django docker exited with code 0 in request GET or POST - django

Question

Good friends, I am developing an application in django 1.11 with docker on windows, recently update the git repository of the project and also made some changes with docker containers.
The problem is that when entering the main page and some other URLS nothing happens, but when I try to login to the administrator, the django container is closed and I do not even get any error by the browser, console or log .
Example:
When I come in here they are fine
GET / 200 OK
POST / 403 Forbidden
GET / api / auth / 405 Method not allowed
But when I enter these, without showing any message, close the docker container (proyect_django_1 exited with code 0)
GET / admin No answer
POST / api / auth / No answer
My docker-compose
version: '3'
services:
db:
build: docker/postgres
volumes:
- ./docker/data/postgres:/var/lib/postgresql/data
environment:
- POSTGRES_PASSWORD=postgres
- POSTGRES_USER=postgres
- POSTGRES_DB=project
redis:
image: redis:3.2-alpine
volumes:
- ./docker/data/redis:/data
rabbit:
image: rabbitmq:3-management-alpine
environment:
- RABBITMQ_DEFAULT_USER=admin
- RABBITMQ_DEFAULT_PASS=admin
django:
build:
context: .
args:
- REQUIREMENTS=development.txt
command: python3.6 manage.py runserver 0.0.0.0:8008
volumes:
- ./:/code
working_dir: /code/project
env_file: ./docker/DevelopmentEnv
ports:
- "8008:8008"
links:
- db
- rabbit
- redis
depends_on:
- db
celeryworker:
build:
context: .
args:
- REQUIREMENTS=development.txt
working_dir: /code/project
volumes:
- ./:/code
env_file: ./docker/DevelopmentEnv
links:
- db
- rabbit
command: celery -A config worker -l INFO -Q celery
frontend:
image: node:8.4-alpine
volumes:
- ./:/code
working_dir: /code/frontend
command: ash -c "yarn install --no-bin-links && yarn run build"
socketio:
image: node:8.4-alpine
volumes:
- ./:/code
working_dir: /code/sockets
command: ash -c "yarn install --no-bin-links && yarn start"
ports:
- "3000:3000"
links:
- redis
- django
depends_on:
- redis
My dockerfile
FROM python:3.6.2-alpine3.6
ARG REQUIREMENTS
RUN apk update
RUN apk add postgresql-dev postgresql-client
RUN apk add libffi-dev gcc
RUN apk add musl-dev zlib-dev jpeg-dev
RUN apk add --no-cache --virtual .build-deps-testing \
--repository http://dl-cdn.alpinelinux.org/alpine/edge/testing \
gdal-dev
RUN mkdir /code
ADD ./ /code/
WORKDIR /code
RUN pip3.6 install -r requirements/$REQUIREMENTS
WORKDIR /code/project

Answer 1

You could add restart: always to your django service definition. This will start a new django container if the previous one exits for any reason.
You should be getting some logs about why the process is exiting. Try running docker inspect <container-name> to see if there are any clues about why the process exits. There is probably a bug in your Python code triggered by some URLs, and it causes the process to exit.