standard_init_linux.go:211: exec user process caused "no such file or directory"?

Question

Dockerfile

FROM python:3.7.4-alpine

ENV PYTHONUNBUFFERED 1
ENV PYTHONDONTWRITEBYTECODE 1
ENV LANG C.UTF-8
MAINTAINER "mail@gmail.com"


RUN apk update && apk add postgresql-dev gcc musl-dev
RUN apk --update add build-base jpeg-dev zlib-dev
RUN pip install --upgrade setuptools pip

RUN mkdir /code
WORKDIR /code

COPY requirements.txt /code/
RUN pip install -r requirements.txt
COPY . /code/

#CMD ["gunicorn", "--log-level=DEBUG", "--timeout 90", "--bind", "0.0.0.0:8000", "express_proj.wsgi:application"]
ENTRYPOINT ["./docker-entrypoint.sh"]

docker-entrypoint.sh

#!/bin/bash

# Prepare log files and start outputting logs to stdout
touch /code/gunicorn.log
touch /code/access.log
tail -n 0 -f /code/*.log &

# Start Gunicorn processes
echo Starting Gunicorn.
exec gunicorn express_proj.wsgi:application \
    --name express \
    --bind 0.0.0.0:8000 \
    --log-level=info \
    --log-file=/code/gunicorn.log \
    --access-logfile=/code/access.log \
    --workers 2 \
    --timeout 90 \
    "$@"

Getting Error standard_init_linux.go:211: exec user process caused "no such file or directory" Need help. Some saying to use dos2unix(i do not know hoe to use it.)

Does your image actually contain GNU bash? (Alpine images typically don't.) I don't think your script uses any bash-specific features so the POSIX shell in Alpine should work fine; does changing the shebang line to `#!/bin/sh` work? — David Maze, Apr 20 '20 at 17:56

score 18 · Accepted Answer · answered Apr 20 '20 at 20:02

18

The "shebang" line at the start of a script says what interpreter to use to run it. In your case, your script has specified #!/bin/bash, but Alpine-based Docker images don't typically include GNU bash; instead, they have a more minimal /bin/sh that includes just the functionality in the POSIX shell specification.

Your script isn't using any of the non-standard bash extensions, so you can just change the start of the script to

#!/bin/sh

answered Apr 20 '20 at 20:02

David Maze

130,717
29
175
215

After changing it, am getting *[2020-04-21 06:18:28 +0000] [10] [INFO] Booting worker with pid: 10*. What does it mean? I cann't solution for it anywhere. – Nurbek Batyrzhan uulu Apr 21 '20 at 06:56
That seems like normal output from `gunicorn` and I don't think you need to do anything about it. Just start making requests. – David Maze Apr 21 '20 at 09:43

score 4 · Answer 2 · answered Jul 14 '20 at 11:18

4

This can also happen if the line endings of the script is wrong, i.e. \r\n instead of \r

this can be checked using the file path/to/script.sh command which tells if the script has CR-LF line endings

If its a one off script dos2unix can be used to change it to \r\n to \n

If its a git repository setting the autocrlf option to input would work

How to change line-ending settings

answered Jul 14 '20 at 11:18

tejas

1,795
1
16
34

Thank you! Saved me days of clueless researches!! – It's a pretty uncommon solution but that was my tricky-to-diagnose problem. My project worked on Linux and MacOS, but **was failing in running the shel scripts, when cloned and rebuilt on Windows (WSL)**. Adding `*.sh eol=lf` and `*.bat eol=crlf` into a newly create `.gitattributes` (committed to the repository), solved the issue on the Windows machine after a `git pull --rebase` to get the attributes rules (hence fixing the line-ending when dealing with `git-checkout` of shell scripts). – Kamafeather Oct 26 '20 at 14:58

standard_init_linux.go:211: exec user process caused "no such file or directory"?

2 Answers2

Linked