Merge branch 'master' into integrate_sanity_tests_with_pytest

.github/dependabot.yml

@@ -0,0 +1,24 @@
+# To get started with Dependabot version updates, you'll need to specify which
+# package ecosystems to update and where the package manifests are located.
+# Please see the documentation for all configuration options:
+# https://docs.github.com/github/administering-a-repository/configuration-options-for-dependency-updates
+
+version: 2
+updates:
+  # Configuration for Python dependencies
+  - package-ecosystem: "pip"
+    directory: "/"
+    schedule:
+      interval: "monthly"
+
+  # Configuration for frontend/gradle.properties
+  - package-ecosystem: "gradle"
+    directory: "/frontend"
+    schedule:
+      interval: "monthly"
+
+  # Configuration for plugins/gradle.properties
+  - package-ecosystem: "gradle"
+    directory: "/plugins"
+    schedule:
+      interval: "monthly"

.github/pytorch-probot.yml

@@ -0,0 +1,13 @@
+tracking_issue: 2724
+
+# List of workflows that will be re-run in case of failures
+# https://github.com/pytorch/test-infra/blob/main/torchci/lib/bot/retryBot.ts
+retryable_workflows:
+- Run Regression Tests on Docker
+- Run Regression Tests for CPU nightly binaries
+- Push torchserve nightly
+- Push Docker Nightly
+- Docker CI
+- CI CPU
+- CI GPU
+- Benchmark torchserve nightly

.github/workflows/ci_cpu.yml

@@ -1,6 +1,7 @@
 name: CI CPU
 
 on:
+  workflow_dispatch:
   push:
     branches:
       - master

.github/workflows/ci_gpu.yml

@@ -1,6 +1,7 @@
 name: CI GPU
 
 on:
+  workflow_dispatch:
   push:
     branches:
       - master
@@ -20,11 +21,12 @@ jobs:
       - name: Clean up previous run
         run: |
           echo "Cleaning up previous run"
-          cd $RUNNER_WORKSPACE
-          pwd
-          cd ..
-          pwd
-          rm -rf _tool
+          ls -la ./
+          # deletes all files in the current directory
+          sudo rm -rf ./* || true
+          # deletes all hidden files in a directory
+          sudo rm -rf ./.??* || true
+          ls -la ./
       - name: Setup Python 3.8
         uses: actions/setup-python@v4
         with:
@@ -39,7 +41,7 @@ jobs:
         uses: actions/checkout@v3
       - name: Install dependencies
         run: |
-          python ts_scripts/install_dependencies.py --environment=dev --cuda=cu117
+          python ts_scripts/install_dependencies.py --environment=dev --cuda=cu121
       - name: Torchserve Sanity
         uses: nick-fields/retry@v2
         with:

.github/workflows/codeql.yml

@@ -12,6 +12,7 @@
 name: "CodeQL"
 
 on:
+  workflow_dispatch:
   # run every day at 11:15am
   schedule:
     - cron:  '15 11 * * *'

.github/workflows/docker-nightly-build.yml

@@ -3,11 +3,18 @@ name: Push Docker Nightly
 on:
   # run every day at 1:15pm
   schedule:
-    - cron:  '15 13 * * *'
+    - cron: "15 13 * * *"
 jobs:
   nightly:
-    runs-on: ubuntu-20.04
+    runs-on: [self-hosted, ci-gpu]
     steps:
+      - name: Clean up previous run
+        run: |
+          echo "Cleaning up previous run"
+          ls -la ./
+          sudo rm -rf ./* || true
+          sudo rm -rf ./.??* || true
+          ls -la ./
       - name: Setup Python 3.8
         uses: actions/setup-python@v4
         with:
@@ -22,13 +29,17 @@ jobs:
       - name: Push Docker Nightly
         run: |
           cd docker
-          python docker_nightly.py
+          python docker_nightly.py --cleanup
+      - name: Push KServe Docker Nightly
+        run: |
+          cd kubernetes/kserve
+          python docker_nightly.py --cleanup
 
       - name: Open issue on failure
         if: ${{ failure() && github.event_name  == 'schedule' }}
         uses: dacbd/create-issue-action@v1
         with:
           token: ${{ secrets.GITHUB_TOKEN }}
           title: Nightly Docker build failed
-          body:  Commit ${{ github.sha }} daily scheduled [CI run](https://github.com/${{ github.repository }}/actions/runs/${{ github.run_id }}) failed, please check why
-          assignees: ''
+          body: Commit ${{ github.sha }} daily scheduled [CI run](https://github.com/${{ github.repository }}/actions/runs/${{ github.run_id }}) failed, please check why
+          assignees: ""

.github/workflows/lint.yml

@@ -51,12 +51,6 @@ jobs:
           pip install pre-commit
           pre-commit install
 
-      - name: Check links in all markdown files
-        uses: gaurav-nelson/github-action-markdown-link-check@1.0.13
-        with:
-          use-verbose-mode: 'yes'
-          config-file: "ts_scripts/markdown_link_check_config.json"
-
       - name: Get changed files
         id: changed-files
         uses: tj-actions/changed-files@v29.0.4

.github/workflows/official_release.yml

@@ -4,16 +4,10 @@ on: workflow_dispatch
 
 jobs:
   official-release:
-    # creates workflows for the 3 OS: ubuntu, macOS & windows
-    runs-on: ${{ matrix.os }}
-
+    runs-on: ubuntu-20.04
     # The official-release environment requires 2 manual approvals
     environment:
       name: official-release
-    strategy:
-      fail-fast: true
-      matrix:
-        os: [ubuntu-20.04, macOS-latest, windows-latest]
     steps:
       - name: Setup Conda
         uses: s-weigand/setup-conda@v1
@@ -41,22 +35,4 @@ jobs:
       - name: Push PyPI binaries
         env:
           TWINE_PASSWORD: ${{ secrets.PYPI_TOKEN }}
-        if: matrix.os == 'ubuntu-20.04'
         run: python binaries/upload.py --upload-pypi-packages
-      - name: Login to Docker
-        env:
-          DOCKER_PASSWORD: ${{secrets.DOCKER_PASSWORD}}
-        if: matrix.os == 'ubuntu-20.04'
-        run: docker login --username pytorchbot --password "$DOCKER_PASSWORD"
-      - name: Build & Upload pytorch/torchserve Docker images
-        if: matrix.os == 'ubuntu-20.04'
-        run: |
-          cd docker
-          python build_upload_release.py
-          cd ..
-      - name: Build & Upload pytorch/torchserve-kfs Docker images
-        if: matrix.os == 'ubuntu-20.04'
-        run: |
-          cd kubernetes/kserve
-          python build_upload_release.py
-          cd ..

.github/workflows/official_release_docker.yml

@@ -0,0 +1,46 @@
+name: Make an official Release of Docker Images
+
+on:
+  workflow_dispatch:
+    inputs:
+          upload_docker:
+            description: 'Upload TorchServe Docker Images?'
+            required: true
+            default: 'no'
+          upload_kfs:
+            description: 'Upload TorchServe-kfs Docker Images?'
+            required: true
+            default: 'no'
+
+jobs:
+  official-release-docker:
+    runs-on: [self-hosted, ci-gpu]
+    steps:
+      - name: Clean up previous run
+        run: |
+          echo "Cleaning up previous run"
+          ls -la ./
+          sudo rm -rf ./* || true
+          sudo rm -rf ./.??* || true
+          ls -la ./
+      - name: Setup Python 3.8
+        uses: actions/setup-python@v4
+        with:
+          python-version: 3.8
+          architecture: x64
+      - name: Checkout TorchServe
+        uses: actions/checkout@v3
+      - name: Login to Docker
+        env:
+          DOCKER_PASSWORD: ${{secrets.DOCKER_PASSWORD}}
+        run: docker login --username pytorchbot --password "$DOCKER_PASSWORD"
+      - name: Build & Upload pytorch/torchserve Docker images
+        if: github.event.inputs.upload_docker == 'yes'
+        run: |
+          cd docker
+          python build_upload_release.py --cleanup
+      - name: Build & Upload pytorch/torchserve-kfs Docker images
+        if: github.event.inputs.upload_kfs == 'yes'
+        run: |
+          cd kubernetes/kserve
+          python build_upload_release.py --cleanup

.github/workflows/regression_tests_cpu_binaries.yml

@@ -0,0 +1,44 @@
+name: Run Regression Tests for CPU nightly binaries
+
+on:
+  # run every day at 6:15am
+  schedule:
+    - cron:  '15 6 * * *'
+
+concurrency:
+  group: ci-cpu-${{ github.workflow }}-${{ github.ref == 'refs/heads/master' && github.run_number || github.ref }}
+  cancel-in-progress: true
+
+jobs:
+  regression-cpu-nightly-binaries:
+    # creates workflows for OS: ubuntu, macOS
+    runs-on: ${{ matrix.os }}
+    strategy:
+      fail-fast: false
+      matrix:
+        os: [ubuntu-20.04, macOS-latest]
+        python-version: ["3.8", "3.9", "3.10"]
+        binaries: ["pypi", "conda"]
+    steps:
+      - uses: actions/checkout@v3
+      - name: Setup conda with Python ${{ matrix.python-version }}
+        uses: s-weigand/setup-conda@v1
+        with:
+          update-conda: true
+          python-version: ${{ matrix.python-version }}
+          conda-channels: anaconda, conda-forge
+      - run: conda --version
+      - run: python --version
+      - name: Setup Java 17
+        uses: actions/setup-java@v3
+        with:
+          distribution: 'zulu'
+          java-version: '17'
+      - name: Checkout TorchServe
+        uses: actions/checkout@v3
+      - name: Install dependencies
+        run: |
+          python ts_scripts/install_dependencies.py --environment=dev
+      - name: Torchserve Regression Tests
+        run: |
+          python test/regression_tests.py --binaries --${{ matrix.binaries }} --nightly

.github/workflows/regression_tests_docker.yml

@@ -1,6 +1,7 @@
 name: Run Regression Tests on Docker
 
 on:
+  workflow_dispatch:
   # run every day at 5:15am
   schedule:
     - cron:  '15 5 * * *'
@@ -25,7 +26,7 @@ jobs:
           sudo rm -rf ./* || true
           sudo rm -rf ./.??* || true
           ls -la ./
-          docker system prune -f
+          docker system prune --all --volumes -f
       - name: Checkout TorchServe
         uses: actions/checkout@v3
       - name: Branch name
@@ -40,7 +41,7 @@ jobs:
         if: false == contains(matrix.hardware, 'ubuntu')
         run: |
           cd docker
-          ./build_image.sh -g -cv cu117 -bt ci -n -b $GITHUB_REF_NAME -t pytorch/torchserve:ci
+          ./build_image.sh -g -cv cu121 -bt ci -n -b $GITHUB_REF_NAME -t pytorch/torchserve:ci
       - name: Torchserve GPU Regression Tests
         if: false == contains(matrix.hardware, 'ubuntu')
         run: |
@@ -53,3 +54,11 @@ jobs:
         if: success()
         run: |
           docker system prune -f && docker rmi pytorch/torchserve:ci
+      - name: Open issue on failure
+        if: ${{ failure() && github.event_name  == 'schedule' }}
+        uses: dacbd/create-issue-action@v1
+        with:
+          token: ${{ secrets.GITHUB_TOKEN }}
+          title: Docker Regression on ${{ matrix.hardware }} failed
+          body:  The daily scheduled [Docker Regression Run](https://github.com/${{ github.repository }}/actions/runs/${{ github.run_id }}) failed, please check why
+          assignees: ''

.github/workflows/regression_tests_gpu.yml

@@ -1,6 +1,7 @@
 name: Run Regression Tests on GPU
 
 on:
+  workflow_dispatch:
   push:
     branches:
       - master
@@ -15,7 +16,7 @@ concurrency:
 
 jobs:
   regression-gpu:
-    # creates workflows for CUDA 11.6 & CUDA 11.7 on ubuntu
+    # creates workflows on self hosted runner
     runs-on: [self-hosted, regression-test-gpu]
     steps:
       - name: Clean up previous run
@@ -43,7 +44,8 @@ jobs:
         uses: actions/checkout@v3
       - name: Install dependencies
         run: |
-          python ts_scripts/install_dependencies.py --environment=dev --cuda=cu117
+          python ts_scripts/install_dependencies.py --environment=dev --cuda=cu121
       - name: Torchserve Regression Tests
         run: |
+          export TS_RUN_IN_DOCKER=False
           python test/regression_tests.py

.github/workflows/regression_tests_gpu_binaries.yml

@@ -0,0 +1,54 @@
+name: Run Regression Tests for GPU nightly binaries
+
+on:
+  # run every day at 6:15am
+  workflow_dispatch:
+
+concurrency:
+  group: ci-cpu-${{ github.workflow }}-${{ github.ref == 'refs/heads/master' && github.run_number || github.ref }}
+  cancel-in-progress: true
+
+jobs:
+  regression-gpu-nightly-binaries:
+    # creates workflows for OS: ubuntu, macOS
+    runs-on: [self-hosted, regression-test-gpu]
+    strategy:
+      fail-fast: false
+      matrix:
+        python-version: ["3.8", "3.9", "3.10"]
+        binaries: ["pypi", "conda"]
+    steps:
+      - name: Clean up previous run
+        run: |
+          echo "Cleaning up previous run"
+          ls -la ./
+          sudo rm -rf ./* || true
+          sudo rm -rf ./.??* || true
+          ls -la ./
+      - name: Setup Java 17
+        uses: actions/setup-java@v3
+        with:
+          distribution: 'zulu'
+          java-version: '17'
+      - name: Checkout TorchServe
+        uses: actions/checkout@v3
+      - uses: conda-incubator/setup-miniconda@v2
+        with:
+          miniconda-version: "latest"
+      - name: Setup Conda
+        uses: s-weigand/setup-conda@v1
+        with:
+          python-version: ${{ matrix.python-version }}
+      - name: Setup Python ${{ matrix.python_version }}
+        uses: actions/setup-python@v3
+        with:
+          python-version: ${{ matrix.python-version }}
+          architecture: x64
+      - run: conda --version
+      - run: python --version
+      - name: Install dependencies
+        run: |
+          python ts_scripts/install_dependencies.py --environment=dev --cuda=cu118
+      - name: Torchserve Regression Tests
+        run: |
+          python test/regression_tests.py --binaries --${{ matrix.binaries }} --nightly