
+          prometheus_username: ${{ secrets.PROMETHEUS_ID || '' }}
+          prometheus_password: ${{ secrets.PROMETHEUS_PASSWORD || '' }}
+          loki_username: ${{ secrets.LOKI_ID || '' }}
+          loki_password: ${{ secrets.LOKI_PASSWORD || '' }}

Taskfile.yml

@@ -199,11 +199,12 @@ tasks:
       - cmd: bash -x ./scripts/tests.e2e.kube.sh {{.CLI_ARGS}}
 
   test-e2e-kube-ci:
-    desc: Runs e2e tests against a network deployed to kube [serially]
+    # Free github action runners do not have sufficient resources to reliably run a full e2e run against a kube-hosted network
+    desc: Runs the xsvm e2e tests in serial against a network deployed to kube
     env:
       E2E_SERIAL: 1
     cmds:
-      - task: test-e2e-kube
+      - cmd: bash -x ./scripts/tests.e2e.kube.sh --ginkgo.focus-file=xsvm.go {{.CLI_ARGS}}
 
   # To use a different fuzz time, run `task test-fuzz FUZZTIME=[value in seconds]`.
   # A value of `-1` will run until it encounters a failing output.
@@ -233,15 +234,9 @@ tasks:
       - task: build
       - cmd: go run ./tests/load/c/main --avalanchego-path=./build/avalanchego {{.CLI_ARGS}}
 
-  test-load-kube:
-    desc: Runs load tests against a network deployed to kube
-    cmds:
-      - task: generate-load-contract-bindings
-      - cmd: bash -x ./scripts/tests.load.kube.sh {{.CLI_ARGS}}
-
   test-load2:
     desc: Runs second iteration of load tests
-    cmds: 
+    cmds:
       - task: build
       - cmd: go run ./tests/load2/main --avalanchego-path=./build/avalanchego {{.CLI_ARGS}}
 
@@ -250,6 +245,18 @@ tasks:
     cmds:
       - cmd: go run ./tests/load/c/main --runtime=kube --kube-use-exclusive-scheduling {{.CLI_ARGS}}
 
+  test-load-kube:
+    desc: Runs load tests against a kubernetes cluster
+    cmds:
+      - task: generate-load-contract-bindings
+      - cmd: go run ./tests/load/c/main --runtime=kube {{.CLI_ARGS}}
+
+  test-load-kube-kind:
+    desc: Runs load tests against a kind cluster
+    cmds:
+      - task: generate-load-contract-bindings
+      - cmd: bash -x ./scripts/tests.load.kube.kind.sh {{.CLI_ARGS}}
+
   test-unit:
     desc: Runs unit tests
     # Invoking with bash ensures compatibility with CI execution on Windows