Hello everyone,

Ive seem to be stuck for a couple of days trying to deploy LibreTranslate to Heroku using Docker.
Locally, i have no issue whatsoever, the docker image is working as intended.

The app is building successfully but when running "run.sh", i am getting the following error in heroku console:

2024-06-22T00:47:40.000000+00:00 app[api]: Build succeeded
2024-06-22T00:47:40.618736+00:00 app[api]: Release v22 created by user xxx@gmail.com
2024-06-22T00:47:40.618736+00:00 app[api]: Deploy 371a439d by user xxx@gmail.com
2024-06-22T00:47:41.793013+00:00 heroku[web.1]: State changed from crashed to starting
2024-06-22T00:48:04.419660+00:00 heroku[web.1]: Starting process with command /bin/sh -c /app/run.sh
2024-06-22T00:48:05.068267+00:00 app[web.1]: Starting LibreTranslate...
2024-06-22T00:48:05.068330+00:00 app[web.1]: PORT: 24559
2024-06-22T00:48:05.068330+00:00 app[web.1]: Using port: 24559
2024-06-22T00:49:04.460710+00:00 heroku[web.1]: Error R10 (Boot timeout) -> Web process failed to bind to $PORT within 60 seconds of launch
2024-06-22T00:49:04.472926+00:00 heroku[web.1]: Stopping process with SIGKILL
2024-06-22T00:49:04.561711+00:00 heroku[web.1]: Process exited with status 137
2024-06-22T00:49:04.585905+00:00 heroku[web.1]: State changed from starting to crashed

Here is my Dockerfile:

`
FROM python:3.11.9-slim-bullseye as builder

WORKDIR /app

ARG DEBIAN_FRONTEND=noninteractive
RUN apt-get update -qq
&& apt-get -qqq install --no-install-recommends -y pkg-config gcc g++
&& apt-get upgrade --assume-yes
&& apt-get clean
&& rm -rf /var/lib/apt

RUN python -m venv venv && ./venv/bin/pip install --no-cache-dir --upgrade pip

COPY . .

RUN ./venv/bin/pip install Babel==2.12.1 && ./venv/bin/python scripts/compile_locales.py
&& ./venv/bin/pip install torch==2.0.1 --extra-index-url https://download.pytorch.org/whl/cpu
&& ./venv/bin/pip install .
&& ./venv/bin/pip cache purge

FROM python:3.11.9-slim-bullseye

ARG with_models=false
ARG models=""

RUN addgroup --system --gid 1032 libretranslate && adduser --system --uid 1032 libretranslate && mkdir -p /home/libretranslate/.local && chown -R libretranslate:libretranslate /home/libretranslate/.local
USER libretranslate

COPY --from=builder --chown=1032:1032 /app /app
WORKDIR /app

COPY --from=builder --chown=1032:1032 /app/venv/bin/ltmanage /usr/bin/

RUN if [ "$with_models" = "true" ]; then \

if [ ! -z "$models" ]; then
./venv/bin/python scripts/install_models.py --load_only_lang_codes "$models";
else
./venv/bin/python scripts/install_models.py;
fi
fi

EXPOSE 5000

ENTRYPOINT ["/bin/bash", "/app/run.sh"]

`

And my run.sh:

set -eo pipefail
echo "Starting LibreTranslate..."
echo "PORT: ${PORT}"

PORT=${PORT:-5000}

echo "Using port: ${PORT}"

./venv/bin/libretranslate --host 0.0.0.0 --port $PORT &> /app/libretranslate.log