Conatiner Based Development

Log in to https://labs.play-with-docker.com/

Simple App

alias build-machine='docker run --rm -it -v $PWD:/my -w /my maven:3-jdk-9-slim'
build-machine bash
# inside the container
mvn -q archetype:generate # filter: jdk9, item #1, group: grp, artifactId:yourname-jdk
cd yourname-jdk
mvn -q package
java -jar target/yourname-jdk-1.0-SNAPSHOT.jar
# close the container ^d
docker build -t test yourname-jdk
docker run --rm test

REST Service

alias build-machine='docker run --rm -it -v $PWD:/my -w /my maven:3-jdk-9-slim'
build-machine mvn -q archetype:generate
# filter: dropwizard-app, item #1, group: grp, artifactId:yourname-dw
cd yourname-dw
build-machine mvn -q package
docker build -t test .
docker run --rm -p 8080:8080 test 

Using docker compose

docker-compose up -d
# verify the services are up using
docker-compose ps

Add DB

Uncomment db definition in docker-compose.yml. Repeat last step. Shutdown the cluster using docker-compose down. Launch it again and verify that the data is there.

Scale

Try scaling using docker-compose scale yourname-dw=3. Doesn't work.

Edit the docker-compose.yml, removing 8081, 3306 ports, and remove mapping for 8080.

version: "3"
services:
  yourname-dw:
    build: .
    image: yourname-dw:1.0-SNAPSHOT
    ports:
      - "8080"
    environment:
      JDBC_DRIVER: "com.mysql.cj.jdbc.Driver"
      JDBC_URL: "jdbc:mysql://db:3306/mydb?useSSL=false"

  db:
    image: mysql
    environment:
      MYSQL_ROOT_PASSWORD: root
      MYSQL_USER: root
      MYSQL_PASSWORD: 123456
      MYSQL_DATABASE: mydb
    volumes:
      - yourname-dw-volume:/var/lib/mysql

volumes:
  yourname-dw-volume

run:

docker-compose down
docker-compose up -d --scale yourname-dw=3

Swarm

Single Node Swarm

docker swarm init --advertise-addr $YOUR_IP
wget https://eitan101.github.io/containers/manifests-examples/myapp/docker-compose.yml
docker deploy -c docker-compose.yml my
docker service ls
docker service scale my_myapp=3
# open http://{host_ip}:8080
docker service ls
docker service logs -f my_myapp
docker stack rm my

Swarm Cluster

In the play with docker site:

• Delete your instance
• Press the wrench icon
• Choose 2 Managers and 3 workers template.

From the manager1 instance repeat the previous scenario, skipping the swarm init command.

Add Visual Dashboard

Repeat previous section but add before:

docker run -it -d -p 18080:8080 -v /var/run/docker.sock:/var/run/docker.sock dockersamples/visualizer

Do the same with

docker run -d -p 9000:9000 -v /var/run/docker.sock:/var/run/docker.sock portainer/portainer

Run on single node with local images

alias build-machine='docker run --rm -it -v $PWD:/my -w /my maven:3-jdk-9-slim'
build-machine mvn -q archetype:generate
# filter: dropwizard-app, item #1, group: grp, artifactId:myapp
cd myapp
build-machine mvn -q package
docker-compose build
docker swarm init --advertise-addr $YOUR_IP
docker deploy -c docker-compose.yml my
docker service ls
docker service scale my_myapp=3
# open http://{host_ip}:8080
docker service ls
docker service logs -f my_myapp
docker stack rm my

Kubernetes

Prerequisites

Signup to Katacoda

Minikube

Scenario can be found here

Minikube with local images

Scenario can be found here

Run minikube on your device locally

Install minikube.

alias build-machine='docker run --rm -it -v $PWD:/my -w /my maven:3-jdk-9-slim'
build-machine mvn -q archetype:generate
# filter: dropwizard-app, item #1, group: grp, artifactId:myapp
# or: git clone https://github.com/eitan101/containers.git && cd containers
cd myapp
build-machine mvn -q package
minikube start
eval $(minikube docker-env)
docker-compose build
kubectl apply -f k8s.yml
watch kubectl get pods
minikube service myapp
# update app "replicas: 1" to 3
kubectl apply -f k8s.yml
# minikube stop/start - to test data persistancy
kubectl delete -f k8s.yml

Add Weave Scope Visualizer

kubectl apply -f "https://cloud.weave.works/k8s/scope.yaml?k8s-service-type=NodePort&k8s-version=$(kubectl version | base64 | tr -d '\n')"
minikube service -n weave weave-scope-app

Kafka Clients

Using Swarm

#docker swarm init --advertise-addr $YOUR_IP
wget https://eitan101.github.io/containers/manifests-examples/mykafka/docker-compose.yml
docker deploy -c docker-compose.yml my
watch docker service ls
docker run -d -p 9000:9000 -v /var/run/docker.sock:/var/run/docker.sock portainer/portainer
docker service logs -f my_mykafka-consumer
docker service scale my_mykafka-consumer=3
docker service logs -f my_mykafka-consumer
docker stack rm mystack

Using Swarm with local images

alias build-machine='docker run --rm -it -v $PWD:/my -w /my maven:3-jdk-9-slim'
build-machine mvn -q archetype:generate -Dfilter=kafka-app
cd mykafka/
build-machine mvn -q package
#docker swarm init --advertise-addr $YOUR_IP
docker-compose build
docker deploy -c docker-compose.yml mystack
watch docker service ls
docker service logs -f mystack_mykafka-consumer
docker service scale mystack_mykafka-consumer=3
docker service logs -f mystack_mykafka-consumer
docker stack rm mystack

Using minikube

Scenario can be found here

Run minikube on your device locally

Install minikube.

alias build-machine='docker run --rm -it -v $PWD:/my -w /my maven:3-jdk-9-slim'
build-machine mvn -q archetype:generate -Dfilter=kafka-app
# filter: kafka-app, item #1, group: grp, artifactId:mykafka
cd mykafka
build-machine mvn -q package
minikube start
eval $(minikube docker-env)
docker-compose build
kubectl apply -f k8s.yml
open http://0.0.0.0:30000

Run minikube on your device locally

Install minikube.

minikube start
eval $(minikube docker-env)
kubectl apply -f  https://eitan101.github.io/containers/manifests-examples/mykafka/k8s.yml

Diagram