not able to access Postgres container in docker compose - postgresql

I have a docker-compose.yml with two services, a custom image (the service called code) and a Postgres server. Below I attach the Dockerfile to built the image called app of the first service and next the docker-compose.yml:
# Dockerfile of custom image
FROM ubuntu:latest
RUN apt-get update \
&& apt-get install -y python3-pip python3-dev \
&& cd /usr/local/bin \
&& ln -s /usr/bin/python3 python \
&& pip3 install --upgrade pip
WORKDIR /usr/app
COPY ./* ${PWD}/
ADD ./requirements.txt ./
RUN pip install -r requirements.txt
ADD ./ ./
# docker-compose.yml
version: '3.2'
services:
code:
image: app:latest
ports:
- 5001:5001
networks:
- docker-elk_elk
postgres:
image: postgres:9.5-alpine
environment:
POSTGRES_USER: postgres # define credentials
POSTGRES_PASSWORD: postgres # define credentials
POSTGRES_DB: postgres # define database
ports:
- 5432:5432 # Postgres port
networks:
- docker-elk_elk
networks:
docker-elk_elk:
external: true
Also here docker-elk_elk points to another network where another docker-compose stack runs and I want the above docker-compose to join too. However when I run docker-compose run code bash and obtain a shell in code service, a curl https://postgres:5432 gives the following message : curl: (35) OpenSSL SSL_connect: SSL_ERROR_SYSCALL in connection to postgres:5432. I've tried also curl http://postgres:5432 which returned curl: (52) Empty reply from server. Furthermore, the docker-elk_elk network (clearly created by elasticsearch-logtash-kibaba stack) when doing docker network ls gives
NETWORK ID NAME DRIVER SCOPE
8a54fe394fe8 docker-elk_elk bridge local
I'm really lost and confused, can someone help me out? If there is any piece of info that might be necessary or helpful and wasn't included above let me know please.
I forgot to mention that app is just a simple python application (not a web app or other python sophisticated libraries).
P.S. Something that perhaps I should have mentioned above. What I want to do, is using the ubuntu container with the app inside to query (and send data) both to the postgres and Elasticsearch (which is in the other docker-compose stack) db.

Related

Docker Volume Data is not Persistent

I want to create two Docker volumes and have their data be persistent. I run sudo docker compose up -d, post some data to my website (text that is stores in a sqlite database and an image stored in the filesystem), then run sudo docker compose down. When I run sudo docker compose up -d again, all the data I posted is gone. With the following configs, I expect the data to still be present.
Dockerfile:
FROM python:3.9.16-buster
RUN pip install --upgrade pip
# The Debian/Buster default is to disable the password.
RUN adduser nonroot
RUN mkdir /home/site/ && chown -R nonroot:nonroot /home/site
RUN chown -R nonroot:nonroot /var/log/site
# two volumes created
VOLUME /home/site/db /home/site/static
WORKDIR /home/site
USER nonroot
# folders ./site/static and ./site/db exist in my host directory
COPY --chown=nonroot:nonroot . .
CMD ["python", "./site/main.py"]
compose.yaml:
services:
site:
build: flask
restart: always
ports:
- '8081:8081'
volumes:
- site_db:/home/site/db # same path as the volumes created in the Dockerfile
- site_static:/home/site/static
command: gunicorn -w 1 -t 3 -b 0.0.0.0:8081 --chdir ./site main:app
volumes:
site_db: # I find it odd these volumes keys don't have values, but that's what I have see other people do
site_static:
docker compose up and docker compose down delete my volumes.
docker compose start and docker compose stop do NOT delete my volumes.
Through the Flask app, check where you are uploading the files to, as well as where the sqlite3 db file is. If these paths do not align with the volumes paths, data will not persist.

Running psql command from a docker container

I'm having trouble connecting postgresDB to my app using docker-compose
My understanding of docker compose is that I can combine two containers so that they can communicate with each other. Suppose I have an app in the appcontainer that runs the psql command (just a one liner python script with os.command("psql")). Since the app container does not have postgres installed, it won't be able to run psql by itself. However, I thought combining two containers in docker-compose.yml would let me run psql but apparently not.
What am I missing here?
I am using 2 postgres images because I'm trying to find regression bugs between two dbms
version: "3"
services:
app:
image: "app:1.0"
depends_on:
- postgres9
- postgres12
ports:
- 8080:80
postgres9:
image: postgres:9.6
environment:
POSTGRES_PASSWORD: mysecretpassword
POSTGRES_USER: postgres
POSTGRES_DB: test_bd
ports:
- '5432:5432'
postgres12:
image: postgres:12
environment:
POSTGRES_PASSWORD: mysecretpassword
POSTGRES_USER: postgres
POSTGRES_DB: test_bd
ports:
- '5435:5435'
Each Docker container has a self-contained filesystem. You can never directly run commands from the host or from other containers' filesystems; anything you want to run needs to be installed in the container (really, in its image's Dockerfile).
If you want to run a tool like psql, it needs to be installed in your image. You don't say what your base image is, but if it's based on Debian or Ubuntu, you need to install the postgresql-client package:
RUN apt-get update \
&& DEBIAN_FRONTEND=noninteractive \
apt-get install --no-install-recommends --assume-yes \
postgresql-client
The right approach here is to add a standard Python PostgreSQL client library, like psycopg2, to your project's Python Pipfile, setup.py, and/or requirements.txt, and use that library instead of shelling out to psql. You will also need the PostgreSQL C library header files to install that package; instead of postgresql-client, install the Debian libpq-dev package.
In your case, those two containers with postgres instance in each, are running on different hosts (other than host with app). What you need is to specify correct host in psql command. It might look like (for postgres12 container):
PGPASSWORD="mysecretpassword" psql -h postgres12 -d test_bd -U postgres

Having some issues with docker-compose

Im pretty new to flask , postgres and docker. I am trying to dockerize our application so that a new dev does not have to worry about installing python depenedencies to start developing. I currently have two containers, one for the flask app and one for the postgres db.
The issue I am having is that when I put the two in a compose they cannot connect to each other. I have defined my own network to ensure that they are running on the same one but that still does not seem to solve the problem. When I run the flask app outside of a container and try to connect to the containerized postgres db I am having no trouble connecting and it works fine.
Any help would be much appreciated
Thanks in advance
Dockerfile for the flask app
FROM python:3.7-alpine3.7
LABEL maintainer="uwblueprint"
LABEL org.label-schema.schema-version="1.0"
LABEL org.label-schema.name="elevate-api"
LABEL org.label-schema.vcs-url="https://gitlab.com/uwblueprint/elevate-api/"
LABEL org.label-schema.vendor="UW Blueprint"
## Copy source and environment-config files.
COPY app/ ./app/
WORKDIR /app
COPY Pipfile Pipfile.lock ./
## Install external dependencies.
RUN apk add --no-cache libpq
## Install application dependencies.
RUN apk add --no-cache --virtual build-deps \
gcc musl-dev postgresql-dev libffi-dev && \
pip3 install --upgrade pip pipenv gunicorn && \
pipenv --python 3.7 && \
pipenv install --system --deploy && \
apk del build-deps
EXPOSE 5000
CMD ["flask", "run"]
Docker Compose File
version: '3.7'
services:
database:
image: registry.gitlab.com/uwblueprint/elevate-api/postgres:latest
build:
context: ./external/postgres
cache_from:
- registry.gitlab.com/uwblueprint/elevate-api/postgres:latest
- postgres:10.5-alpine
container_name: database
env_file: ./external/postgres/configs/.env
volumes:
- postgres.data:/var/lib/postgresql/data # persist data
ports:
- "5432:5432"
networks:
- "api_net"
api:
build: .
ports:
- "5000:5000"
networks:
- "api_net"
networks:
api_net:
volumes:
postgres.data:
Python SQL Alchemy Code
import os
from flask_sqlalchemy import SQLAlchemy
from app import app
# Make sure that DB_PASS is an environment variable:
##if 'DB_PASS' not in os.environ:
## raise EnvironmentError("Could not find environment variable 'DB_PASS',")
##user = os.environ.get("DB_USER", "robot")
user ="fakeUser"
password = "fakePassword"
##password = os.environ["DB_PASS"]
host = os.environ.get("DB_HOST", "0.0.0.0")
port = os.environ.get("DB_PORT", "5432")
name = os.environ.get("DB_NAME", "elevate")
# Configure flask-sqlalchemy.
app.config["SQLALCHEMY_DATABASE_URI"] = "postgresql://%s:%s#%s:%s/%s" % (
user, password, host, port, name
)
app.config["SQLALCHEMY_TRACK_MODIFICATIONS"] = False
# TODO: Find the optimal pool recycle time for PostgreSQL.
app.config["SQLALCHEMY_POOL_RECYCLE"] = 7200 # in milliseconds
# The db object is aware of the Flask application lifecycle, and will do things
# like close the database session when a Flask request ends.
#
# This saves a lot of headaches compared to the plain SQLAlchemy library, where
# I kept running into issues about database-session related objects trying
# to access the session when I had closed it prematurely (due to SQLAlchemy's
# lazy-loading data acess methods).
db = SQLAlchemy(app)
Set DB_HOST to "database" which is the name of the container.
Inside docker-compose you can make connection between services by service names. You can read about this in documentation.
By default Compose sets up a single network for your app. Each container for a service joins the default network and is both reachable by other containers on that network, and discoverable by them at a hostname identical to the container name.
Set host or DB_HOST environ to database. 127.0.0.1 won't work.

docker-compose succeed but server does not response when request

I have built a RESTful API web service using Flask framework, Redis as main database, MongoDB as a backup store and Celery as task queue to store data into MongoDB in background
Then I dockerize my application using docker-compose. Here is my docker-compose.yml:
version: '3'
services:
web:
build: .
ports:
- "5000:5000"
volumes:
- .:/app
redis:
image: "redis:alpine"
ports:
- "6379:6379"
mongo:
image: "mongo:3.6.5"
ports:
- "27017:27017"
environment:
MONGO_INITDB_DATABASE: syncapp
Here is my Dockerfile:
# base image
FROM python:3.5-alpine
MAINTAINER xhoix <145giakhang#gmail.com>
# copy just the requirements.txt first to leverage Docker cache
# install all dependencies for Python app
COPY ./requirements.txt /app/requirements.txt
WORKDIR /app
# install dependencies in requirements.txt
RUN pip install -r requirements.txt
# copy all content to work directory /app
COPY . /app
# specify the port number the container should expose
EXPOSE 5000
# run the application
CMD ["python", "/app/app.py"]
After run command docker-compose up, the app server, Redis and Mongo server just run well. But when I use Postman or curl to call the API, for example http://127.0.0.1:5000/sync/api/v1.0/users, which should return JSON format of all users, but the result is Could not get any response: There was an error connecting to http://127.0.0.1:5000/sync/api/v1.0/users.
I have no idea why this happens.
Thanks for any help and suggestion!
I found the cause of the issue:
After an hour debug, it turns out that I only need to change the app host to 0.0.0.0. Maybe when mapping port, docker default will be 0.0.0.0, since when I run command docker-compose ps, the PORTS column of each container has format 0.0.0.0:<port> -> <port>. I don't know this is the cause of the issue or not, but I did it and the problem is solved
If operating system Linux then use :
ifconfig -a
If operating system Windows then use :
ipconfig /all
Then check the interface like docker or something with virtualization, and use the ipv4 or inet
Or Just use the docker command:
docker network inspect bridge
Then use the gateway ip on IPAM

Networking using docker-compose in docker executor in circleci

this is a circleci question I guess.
I am quite happy with circleci but now I ran into a problem and I don't know what I'm doing wrong.
Maybe this is something very easy, but I don't see the it.
In short
I can't make containers talk to each other on circleci.
Problem
Basically what I wanted to do is start a server container and a client container, and then let them talk to each other.
I created a minimal example here: https://github.com/mRcSchwering/circleci-integration-test
The README.md basically explains the desired outcome.
I have a .circleci/config.yml like this:
version: 2
jobs:
build:
docker:
- image: docker:18.03.0-ce-git
steps:
- checkout
- setup_remote_docker
- run:
name: Install docker-compose
command: |
apk --update add py2-pip
/usr/bin/pip2 install docker-compose
docker-compose --version
- run:
name: Start Container
command: |
docker-compose up -d
docker-compose ps
- run:
name: Let client talk to server
command: |
docker-compose run client psql -h server -p 5432 -U postgres -c "\l"
In a docker container, docker-compose is installed, which is then used to start a server and a client (postgres here). In the last step I am telling the client to query the server. However, it cannot find the server:
#!/bin/sh -eo pipefail
docker-compose run client psql -h server -p 5432 -U postgres -c "\l"
Starting project_server_1 ...
^#^#psql: could not connect to server: Connection refused
Is the server running on host "server" (172.18.0.2) and accepting
TCP/IP connections on port 5432?
Exited with code 2
Files
The docker-compose.yml looks like this
version: '2'
services:
server:
image: postgres:9.5.12-alpine
networks:
- internal
expose:
- '5432'
client:
build:
context: .
networks:
- internal
depends_on:
- server
networks:
internal:
driver: bridge
where the client is built from a dockerfile like this
FROM alpine:3.7
RUN apk --no-cache add postgresql-client && rm -rf /var/cache/apk/*
Note
If I repeat everything on my Linux (also with docker-in-docker) it works.
But I guess some things work completely different on circleci.
I found some people mentioning that on circleci networking and bind mounts can be tricky but I didn't find anything that can help me.
There is this doc but I thought I am doing this already.
Then there is this project where someone seems to do the same thing on circleci successfully.
But I cannot figure out what's different there...
Anyway I would really appreciate your help. So far I have given up on this.
Best
Marc
Ok, in the meanwhile I (no actually it was halfer from the circleci forum) noticed that docker-compose run client psql -h server -p 5432 -U postgres -c "\l" was run before the server was up and running. A simple sleep 5 after docker-compose up -d fixes the problem.