manage.py collectstatic: error: unrecognized arguments: --noinput in shell script launched by Docker

Question

I'm working to launch a django-tornado hybrid app in a Docker container from a shell script and and getting --noinput as an unrecognized argument for django commands:

usage: manage.py collectstatic [-h] [--version] [-v {0,1,2,3}]
                               [--settings SETTINGS] [--pythonpath PYTHONPATH]
                               [--traceback] [--no-color] [--noinput]
                               [--no-post-process] [-i PATTERN] [-n] [-c] [-l]
                               [--no-default-ignore]
manage.py collectstatic: error: unrecognized arguments: --noinput

Why would I be getting --noinput as an unrecognized argument? My Dockerfile calls a deployment shell script which performs the collectstatic and migrate commands (both with the --noinput argument, which is failing for both. I've played around with removing extraneous lines, adjusting whitespace around the command, etc, to no avail. I can run the shell script locally without any issues; it seems to only be a problem in the Docker container RUN call to the shell script.

Dockerfile:

FROM python:2.7

RUN apt-get update
RUN apt-get install -y build-essential
RUN apt-get install -y python-dev
RUN apt-get install -y libmysqlclient-dev

RUN mkdir /code
ADD . /code/
WORKDIR /code
RUN pip install -r requirements.txt

CMD ["sh","./deploy.sh"]

EXPOSE 8888

deploy.sh:

#!/bin/sh
python manage.py collectstatic --noinput
python manage.py migrate --noinput
python main.py

If I run the django commands in the Dockerfile with RUN python manage.py collectstatic --noinput there is no issue, but I'm trying to get the application-specific commands in the CMD call, since I need to have database environment vars from Elastic Beanstalk for my deployment environment.

Answer 1

Looks like my issue was line endings in the shell script. I think sh was feeding in --noinput\R into python, so it was presenting itself in the terminal as looking like --noinput, but really it was getting a CR character as well that it was matching against.

When I was locally testing, it was in the Docker Quickstart terminal (where it worked), and the Docker containers were always running in Ubuntu (where it was failing).

I've hit this way in the past before where different line endings in shell scripts that were written on Windows messed things up in a Linux environment, and I need to remember how important it is to set up line endings correctly in my editors...

Answer 2

I Solve this problem using notepad++

In your notepad++

Do Inspection
View > Show Symbol > Show All Characters
Look for CR LF line endings, rather than LF line endings!

Solving
Search > Replace (Ctrl+H)
Make sure Search Mode is set to Extended
Find \r\n and replace with \n

main problem in this case is there are CR aka carriage return in your sh file