Category: Docker – ServerOK

Docker delete container after running

January 1, 2020

by ServerOk with No Comment Docker

When you run a docker image, it create a container and run it. After docker container stop, container stays, so you can use it again if required.

Some times, you need to delete docker container after it is run. This is useful if your docker container is just a command line executable. In my case, docker container have ffmpeg in it, i don’t want container left over after i executed ffmpeg command.

To auto delete container after execution, use –rm option.

docker run --rm DOCKER_IMAGE

1	docker run --rm DOCKER_IMAGE

Example

docker run --rm --name ffmpeg jrottenberg/ffmpeg:3.2-ubuntu -format

1	docker run --rm --name ffmpeg jrottenberg/ffmpeg:3.2-ubuntu -format

Run PostgreSQL in Docker

June 30, 2019

by ServerOk with No Comment Docker

To run PostgreSQL on docker, create a directory for saving the data presistant

mkdir -p /opt/my-postgresql

1	mkdir -p /opt/my-postgresql

run docker container

docker run --name my-postgresql \
    -p 5432:5432 \
    -e POSTGRES_PASSWORD=serverok123 \
    -e POSTGRES_USER=serverok \
    -e POSTGRES_DB=serverok \
    -v /opt/my-postgresql:/var/lib/postgresql/data \
    -d postgres

docker run --name my-postgresql \

-p 5432:5432 \

-e POSTGRES_PASSWORD=serverok123 \

-e POSTGRES_USER=serverok \

-e POSTGRES_DB=serverok \

-v /opt/my-postgresql:/var/lib/postgresql/data \

-d postgres

In above, change the value for POSTGRES_DB, POSTGRES_USER and POSTGRES_PASSWORD as needed.

Connect to PostgreSQL server

To connect to PostgreSQL server, run

docker container exec -ti my-postgresql bash

1	docker container exec -ti my-postgresql bash

Now you are inside PostgreSQL conainer, to login, run

psql -U serverok -W

1	psql -U serverok -W

It will ask for password. Once you enter password, you will be in PostgreSQL command line.

See docker

Docker compose start container on boot

March 3, 2019

by ServerOk with No Comment Docker

I have a docker container, that i need to start on server boot.

The docker-compose.yml file i used to create this docker container is

[root@cp03 serverok-rubycms]# cat docker-compose.yml 
version: '3'
services:
  web:
    image: serverok/rubycms:1.0
    command: bash -c "rm -f tmp/pids/server.pid && cd /myapp && source /etc/profile.d/rvm.sh && script/server -e production"
    volumes:
      - /home/naiwnyc/cms:/myapp
    ports:
      - "3000:3000"
[root@cp03 serverok-rubycms]#

[root@cp03 serverok-rubycms]# cat docker-compose.yml

version: '3'

services:

web:

image: serverok/rubycms:1.0

command: bash -c "rm -f tmp/pids/server.pid && cd /myapp && source /etc/profile.d/rvm.sh && script/server -e production"

volumes:

- /home/naiwnyc/cms:/myapp

ports:

- "3000:3000"

[root@cp03 serverok-rubycms]#

With above docker-compose.yml file. i have to start docker container manually after server reboot.

To make it auto start, add the line

    restart: always

1	restart: always

Here is the modified docker-compose.yml file.

[root@cp03 serverok-rubycms]# cat docker-compose.yml 
version: '3'
services:
  web:
    image: serverok/rubycms:1.0
    restart: always
    command: bash -c "rm -f tmp/pids/server.pid && cd /myapp && source /etc/profile.d/rvm.sh && script/server -e production"
    volumes:
      - /home/naiwnyc/cms:/myapp
    ports:
      - "3000:3000"
[root@cp03 serverok-rubycms]#

[root@cp03 serverok-rubycms]# cat docker-compose.yml

version: '3'

services:

web:

image: serverok/rubycms:1.0

restart: always

command: bash -c "rm -f tmp/pids/server.pid && cd /myapp && source /etc/profile.d/rvm.sh && script/server -e production"

volumes:

- /home/naiwnyc/cms:/myapp

ports:

- "3000:3000"

[root@cp03 serverok-rubycms]#

You need to rebuild docker container based on this docker-compose.yml file

Change to the folder where your docker-compse.yml file is, in my case it was

cd  ~/serverok-rubycms

1	cd ~/serverok-rubycms

Run

docker-compose down
docker-compose up -d

1 2	docker-compose down docker-compose up -d

docker-compose: error while loading shared libraries

March 3, 2019

by ServerOk with No Comment Docker

When running docker-compose on CentOS 7, i get following error

[root@cp03 serverok-rubycms]# docker-compose up 
docker-compose: error while loading shared libraries: libz.so.1: failed to map segment from shared object: Operation not permitted
[root@cp03 serverok-rubycms]#

[root@cp03 serverok-rubycms]# docker-compose up

docker-compose: error while loading shared libraries: libz.so.1: failed to map segment from shared object: Operation not permitted

[root@cp03 serverok-rubycms]#

To fix the error, do the following

mv /usr/local/bin/docker-compose /usr/local/bin/docker-compose-bin

1	mv /usr/local/bin/docker-compose /usr/local/bin/docker-compose-bin

Now create a new file

vi /usr/local/bin/docker-compose

1	vi /usr/local/bin/docker-compose

Add following content

#!/bin/bash

export TMPDIR=/opt/compose-tmp
/usr/local/bin/docker-compose-bin "$@"

#!/bin/bash

export TMPDIR=/opt/compose-tmp

/usr/local/bin/docker-compose-bin "[email protected]"

Make it executable

chmod 755 /usr/local/bin/docker-compose

1	chmod 755 /usr/local/bin/docker-compose

Create temp folder

mkdir -p /opt/compose-tmp

1	mkdir -p /opt/compose-tmp

Now docker-compose will work.

Dockerfile

February 10, 2019

by ServerOk with No Comment Docker

Create Python Flask Docker Container

February 10, 2019

by ServerOk with No Comment Docker

Create a folder and change to the folder

mkdir python-flask
cd python-flask

1 2	mkdir python-flask cd python-flask

Create file

vi Dockerfile

1	vi Dockerfile

Add following content

FROM python:2.7-slim

WORKDIR /app
COPY . /app

RUN pip install -r requirments.txt

EXPOSE 8080

CMD ["python", "app.py"]

FROM python:2.7-slim

WORKDIR /app

COPY . /app

RUN pip install -r requirments.txt

EXPOSE 8080

CMD ["python", "app.py"]

Create file requirements.txt, add “Flask” to it.

echo "Flask" > requirments.txt

1	echo "Flask" > requirments.txt

Now lets create our Python Flask Application

vi app.py

vi app.py

Add following content

from flask import Flask
import os
import socket

app = Flask(__name__)

@app.route("/")
def hello():
	return  "<h1>Ok, World</h1>"

if __name__ == "__main__":
	app.run(host="0.0.0.0", port=8080)

from flask import Flask

import os

import socket

app = Flask(__name__)

@app.route("/")

def hello():

return "<h1>Ok, World</h1>"

if __name__ == "__main__":

app.run(host="0.0.0.0", port=8080)

To test Flask application locally, install Flask using pip

sudo pip install Flask

1	sudo pip install Flask

Now run the application using

python app.py

1	python app.py

Now you will be able to see web application at

http://localhost:8080

1	http://localhost:8080

Press CTRL+C to stop the application.

To build Docker image, run

docker build -t python-flask-app .

1	docker build -t python-flask-app .

-t specify tag.

If all worked properly, you will see

You can see the container image listed

docker images

1	docker images

Now your docker image is ready. To build a container using this image, run

docker container run -d --name my-flask-app -p 8080:8080 python-flask-app

1	docker container run -d --name my-flask-app -p 8080:8080 python-flask-app

You can access your python application running inside docker container on URL

http://localhost:8080

1	http://localhost:8080

To stop the application, run

docker stop my-flask-app

1	docker stop my-flask-app

To start it, run

docker start my-flask-app

1	docker start my-flask-app

Create Kubernetes cluser using gcloud

February 10, 2019

by ServerOk with No Comment Docker

gcloud auth login

1	gcloud auth login

Set default region

gcloud config set compute/zone us-central1-a

1	gcloud config set compute/zone us-central1-a

To create a cluser, run

gcloud container clusters create CLUSTER-NAME

1	gcloud container clusters create CLUSTER-NAME

To create a cluster with 2 nodes, run

gcloud container clusters create CLUSTER-NAME --num-nodes 2

1	gcloud container clusters create CLUSTER-NAME --num-nodes 2

To specify compute engine type during cluster creation, run

gcloud container clusters create CLUSTER-NAME --num-nodes 2 --machine-type n1-standard-2

1	gcloud container clusters create CLUSTER-NAME --num-nodes 2 --machine-type n1-standard-2

You can also specify scopes, this allow you to give the cluster permission to access other google cloud resources. For example to allow cluster access Cloud Source Repositories and Google Container Registry. run

gcloud container clusters create CLUSTER-NAME --num-nodes 2 --machine-type n1-standard-2 --scopes "https://www.googleapis.com/auth/projecthosting,cloud-platform"

1	gcloud container clusters create CLUSTER-NAME --num-nodes 2 --machine-type n1-standard-2 --scopes "https://www.googleapis.com/auth/projecthosting,cloud-platform"

To list your container, run

gcloud container clusters list

1	gcloud container clusters list

Before you can use cluser, you need to get authentication credentials with command

gcloud container clusters get-credentials CLUSTER-NAME

1	gcloud container clusters get-credentials CLUSTER-NAME

To interact with Kubernetes, you need kubectl, on Ubuntu, yu can install it with

sudo snap install kubectl --classic

1	sudo snap install kubectl --classic

To deploy a container, run

kubectl run hello-server --image=gcr.io/google-samples/hello-app:1.0 --port 8080

1	kubectl run hello-server --image=gcr.io/google-samples/hello-app:1.0 --port 8080

This will deploy google provided hello-app container to the cluser.

To make the application available to public, create a loadbalancer service.

kubectl expose deployment hello-server --type="LoadBalancer"

1	kubectl expose deployment hello-server --type="LoadBalancer"

Now find IP of the LoadBalacner service with command

kubectl get services

1	kubectl get services

From the result, we found EXTERNAL-IP for LoadBalancer is 35.232.226.216. To access the application, go to url

http://35.232.226.216:8080

See Kubernetes

Docker container start on boot

January 2, 2019

by ServerOk with No Comment Docker

To set a docker container to start on system reboot, run

docker update --restart=unless-stopped CONTAINER_ID

1	docker update --restart=unless-stopped CONTAINER_ID

Replace CONTAINER_ID with actual ID of your container.

Docker Nginx Proxy

December 23, 2018

by ServerOk with No Comment Docker

Docker Nginx Proxy allow you to run multiple docker containers on same server behind nginx proxy. This is done using

https://github.com/jwilder/nginx-proxy

To do this, you need a server with port 80 and 443 unused.

To setup nginx proxy, run following

cd /root/
git clone https://github.com/evertramos/docker-compose-letsencrypt-nginx-proxy-companion.git
cd docker-compose-letsencrypt-nginx-proxy-companion
cp .env.sample .env
./start.sh

cd /root/

git clone https://github.com/evertramos/docker-compose-letsencrypt-nginx-proxy-companion.git

cd docker-compose-letsencrypt-nginx-proxy-companion

cp .env.sample .env

./start.sh

This will start nginx proxy. You can modify .env file if you want.

Starting a Docker Web App Behind Proxy

To start a web app, all you need is to start docker container on same network as nginx proxy. By default it is “webproxy”.

Here is an example command to start a web server.

docker run -d 
    -e VIRTUAL_HOST=test.serverok.in \
    -e LETSENCRYPT_HOST=test.serverok.in \
    -e LETSENCRYPT_EMAIL=admin@serverok.in \
    --network=webproxy \
    --name my_app \
    httpd:alpine

docker run -d

-e VIRTUAL_HOST=test.serverok.in \

-e LETSENCRYPT_HOST=test.serverok.in \

-e LETSENCRYPT_EMAIL=admin@serverok.in \

--network=webproxy \

--name my_app \

httpd:alpine

This will start a test web server. You need to point the domain specified to this servers IP, then only nginx proxy can get LetsEncrypt SSL installed.

Replace test.serverok.in with your actual domain.

If you don’t want LetsEncrypt SSL installed, you can remove following 2 options

    -e LETSENCRYPT_HOST=test.serverok.in \
    -e LETSENCRYPT_EMAIL=admin@serverok.in \

1 2	-e LETSENCRYPT_HOST=test.serverok.in \ -e LETSENCRYPT_EMAIL=admin@serverok.in \

Docker Build

May 4, 2018

by ServerOk with No Comment Docker

docker build command is used to create a docker image from Dockerfile.

To create an image, first lets create a working directory and change to the directory.

mkdir ~/my-docker-image
cd ~/my-docker-image

1 2	mkdir ~/my-docker-image cd ~/my-docker-image

Create a file with name Dockerfile

vi Dockerfile

1	vi Dockerfile

Put following content in it

FROM ubuntu:18.04

RUN apt update && apt upgrade -y

CMD ["/bin/bash"]

FROM ubuntu:18.04

RUN apt update && apt upgrade -y

CMD ["/bin/bash"]

Now build the docker image with

docker build -t my-ubuntu .

1	docker build -t my-ubuntu .

This will create a docker image with name “my-ubuntu”.

To run this, we need to create a container using this image, this can be done with command

docker run -ti my-ubuntu

1	docker run -ti my-ubuntu

Here i used -ti as we need to run interactively using terminal, this is because our the command we set to start is “/bin/bash” in the Dockerfile.

Docker