This continution of could provisioning of EC2 Docker Host machine, This Repository deals with java based Dockerized Application Container Deployment through Docker compose and Docker file scripting, Further shipping Re-producible docker images through Jenkins Pipeline(Continuous-Integration and Continuous-Deployment Pipeline).
This repository is tries to exemplify how to automatically manage the process of building and deployment phases.
First we will setup the Container infrastructure through docker-compose and docker file scripting then ensure to build reproducible docker images with Jenkins pipeline whenver we find any changes in application code.
To ensure building of reproducible docker images, we will be provisioning a Jenkins container then we will add Jenkins Pipeline script and Git Webhook Configuration to oensure ur pipeline works well after each code being pushed.
We can further Scale and Orchestrate the above Application through Kubernetes or other scaling solutions.
Following are the pipeline stages of Jenkins Docker Image Build Pipeline:
- Code checkout
- Create Docker image
- Push the image to Docker Hub
- Create the container with port mappings
Since one of the goals is to obtain the sonarqube
report of our project, we should be able to access sonarqube from the jenkins service. Docker compose
is a best choice to run services working together. We configure our application services in a yaml file as mentioned below.
docker-compose.yml
version: "3.1"
services:
reverse_proxy:
build: ./reverse_proxy
user: nginx
appserver:
build:
context: app
dockerfile: Dockerfile
container_name: appserver
user: app
ports:
- "5005:5005"
networks:
- front-tier
- back-tier
jenkins:
build:
context: jenkins/
privileged: true
user: root
ports:
- "8080:8080"
container_name: jenkins
volumes:
- /tmp/jenkins:/var/jenkins_home #Remember that, the tmp directory is designed to be wiped on system reboot.
- /var/run/docker.sock:/var/run/docker.sock
networks:
front-tier:
back-tier:
driver: overlay
Self-Signed SSL Certificate Generation:
This application uses Docker secrets to secure the application components such as self-signed certificates. The reverse proxy requires creating a certificate that is stored as a secret. To create a certificate and add as a secret, run the following commands in Docker host machine:
mkdir certs
openssl req -newkey rsa:4096 -nodes -sha256 -keyout certs/domain.key -x509 -days 365 -out certs/domain.crt
docker secret create revprox_cert certs/domain.crt
docker secret create revprox_key certs/domain.key
Paths of docker files of the containers are specified at context attribute in the docker-compose file. Content of these files as follows.
jenkins/Dockerfile
FROM jenkins:2.60.3
If we run the following command in the same directory as the docker-compose.yml
file, the Sonarqube and Jenkins containers will up and run.
docker-compose -f docker-compose.yml up --build
docker ps
CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES
87105432d655 pipeline_jenkins "/bin/tini -- /usr..." About a minute ago Up About a minute 0.0.0.0:8080->8080/tcp, jenkins
f5bed5ba3266 appserver "./bin/run.sh" About a minute ago Up About a minute 0.0.0.0:5005->5005/tcp, appserver
f5bed5ba3288 reverse_proxy "./bin/run.sh" About a minute ago Up About a minute 0.0.0.0:80->80/tcp, 0.0.0.0:443->443/tcp reverse_proxy
-
We’ll define a service on Github to call the
Jenkins Github webhook
because we want to trigger the pipeline. To do this go to Settings -> Integrations & services. TheJenkins Github plugin
-
After this, we should add a new service by typing the URL of the dockerized Jenkins container along with the
/github-webhook/
path. -
The next step is that create an
SSH key
for a Jenkins user and define it asDeploy keys
on our GitHub repository. -
If everything goes well, the following connection request should return with a success.
ssh [email protected]
PTY allocation request failed on channel 0
Hi <your github username>/<repository name>! You've successfully authenticated, but GitHub does not provide shell access.
Connection to github.com closed.
We have configured Jenkins in the docker compose file to run on port 8080 therefore if we visit http://localhost:8080 we will be greeted with a screen like this.
We need the admin password to proceed to installation. It’s stored in the /var/jenkins_home/secrets/initialAdminPassword
directory and also It’s written as output on the console when Jenkins starts.
jenkins | *************************************************************
jenkins |
jenkins | Jenkins initial setup is required. An admin user has been created and a password generated.
jenkins | Please use the following password to proceed to installation:
jenkins |
jenkins | 45638c79cecd4f43962da2933980197e
jenkins |
jenkins | This may also be found at: /var/jenkins_home/secrets/initialAdminPassword
jenkins |
jenkins | *************************************************************
To access the password from the container.
docker exec -it jenkins sh
/ $ cat /var/jenkins_home/secrets/initialAdminPassword
After entering the password, we will download recommended plugins and define an admin user
.
After clicking Save and Finish and Start using Jenkins buttons, we should be seeing the Jenkins homepage. One of the seven goals listed above is that we must have the ability to build an image in the Jenkins being dockerized. Take a look at the volume definitions of the Jenkins service in the compose file.
- /var/run/docker.sock:/var/run/docker.sock
The purpose is to communicate between the Docker Daemon
and the Docker Client
(we will install it on Jenkins) over the socket. Like the docker client, we also need Maven
to compile the application. For the installation of these tools, we need to perform the Maven
and Docker Client
configurations under Manage Jenkins -> Global Tool Configuration menu.
We have added the Maven and Docker installers
and have checked the Install automatically
checkbox. These tools are installed by Jenkins when our script(Jenkins file) first runs. We give myMaven
and myDocker
names to the tools. We will access these tools with this names in the script file.
Since we will perform some operations such as checkout codebase
and pushing an image to Docker Hub
, we need to define the Docker Hub Credentials
. Keep in mind that if we are using a private repo, we must define Github credentials
. These definitions are performed under Jenkins Home Page -> Credentials -> Global credentials (unrestricted) -> Add Credentials menu.
We use the value we entered in the ID
field to Docker Login in the script file. Now, we define pipeline under Jenkins Home Page -> New Item menu.
In this step, we select GitHub hook trigger for GITScm pooling
options for automatic run of the pipeline by Github hook
call.
Also in the Pipeline section, we select the Pipeline script from SCM
as Definition, define the GitHub repository and the branch name, and specify the script location (Jenkins file).
stage('Initialize'){
def dockerHome = tool 'myDocker'
def mavenHome = tool 'myMaven'
env.PATH = "${dockerHome}/bin:${mavenHome}/bin:${env.PATH}"
}
The Maven
and Docker client
tools we have defined in Jenkins under Global Tool Configuration menu are added to the PATH environment variable
for using these tools with sh command
.
stage('Push to Docker Registry'){
withCredentials([usernamePassword(credentialsId: 'dockerHubAccount', usernameVariable: 'USERNAME', passwordVariable: 'PASSWORD')]) {
pushToImage(CONTAINER_NAME, CONTAINER_TAG, USERNAME, PASSWORD)
}
}
With reverse proxy configuration, we can achieve non https redirection to https protocol
You can access the application at https://localhost
@TODO:
I will try further to add High-availability of the application though Kubernetes or sclaing Docker server through Auto-scaling with scaling policies. As part of above process we are building the docker image with respective tags and pushing back to the DockeHub Registry.
In this scenraio usage of Kubenetes is much better for scaling the application across multiple containers(Horizontal scaling) instead host machine scaling(vertical scaling) everytime which is not a good practice.
We can use Deployment Patterns in "Kubernetes" for Application scaling and Deployment such as Canary deployment, provided proper moniotring is integrated such as Prometheus