Incremental caching using S3 compatibale object storage of Linode

name: Save and Restore

on:
  push:
  pull_request:
  workflow_dispatch:

jobs:
  build:
    runs-on: ubuntu-latest
    name: Build
    steps:
      - uses: actions/checkout@v6

      - name: prepare
        run: |
          sudo apt-get update
          sudo apt-get install -y s3cmd
          echo "[default]" > s3cfg.ini
          echo "host_base = us-southeast-1.linodeobjects.com" >> s3cfg.ini
          echo "host_bucket = us-southeast-1.linodeobjects.com" >> s3cfg.ini
          echo "access_key = ${{ secrets.ACCESS_KEY }}" >> s3cfg.ini
          echo "secret_key = ${{ secrets.SECRET_KEY }}" >> s3cfg.ini

      #- name: listing
      #  run: |
      #    echo restore
      #    s3cmd --config=s3cfg.ini ls s3://diggers/

      # I had to put in an external script as GitHub runs the shell commands using the -e flag
      # meaning if any of the commands fails then the whole step fails.
      # The resote will fail the first time it run as there is still nothing to restore.
      # So now we sweep in under the carpet.
      # In a better solution we would look at the exit code and have a configuration option to
      # disregard this failure.
      - name: restore
        run: ./restore.sh

      - name: Pretend to add more data...
        run: |
          mkdir -p data
          date >> data/time.txt
          cat data/time.txt

      - name: save
        run: |
          echo save
          s3cmd --config=s3cfg.ini put data/time.txt s3://diggers/

• Incremental caching demo using S3 compatibale object storage of Linode