Docker 图像未运行，数据库错误：django.db.utils.OperationalError 无法将主机名“postgres”转换为地址：名称或服务未知

Question

So I've managed to build my Docker image locally using docker-compose build and I've pushed the image to my Docker Hub repository.

Now I'm trying to get it working on DigitalOcean so I can host it. I've pulled the correct version of the image and I am trying to run it with the following command:

root@my-droplet:~# docker run --rm -it -p 8000:8000/tcp mycommand/myapp:1.1 (yes, 1.1)

However I soon run into these two errors:

 ... File "/usr/local/lib/python3.8/dist-packages/django/db/backends/postgresql/base.py", line 185, in get_new_connection connection = Database.connect(**conn_params) File "/usr/local/lib/python3.8/dist-packages/psycopg2/__init__.py", line 127, in connect conn = _connect(dsn, connection_factory=connection_factory, **kwasync) psycopg2.OperationalError: could not translate host name "postgres" to address: Name or service not known ```

 ... File "/usr/local/lib/python3.8/dist-packages/django/db/backends/base/base.py", line 197, in connect self.connection = self.get_new_connection(conn_params) File "/usr/local/lib/python3.8/dist-packages/django/utils/asyncio.py", line 26, in inner return func(*args, **kwargs) File "/usr/local/lib/python3.8/dist-packages/django/db/backends/postgresql/base.py", line 185, in get_new_connection connection = Database.connect(**conn_params) File "/usr/local/lib/python3.8/dist-packages/psycopg2/__init__.py", line 127, in connect conn = _connect(dsn, connection_factory=connection_factory, **kwasync) django.db.utils.OperationalError: could not translate host name "postgres" to address: Name or service not known ```

This may be due to how I have divided my application (using docker-compose,yml ) into two services and have only pushed the image of the app since my previous post . Here is my docker-compose.yml file:

Here is my docker-compose.yml file:

version: '3'

services:
  db:
    image: postgres
    environment:
      - POSTGRES_DB=postgres
      - POSTGRES_USER=(adminname)
      - POSTGRES_PASSWORD=(adminpassword)
      - CLOUDINARY_URL=(cloudinarykey)
  app:
    build: .
    ports:
      - "8000:8000"
    depends_on:
      - db

Here is my Dockerfile:

FROM (MY FRIENDS ACCOUNT)/django-npm:latest

RUN mkdir usr/src/mprova

WORKDIR /usr/src/mprova

COPY frontend ./frontend
COPY backend ./backend

WORKDIR /usr/src/mprova/frontend

RUN npm install
RUN npm run build

WORKDIR /usr/src/mprova/backend

ENV DJANGO_PRODUCTION=True
RUN pip3 install -r requirements.txt

EXPOSE 8000

CMD python3 manage.py collectstatic && \
    python3 manage.py makemigrations && \
    python3 manage.py migrate && \
    gunicorn mellon.wsgi --bind 0.0.0.0:8000

and here is a snippet of my settings.py file:

...
DATABASES = {
    'default': {
        'ENGINE': 'django.db.backends.postgresql',
        'NAME': 'postgres',
        'USER': 'admin',
        'PASSWORD': '[THE PASSWORD]',
        'HOST': 'db',
        'PORT': 5432,
    }
}
...

How can I fix this issue? I've looked around but can't see what I'm doing wrong.

---

From what I saw online, I added the following to my project, but it doesn't work either:

Here is my base.py (new file):

import psycopg2

conn = psycopg2.connect("dbname='postgres' user='admin' host='db' password='[THE PASSWORD]'")

Additions to Dockerfile :

FROM (MY FRIENDS ACCOUNT)/django-npm:latest

RUN pip3 install psycopg2-binary

COPY base.py base.py

RUN python3 base.py

...

Yet the build fails due to this error:

 Traceback (most recent call last): File "base.py", line 3, in <module> conn = psycopg2.connect("dbname='postgres' user='admin' host='db' password='[THE PASSWORD]'") File "/usr/local/lib/python3.8/dist-packages/psycopg2/__init__.py", line 127, in connect conn = _connect(dsn, connection_factory=connection_factory, **kwasync) psycopg2.OperationalError: could not connect to server: Connection refused Is the server running on host "db" ([SOME-IP-ADDRESS]) and accepting TCP/IP connections on port 5432? ERROR: Service 'app' failed to build: The command '/bin/sh -c python3 base.py' returned a non-zero code: 1 me@My-MacBook-Pro-5 mellon %

I'm unsure as to what to try next but I feel like I am making a simple mistake. Why is the connection being refused at the moment?

Answer 1

When you launch a docker-compose.yml file, Compose will automatically create a Docker.network for you and attach containers to it. Absent any specific networks: declarations in the file, Compose will name the.network default , and then most things Compose creates are prefixed with the current directory name. This setup is described in more detail in Networking in Compose .

docker run doesn't look at the Compose settings at all, so anything you manually do with docker run needs to replicate the settings docker-compose.yml has or would automatically generate. To make a docker run container connect to a Compose.network, you need to do something like:

docker network ls              # looking for somename_default
docker run \
  --rm -it -p 8000:8000/tcp \
  --net somename_default \     # <-- add this option
  mycommand/myapp:1.1

(The specific error message could not translate host name "postgres" to address hints at a Docker.networking setup issue: the database client container isn't on the same.network as the database server and so the hostname resolution fails. Contrast with a "connection refused" type error, which would suggest one container is finding the other but either the database isn't running yet or the port number is wrong.)

In your docker build example, code in a Dockerfile can never connect to a database. At a mechanical level, there's no way to attach it to a Docker.network as we did with docker run . At a conceptual level, building an image produces a reusable artifact; if I docker push the image to a registry and docker run it on a different host, that won't have the database setup, or similarly if I delete and recreate the database container locally.

Answer 2

I am trying to run it with the following command: root@my-droplet:~#

docker run --rm -it -p 8000:8000/tcp mycommand/myapp:1.1 (yes, 1.1)

Docker-compose allows to create a composition of several containers.
Running the django container with docker run defeats its purpose.

What you want is running:

docker-compose build

docker-compose up

or in a single command:

docker-compose up --build                    

Why is the connection being refused at the moment?

Maybe because that is not done at the right moment.

I am not a Django specialist but your second try looks ok while your django service misses the Dockerfile attribute in the compose declaration (typing error I assume).

For your first try, that error:

django.db.utils.OperationalError: could not translate host name "postgres" to address: Name or service not known ```

means that django uses "postgres" as host of the database.
But according to your compose, the database could be reached with a host named "db". These don't match.

Regarding your second try that defines database properties, it looks fine and it also matches to what the official doc states :

In this section, you set up the database connection for Django.

In your project directory, edit the composeexample/settings.py file.

Replace the DATABASES =... with the following:

settings.py

 DATABASES = { 'default': { 'ENGINE': 'django.db.backends.postgresql', 'NAME': 'postgres', 'USER': 'postgres', 'PASSWORD': 'postgres', 'HOST': 'db', 'PORT': 5432, } } These settings are determined by the postgres Docker image specified in docker-compose.yml.

For your second try, that error:

 conn = _connect(dsn, connection_factory=connection_factory, **kwasync) psycopg2.OperationalError: could not connect to server: Connection refused
        Is the server running on host "db" ([SOME-IP-ADDRESS]) and accepting
        TCP/IP connections on port 5432?

may mean that the Postgres DB listener is not ready yet when Django tries to init the connection with.

That attributes of your compose: depends_on: defines order of container start but doesn't wait for that the application behind the container be "ready".

Here two way to addresse that kind of scenario:

• either add a sleep of some seconds before starting your django service
• or set your django compose conf to restart on failure.

For the first way, configure your django service with a CMD such as (not tried):

   app:
     build: .
     ports:
       - "8000:8000"
     depends_on:
       - db
     command: /bin/bash -c "sleep 30; python3 manage.py collectstatic && python3 manage.py makemigrations && python3 manage.py migrate && gunicorn mellon.wsgi --bind 0.0.0.0:8000 "

For the second way, try something like that:

  app:
    build: .
    ports:
      - "8000:8000"
    depends_on:
      - db
    restart: on-failure