Postgres Docker容器：启动时运行命令

Question

I have following Dockerfile to create a Postgres container with a third party app:

FROM postgres

COPY ./setup/third-party-app.deb /opt/third-party-app.deb
COPY ./setup/init_extensions-db.sql /docker-entrypoint-initdb.d/init_extensions-db.sql

RUN apt-get update -y
RUN apt-get install -y default-jre-headless
RUN dpkg -i /opt/third-party-app.deb
RUN service third-party-app init
RUN service third-party-app start

But I noticed that the service third-party-app start was successfully executed during the build phase, but than the service was not started in the running container.
I figured the reason is that Docker images don't snapshot the running process. So I am trying to find a way to automatically execute this command after the container has started.

I tried following code at the end of the Dockerfile:

CMD postgres && service third-party-app start

But the container exits with error status code 1.

I also tried to copy following shell script into the docker-entrypoint-initdb.d of the Postgres image:

COPY ./setup/third-party-app-start.sh /docker-entrypoint-initdb.d/third-party-app-start.sh

third-party-app-start.sh:

#!/bin/bash
set -e
service third-party-app start

But the same happens. The container exits with error status code 1.

But when I execute the command manually via PowerShell after the container has started it works perfectly fine.

Is there a better way to accomplish this?
Or is there a way to get a detailed error log of what went wrong with that command?

Answer 1

A container's main running process is the ENTRYPOINT and/or CMD at the end of the Dockerfile. It is generally recommended that you separate areas of concern by using one service per container. That service may fork into multiple processes (for example, Apache web server starts multiple worker processes). It's ok to have multiple processes, but to get the most benefit out of Docker, avoid one container being responsible for multiple aspects of your overall application. You can connect multiple containers using user-defined networks and shared volumes.

In order to achieve your goal, you will need to use a wrapper script or supervisor, more details: https://docs.docker.com/config/containers/multi-service_container/