FastDeploy/.github/workflows/_golang_router_test.yml

name: GOLANG_ROUTER Tests
description: "Run FastDeploy golang_router tests"

on:
  workflow_call:
    inputs:
      DOCKER_IMAGE:
        description: "Build Images"
        required: true
        type: string
        default: "ccr-2vdh3abv-pub.cnc.bj.baidubce.com/paddlepaddle/paddleqa:fastdeploy-ciuse-cuda126-paddle-dev"
      FASTDEPLOY_ARCHIVE_URL:
        description: "URL of the compressed FastDeploy code archive."
        required: true
        type: string
      FASTDEPLOY_WHEEL_URL:
        description: "URL of the FastDeploy Wheel."
        required: true
        type: string
      FASTDEPLOY_ROUTER_URL:
        description: "URL of the FastDeploy Router"
        required: true
        type: string
      CACHE_DIR:
        description: "Cache Dir Use"
        required: false
        type: string
        default: ""
      MODEL_CACHE_DIR:
        description: "Cache Dir Use"
        required: false
        type: string
        default: ""
    secrets:
      github-token:
        required: true

jobs:
  run_golang_router_tests:
    runs-on: [self-hosted, GPU-h20-2Cards]
    timeout-minutes: 30
    steps:
      - name: Code Prepare
        shell: bash
        env:
          docker_image: ${{ inputs.DOCKER_IMAGE }}
          fd_archive_url: ${{ inputs.FASTDEPLOY_ARCHIVE_URL }}
        run: |
          set -x
          REPO="https://github.com/${{ github.repository }}.git"
          FULL_REPO="${{ github.repository }}"
          REPO_NAME="${FULL_REPO##*/}"
          BASE_BRANCH="${{ github.base_ref }}"
          docker pull ${docker_image}
          # Clean the repository directory before starting
          docker run --rm --net=host -v $(pwd):/workspace -w /workspace \
          -e "REPO_NAME=${REPO_NAME}" \
          ${docker_image} /bin/bash -c '
            CLEAN_RETRIES=3
            CLEAN_COUNT=0

            while [ $CLEAN_COUNT -lt $CLEAN_RETRIES ]; do
              echo "Attempt $((CLEAN_COUNT+1)) to remove ${REPO_NAME}* ..."
              rm -rf "${REPO_NAME}"* || true
              sleep 2

              # Check if anything matching ${REPO_NAME}* still exists
              if ! ls "${REPO_NAME}"* >/dev/null 2>&1; then
                echo "All ${REPO_NAME}* removed successfully"
                break
              fi

              CLEAN_COUNT=$((CLEAN_COUNT + 1))
            done

            if ls "${REPO_NAME}"* >/dev/null 2>&1; then
              echo "ERROR: Failed to clean ${REPO_NAME}* after multiple attempts"
              ls -ld "${REPO_NAME}"*
              echo "Attempting force cleanup with find..."
              find /workspace -mindepth 1 -maxdepth 1 -name "${REPO_NAME}*" -type d -exec chmod -R u+rwx {} \; -exec rm -rf {} + 2>/dev/null || true
              if ls "${REPO_NAME}"* >/dev/null 2>&1; then
                echo "ERROR: Force cleanup still failed"
                exit 1
              else
                echo "Force cleanup succeeded"
              fi
            fi
          '

          wget -q --no-proxy ${fd_archive_url} || {
            echo "ERROR: Failed to download archive from ${fd_archive_url}"
            exit 1
          }

          tar --no-same-owner -xf FastDeploy.tar.gz || {
            echo "ERROR: Failed to extract archive"
            exit 1
          }

          rm -rf FastDeploy.tar.gz
          cd FastDeploy
          git config --global user.name "FastDeployCI"
          git config --global user.email "fastdeploy_ci@example.com"
          git log -n 3 --oneline

      - name: Run Golang_Router Tests
        shell: bash
        env:
          docker_image: ${{ inputs.DOCKER_IMAGE }}
          fd_wheel_url: ${{ inputs.FASTDEPLOY_WHEEL_URL }}
          fd_router_url: ${{ inputs.FASTDEPLOY_ROUTER_URL }}
          CACHE_DIR: ${{ inputs.CACHE_DIR }}
          BASE_REF: ${{ github.event.pull_request.base.ref }}
          MODEL_CACHE_DIR: ${{ inputs.MODEL_CACHE_DIR }}
          IS_PR: ${{ github.event_name == 'pull_request' }}
        run: |
          if [[ "$IS_PR" == "true" ]]; then
            echo "Running on PR"
          else
            echo "Not a PR"
          fi
          runner_name="${{ runner.name }}"
          CARD_ID=$(echo "${runner_name}" | awk -F'-' '{print $NF}')
          DEVICES=$(echo "$CARD_ID" | fold -w1 | paste -sd,)
          DEVICE_PORT=$(echo "$DEVICES" | cut -d',' -f1)

          FLASK_PORT=$((8068 + DEVICE_PORT * 100))
          FD_API_PORT=$((8088 + DEVICE_PORT * 100))
          FD_ENGINE_QUEUE_PORT=$((8058 + DEVICE_PORT * 100))
          FD_METRICS_PORT=$((8078 + DEVICE_PORT * 100))
          FD_CACHE_QUEUE_PORT=$((8098 + DEVICE_PORT * 100))
          FD_ROUTER_PORT=$((8048 + DEVICE_PORT * 100))
          FD_CONNECTOR_PORT=$((8038 + DEVICE_PORT * 100))
          FD_RDMA_PORT=$((8028 + DEVICE_PORT * 100))
          echo "Test ENV Parameter:"
          echo "========================================================="
          echo "FLASK_PORT=${FLASK_PORT}"
          echo "FD_API_PORT=${FD_API_PORT}"
          echo "FD_ENGINE_QUEUE_PORT=${FD_ENGINE_QUEUE_PORT}"
          echo "FD_METRICS_PORT=${FD_METRICS_PORT}"
          echo "FD_CACHE_QUEUE_PORT=${FD_CACHE_QUEUE_PORT}"
          echo "FD_ROUTER_PORT=${FD_ROUTER_PORT}"
          echo "FD_CONNECTOR_PORT=${FD_CONNECTOR_PORT}"
          echo "FD_RDMA_PORT=${FD_RDMA_PORT}"
          echo "DEVICES=${DEVICES}"
          echo "========================================================="

          CACHE_DIR="${CACHE_DIR:-$(dirname "$(dirname "${{ github.workspace }}")")}"
          echo "CACHE_DIR is set to ${CACHE_DIR}"
          if [ ! -f "${CACHE_DIR}/gitconfig" ]; then
            touch "${CACHE_DIR}/gitconfig"
          fi

          PORTS=($FLASK_PORT $FD_API_PORT $FD_ENGINE_QUEUE_PORT $FD_METRICS_PORT $FD_CACHE_QUEUE_PORT)
          LOG_FILE="./port_cleanup_$(date +%Y%m%d_%H%M%S).log"
          echo "==== LOG_FILE is ${LOG_FILE} ===="

          echo "==== PORT CLEAN BEFORE TASK RUN ====" | tee -a $LOG_FILE

          for port in "${PORTS[@]}"; do
              PIDS=$(lsof -t -i :$port || true)
              if [ -n "$PIDS" ]; then
                  echo "Port $port is occupied by PID(s): $PIDS" | tee -a $LOG_FILE
                  echo "$PIDS" | xargs -r kill -9
                  echo "Port $port cleared" | tee -a $LOG_FILE
              else
                  echo "Port $port is free" | tee -a $LOG_FILE
              fi
          done

          echo "==== PORT CLEAN COMPLETE ====" | tee -a $LOG_FILE

          echo "========================================================="
          echo "Ensuring no stale container named ${runner_name} ..."
          if [ "$(docker ps -a -q -f name=${runner_name})" ]; then
            echo "Removing stale container: ${runner_name}"
            docker rm -f ${runner_name} || true
          fi

          export RDMA_DEVICES=$(find /dev/infiniband/uverbs* -maxdepth 1 -not -type d | xargs -I{} echo '--device {}:{}')

          docker run --rm --net=host \
          --name ${runner_name} \
          --cap-add=SYS_PTRACE --cap-add=IPC_LOCK \
          --shm-size=64G \
          ${RDMA_DEVICES} \
          --device=/dev/infiniband/rdma_cm \
          --ulimit memlock=-1:-1 \
          -v $(pwd):/workspace -w /workspace \
          -v "${CACHE_DIR}/gitconfig:/etc/gitconfig:ro" \
          -v "${CACHE_DIR}/.cache:/root/.cache" \
          -v "${CACHE_DIR}/ConfigDir:/root/.config" \
          -v "${MODEL_CACHE_DIR}:/ModelData:ro" \
          -e "MODEL_PATH=/ModelData" \
          -e "FD_API_PORT=${FD_API_PORT}" \
          -e "FD_ENGINE_QUEUE_PORT=${FD_ENGINE_QUEUE_PORT}" \
          -e "FD_METRICS_PORT=${FD_METRICS_PORT}" \
          -e "FLASK_PORT=${FLASK_PORT}" \
          -e "FD_CACHE_QUEUE_PORT=${FD_CACHE_QUEUE_PORT}" \
          -e "FD_ROUTER_PORT=${FD_ROUTER_PORT}" \
          -e "FD_CONNECTOR_PORT=${FD_CONNECTOR_PORT}" \
          -e "FD_RDMA_PORT=${FD_RDMA_PORT}" \
          -e "CLEAN_CUDA=1" \
          -e TZ="Asia/Shanghai" \
          -e "fd_wheel_url=${fd_wheel_url}" \
          -e "fd_router_url=${fd_router_url}" \
          -e "BASE_REF=${BASE_REF}" \
          -e "IS_PR=${IS_PR}" \
          -e "no_proxy=localhost,127.0.0.1,0.0.0.0,bcebos.com,.bcebos.com,bj.bcebos.com,su.bcebos.com,paddle-ci.gz.bcebos.com,apiin.im.baidu.com,baidu-int.com,.baidu.com,aliyun.com,gitee.com,pypi.tuna.tsinghua.edu.cn,.tuna.tsinghua.edu.cn" \
          --gpus "\"device=${DEVICES}\"" ${docker_image} /bin/bash -c '

          git config --global --add safe.directory /workspace/FastDeploy
          cd FastDeploy

          python -m pip install --pre paddlepaddle-gpu -i https://www.paddlepaddle.org.cn/packages/nightly/cu126/
          pip config set global.extra-index-url https://mirrors.tuna.tsinghua.edu.cn/pypi/web/simple

          python -m pip install -r scripts/unittest_requirement.txt
          python -m pip install ${fd_wheel_url}
          rm -rf fastdeploy
          python -m pip install ${fd_wheel_url} --no-deps --target=/workspace/FastDeploy
          export PYTHONPATH=/workspace/FastDeploy/

          # download fd-router binary
          wget -q --no-proxy ${fd_router_url} -O /usr/local/bin/fd-router
          chmod +x /usr/local/bin/fd-router

          bash scripts/run_golang_router.sh
          '

      - name: Terminate and delete the container
        if: always()
        run: |
          set +e
          docker exec -t ${{ runner.name }} /bin/bash -c 'find /workspace -mindepth 1 -delete'
          docker rm -f ${{ runner.name }}