[Python] connection is reestablished when i dockerize a FlaskApp and accessing by my localhost

I created a python flaskapp. everything works well so I decided to Dockerize it.

But when I launch my app with docker and try to access to http://localhost:5000/ I get 'The connection has been reset'.

this is what my docker-compose and my Dockfile look like.

docker-compose :


version: '3.8'

services:
elasticsearch:
container_name: elasticsearch
image: elasticsearch:8.15.2
ports:
- "9200:9200"
- "9300:9300"
environment:
- discovery.type=single-node
- xpack.security.enabled=false
healthcheck:
test: ["CMD", "curl", "-f", "http://elasticsearch:9200"]
interval: 30s
timeout: 10s
retries: 5
networks:
- mynetwork

flaskapp:
container_name: flaskapp
build: .
ports:
- "5000:5000"
depends_on:
elasticsearch:
condition: service_healthy
networks:
- mynetwork

networks:
mynetwork:
driver: bridge


Dockerfile :


FROM python:3.12

WORKDIR /IoT

COPY Pipfile Pipfile.lock /IoT/

RUN pip install pipenv && pipenv install --deploy --ignore-pipfile

COPY . /IoT/

EXPOSE 5000

CMD ["pipenv", "run", "python", "app.py"]


here is my command which launches docker compose:

• docker-compose up -d

Python files that matter:


# app/__init__.py

from flask import Flask
from .routes import main_blueprint
from elasticsearch import Elasticsearch


def create_app():
app = Flask(__name__)

app.elasticsearch = Elasticsearch(["http://elasticsearch:9200"])

app.register_blueprint(main_blueprint)

return app


# app.py

from app import create_app

app = create_app()

if __name__ == "__main__":
app.run(host="0.0.0.0", port=5000)




when I try from my docker container with the curl command http://localhost:5000 it returns my template index.html, so it works.

so I tried to connect with the IP of my container and here's how I got it: docker inspect -f "{{range .NetworkSettings.Networks}}{{.IPAddress}}{{end}}" flaskapp

but it still doesn't work.

I understand that my appflask works correctly within my docker but I don't understand why I can't access it via my pc through my localhost.

thanks for your time and your help !

Continue reading...