+++ +++

Run on Local/Docker service (Flask)

There are 2 options to run at local:

  • Deploy/Run directly from host machine: I use this option for debug/development
  • Deploy/Run on docker: I will build docker image from base image and source code.

Image

Run on local:

  • We can run by:
python -m flask run
  • Or by run directly app.py
python app.py

Build Docker image

  • Create Dockerfile

    FROM python:3.10.13-alpine3.17

WORKDIR /translation-backend

COPY requirements.txt requirements.txt

RUN pip3 install -r requirements.txt

COPY . .

ENV PORT 8080
# Expose port
EXPOSE 8080

# CMD [ "python3", "-m" , "flask", "run", "--host=0.0.0.0"]
CMD ["python3", "app.py"]

  • In the case if we want to build the docker image then we can use docker build. Go to the folder (. in command line) containing the Dockerfile then execute one of options

    docker build . --tag <IMAGE_NAME_CAN_HAVE_VERSION>
    # Tag name with format is URL (which can be used for Cloud/Dockerhub services to resolve information in URL
    # <IMAGE_URI> should be format <REGION>-docker.pkg.dev/<PROJECT-ID>/<REPOSITORY>/<IMAGE:TAG>
    docker build . --tag <IMAGE_URI>
    # Example
    docker build . --tag asia-east1-docker.pkg.dev/hohai-sample-cloud-run/translation-backend/translation-backend:tag1

    # Build with other options
    docker buildx build --platform linux/amd64 -t {project-name} .
    # Example
    docker buildx build --platform linux/amd64 -t translation-backend-image .

    ```

  • We can use other parties (like Goole CLI) to build and submit using gcloud builds

    gcloud builds submit --tag asia-east1-docker.pkg.dev/hohai-sample-cloud-run/translation-backend/translation-backend:tag1

Run Docker image

We can create docker-compose.yml to build and provision services

version: "3.8"

services:
  backend-app:
    container_name: translation-backend
    image: translation-backend-image
    build:
      context: .
      dockerfile: Dockerfile
    volumes:
      - ./app.py:/translation-backend/app.py
      - ./_config.py:/translation-backend/_config.py
    ports:
      - "5000:5000"