adds smoke test workflow

users can dispatch a workflow that runs smoke tests against a selected branch Signed-off-by: James Kunstle <[email protected]>
instructlab · Jan 25, 2025 · 36c446a · 36c446a
1 parent 77ae91e
commit 36c446a
Showing 1 changed file with 96 additions and 0 deletions.
diff --git a/.github/workflows/smoke-tests.yaml b/.github/workflows/smoke-tests.yaml
@@ -0,0 +1,96 @@
+# SPDX-License-Identifier: Apache-2.0
+
+name: "Run smoke tests via Tox::pytest"
+# These tests will be long running and require accelerated hardware.
+# They will help to verify that the library is *functionally* correct but
+# will not try to verify that the libary is *correct*.
+
+on:
+  # TEMP - only runs when manually invoked
+  # and only runs against branches in the repo.
+  workflow_dispatch:
+    inputs:
+      branch:
+        type: string
+        default: main
+
+permissions:
+  contents: read
+
+defaults:
+  run:
+    shell: bash
+
+jobs:
+  start-ec2-runner:
+    uses: ./.github/workflows/reusable-start-ec2-runner.yaml
+    with:
+      aws_region: ${{vars.AWS_REGION}}
+      aws_ami: ${{vars.AWS_EC2_AMI}}
+      aws_ec2_runner_variant: ${{vars.SMOKETEST_EC2_INSTANCE_TYPE}} # requires accelerators
+    secrets: inherit
+
+  run-smoke-tests:
+    needs:
+      - start-ec2-runner
+    runs-on: ${{needs.start-ec2-runner.outputs.runner_label}}
+    # It is important that this job has no write permissions and has
+    # no access to any secrets. This part is where we are running
+    # untrusted code from PRs.
+    permissions: {}
+    steps:
+      - name: "Harden runner"
+        uses: step-security/harden-runner@cb605e52c26070c328afc4562f0b4ada7618a84e # v2.10.1
+        with:
+          egress-policy: audit
+
+      - name: "Install packages"
+        run: |
+          cat /etc/os-release
+          sudo dnf install -y gcc gcc-c++ make git python3.11 python3.11-devel
+
+      - name: "Verify cuda environment is setup"
+        run: |
+          export CUDA_HOME="/usr/local/cuda"
+          export LD_LIBRARY_PATH="$LD_LIBRARY_PATH:/usr/local/cuda/lib64:/usr/local/cuda/extras/CUPTI/lib64"
+          export PATH="$PATH:$CUDA_HOME/bin"
+          nvidia-smi
+
+      - name: "Checkout code"
+        uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 # v4.2.2
+        with:
+          fetch-depth: 0
+          ref: ${{inputs.branch}}
+
+      # installs in $GITHUB_WORKSPACE/venv.
+      # only has to install Tox because Tox will do the other virtual environment management.
+      - name: "Setup Python virtual environment"
+        run: |
+          python3.11 -m venv --upgrade-deps venv
+          . venv/bin/activate
+          pip install tox
+
+      - name: "Show disk utilization BEFORE tests"
+        run: |
+          df -h
+
+      - name: "Run unit tests with Tox and Pytest"
+        run: |
+          source venv/bin/activate
+          tox -e py3-smoke
+
+      - name: "Show disk utilization AFTER tests"
+        run: |
+          df -h
+
+  stop-ec2-runner:
+    needs:
+      - start-ec2-runner
+      - run-smoke-tests
+    if: ${{ always() }}
+    uses: ./.github/workflows/reusable-stop-ec2-runner.yaml
+    with:
+      aws_region: ${{vars.AWS_REGION}}
+      aws_ec2_runner_label: ${{needs.start-ec2-runner.outputs.runner_label}}
+      aws_ec2_runner_instance_id: ${{needs.start-ec2-runner.outputs.runner_instance_id}}
+    secrets: inherit