Merge pull request #260 from nathan-weinberg/aws-runner

ci: move E2E runner from github to AWS

.github/workflows/e2e-nvidia-a10g-x4.yml

@@ -37,7 +37,7 @@ jobs:
           iam-role-name: instructlab-ci-runner
           aws-resource-tags: >
             [
-              {"Key": "Name", "Value": "instructlab-ci-github-runner"},
+              {"Key": "Name", "Value": "instructlab-ci-github-large-runner"},
               {"Key": "GitHubRepository", "Value": "${{ github.repository }}"}
             ]
 

.github/workflows/e2e-nvidia-t4-x1.yml

@@ -0,0 +1,166 @@
+# SPDX-License-Identifier: Apache-2.0
+
+name: E2E (NVIDIA Tesla T4 x1)
+
+on:
+  push:
+    branches:
+      - main
+      - release-*
+  pull_request_target:
+    types:
+      - opened
+      - synchronize
+      - reopened
+    branches:
+      - main
+      - release-*
+
+concurrency:
+  group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }}
+  cancel-in-progress: true
+
+jobs:
+  start-runner:
+    name: Start external EC2 runner
+    runs-on: ubuntu-latest
+    outputs:
+      label: ${{ steps.start-ec2-runner.outputs.label }}
+      ec2-instance-id: ${{ steps.start-ec2-runner.outputs.ec2-instance-id }}
+    steps:
+      - name: Configure AWS credentials
+        uses: aws-actions/configure-aws-credentials@e3dd6a429d7300a6a4c196c26e071d42e0343502 # v4.0.2
+        with:
+          aws-access-key-id: ${{ secrets.AWS_ACCESS_KEY_ID }}
+          aws-secret-access-key: ${{ secrets.AWS_SECRET_ACCESS_KEY }}
+          aws-region: ${{ secrets.AWS_REGION }}
+      - name: Start EC2 runner
+        id: start-ec2-runner
+        uses: machulav/ec2-github-runner@fcfb31a5760dad1314a64a0e172b78ec6fc8a17e # v2.3.6
+        with:
+          mode: start
+          github-token: ${{ secrets.GH_PERSONAL_ACCESS_TOKEN }}
+          ec2-image-id: ami-00c51d9c1374eda97
+          ec2-instance-type: g4dn.2xlarge
+          subnet-id: subnet-02d230cffd9385bd4
+          security-group-id: sg-06300447c4a5fbef3
+          iam-role-name: instructlab-ci-runner
+          aws-resource-tags: >
+            [
+              {"Key": "Name", "Value": "instructlab-ci-github-small-runner"},
+              {"Key": "GitHubRepository", "Value": "${{ github.repository }}"}
+            ]
+
+  e2e:
+    name: E2E Test
+    needs: start-runner
+    runs-on: ${{ needs.start-runner.outputs.label }}
+
+    # It is important that this job has no write permissions and has
+    # no access to any secrets. This part (e2e) is where we are running
+    # untrusted code from PRs.
+    permissions: {}
+
+    # No step-security/harden-runner since this is a self-hosted runner
+    steps:
+      # for debugging
+      - name: Print environment state
+        run: |
+          echo "Current Working Directory: $PWD"
+          echo "Files in Local Directory:"
+          ls -l
+
+      - name: Checkout instructlab/instructlab
+        uses: actions/checkout@692973e3d937129bcbf40652eb9f2f61becf3332 # v4.1.7
+        with:
+          repository: "instructlab/instructlab"
+          path: "instructlab"
+          fetch-depth: 0
+
+      - name: Checkout instructlab/sdg
+        uses: actions/checkout@692973e3d937129bcbf40652eb9f2f61becf3332 # v4.1.7
+        with:
+          repository: "instructlab/sdg"
+          path: "sdg"
+          # https://github.com/actions/checkout/issues/249
+          fetch-depth: 0
+
+      # for debugging
+      - name: Print environment state
+        run: |
+          echo "Current Working Directory: $PWD"
+          echo "Files in Local Directory:"
+          ls -l
+
+      - name: Fetch and checkout PR
+        id: fetch_pr
+        if: github.event_name == 'pull_request_target'
+        working-directory: ./sdg
+        run: |
+          git fetch origin pull/${{ github.event.pull_request.number }}/head:pr-${{ github.event.pull_request.number }}
+          git checkout pr-${{ github.event.pull_request.number }}
+
+      - name: Install system packages
+        run: |
+          cat /etc/os-release
+          sudo dnf install -y gcc gcc-c++ make git python3.11 python3.11-devel
+
+      - name: Install instructlab
+        working-directory: ./instructlab
+        run: |
+          export PATH="/home/ec2-user/.local/bin:/usr/local/cuda/bin:$PATH"
+          python3.11 -m venv --upgrade-deps venv
+          . venv/bin/activate
+          nvidia-smi
+          python3.11 -m pip cache remove llama_cpp_python
+
+          CMAKE_ARGS="-DLLAMA_CUDA=on" python3.11 -m pip install .
+
+          # https://github.com/instructlab/instructlab/issues/1821
+          # install with Torch and build dependencies installed
+          python3.11 -m pip install packaging wheel setuptools-scm
+          python3.11 -m pip install .[cuda]
+        
+      - name: Update instructlab-sdg library
+        working-directory: ./sdg
+        run: |
+          . ../instructlab/venv/bin/activate
+          pip install .
+          pip install .[cuda]        
+
+      - name: Run e2e test
+        working-directory: ./instructlab
+        run: |
+          . venv/bin/activate
+          ./scripts/basic-workflow-tests.sh -m
+
+  stop-runner:
+    name: Stop external EC2 runner
+    needs:
+      - start-runner
+      - e2e
+    runs-on: ubuntu-latest
+    if: ${{ always() }}
+    steps:
+      - name: Configure AWS credentials
+        uses: aws-actions/configure-aws-credentials@e3dd6a429d7300a6a4c196c26e071d42e0343502 # v4.0.2
+        with:
+          aws-access-key-id: ${{ secrets.AWS_ACCESS_KEY_ID }}
+          aws-secret-access-key: ${{ secrets.AWS_SECRET_ACCESS_KEY }}
+          aws-region: ${{ secrets.AWS_REGION }}
+      - name: Stop EC2 runner
+        uses: machulav/ec2-github-runner@fcfb31a5760dad1314a64a0e172b78ec6fc8a17e # v2.3.6
+        with:
+          mode: stop
+          github-token: ${{ secrets.GH_PERSONAL_ACCESS_TOKEN }}
+          label: ${{ needs.start-runner.outputs.label }}
+          ec2-instance-id: ${{ needs.start-runner.outputs.ec2-instance-id }}
+
+  e2e-workflow-complete:
+    # we don't want to block PRs on failed EC2 cleanup
+    # so not requiring "stop-runner" as well
+    needs: ["start-runner", "e2e"]
+    runs-on: ubuntu-latest
+    steps:
+      - name: E2E Workflow Complete
+        run: echo "E2E Workflow Complete"