/rancher-demonstration

Rancher 2.x demonstration

Primary LanguageShell

rancher-demonstration

This repository is a simple demonstration of virtualized environments (vagrant + virtualbox), tailored to exhibit the kubernetes-based rancher-server, and rancher-agent ecosystem. Specifically, a Vagrantfile will automate the processes contained within the provided install scripts.

In this setup, the rancher-server is launched using Centos7x, while the rancher-agent is launched in Bionic64. If the rancher-server needs to be debian-based, or corresponding rancher-agents need to be rhel-based, then additional utility scripts, not included in this repository, will need to be created.

Regardless of implementation, when vagrant completes provisioning, a rancher-server, is available via https://localhost:7895, and can be used to manage various kubernetes-based clusters:

Rancher Login

Upon immediate login, the rancher-server may still be provisioning:

Rancher Provisioning

Note: the provided install scripts, used to provision the corresponding vagrant virtual machine(s), can also be used on production-like environments. However, for high-availability, the install-rancher-server will need to be adjusted.

After a few minutes, the cluster will complete provisioning, and be active:

Rancher Active

Then, the newly created cluster can be reviewed:

Rancher Cluster

Additional hosts

Additional hosts can be added to a desired cluster, by first clicking Edit under the top-right hamburger icon:

rancher-add-host

At the bottom of the associated page, the corresponding docker command can be pasted into the desired cluster host:

rancher-new-host-command

Deployment

The kubectl command can be executed on the rancher-server via the web-browser, or directly within the container:

[root@rancher-server vagrant]# docker ps -a
CONTAINER ID        IMAGE                    COMMAND             CREATED             STATUS              PORTS                                                               NAMES
92dde0d45453        rancher/rancher:v2.1.5   "entrypoint.sh"     38 hours ago        Up 38 hours         0.0.0.0:8890->80/tcp, 0.0                      .0.0:8895->443/tcp   rancher
[root@rancher-server vagrant]# docker exec -it rancher /bin/bash
root@92dde0d45453:/var/lib/rancher# kubectl run \
  --image=nginx \
  --port=80 \
  --env='DOMAIN=cluster' \
  replicas=3

Note: if --env= is passed, environment variables can be read from STDIN using the standard env syntax.

Additionally, a yaml configuration file can be utilized:

[root@rancher-server vagrant]# docker ps -a
CONTAINER ID        IMAGE                    COMMAND             CREATED             STATUS              PORTS                                                               NAMES
92dde0d45453        rancher/rancher:v2.1.5   "entrypoint.sh"     38 hours ago        Up 38 hours         0.0.0.0:8890->80/tcp, 0.0                      .0.0:8895->443/tcp   rancher
[root@rancher-server vagrant]# docker exec -it rancher /bin/bash
root@92dde0d45453:/var/lib/rancher# kubectl create -f ./manifest.yaml
root@92dde0d45453:/var/lib/rancher# kubectl get deployments
NAME               DESIRED   CURRENT   UP-TO-DATE   AVAILABLE   AGE
nginx-deployment   3         3         3            3           18s

The following is an example manifest.yaml:

apiVersion: apps/v1
kind: Deployment
metadata:
  name: nginx-deployment
  labels:
    app: nginx
spec:
  replicas: 3
  selector:
    matchLabels:
      app: nginx
  template:
    metadata:
      labels:
        app: nginx
    spec:
      containers:
        - name: nginx
          image: nginx:1.7.9
          ports:
          - containerPort: 80
        env:
        - name: DOMAIN
          value: cluster

Note: for development purposes, the kind: Pod can be implemented if replicas are not needed. However, in production systems, the kind: Deployment is typically desired.

Note: Kompose can convert an existing docker-compose.yml to a series of kubernetes yaml files.

Alternatively, instead of defining multiple environment variables for each yaml file, a custom configMapRef can be created:

root@92dde0d45453:/var/lib/rancher# cat config/special_config.yml
apiVersion: v1
kind: ConfigMap
metadata:
  name: special-config
  namespace: default
data:
  DOMAIN: cluster
root@92dde0d45453:/var/lib/rancher# kubectl create configmap special_config --from-file=config/special_config.yml

Note: the ConfigMap API resource stores configuration data as key-value pairs. The data can be consumed in pods or provide the configurations for system components such as controllers. ConfigMap is similar to Secrets, but provides a means of working with strings that don’t contain sensitive information.

This allows the above manifest.yaml to refactor as follows:

apiVersion: apps/v1
kind: Deployment
metadata:
  name: nginx-deployment
  labels:
    app: nginx
spec:
  replicas: 3
  selector:
    matchLabels:
      app: nginx
  template:
    metadata:
      labels:
        app: nginx
    spec:
      containers:
      - name: nginx
        image: nginx:1.7.9
        ports:
          - containerPort: 80
        envFrom:
          - configMapRef:
            name: special-config
            key: DOMAIN

Note: if the envFrom > key is omitted, the entire special-config, will be available within the given container.

Configuration

Fork this project in your GitHub account. Then, clone your repository, with one of the following approaches:

  • simple clone: clone the remote master branch.
  • commit hash: clone the remote master branch, then checkout a specific commit hash.
  • release tag: clone the remote branch, associated with the desired release tag.

Installation

In order to proceed with the installation for this project, three dependencies need to be installed:

Once the necessary dependencies have been installed, execute the following command to build the rancher-server:

cd /path/to/rancher-demonstration/
vagrant up

Note: an alternative syntax to vagrant up, is to run vagrant up rancher-server. However, the associated rancher-agent needs to be created as well.

Depending on the network speed, the build can take between 2-4 minutes. So, grab a cup of coffee, and perhaps enjoy a danish while the virtual machine builds.

Note: a more complete refresher on virtualization, can be found within the vagrant wiki page.

Though, the implemented install scripts can be used to provision vagrant, it can be run on non-vagrant systems. For example, the install scripts can easily be run on corresponding virtual machines:

## virtual machine for rancher-server
./install-rancher-server \
  "$server_version" \
  "$server_internal_port" \
  "$server_internal_https_port" \
>> "${project_root}/logs/install-rancher-server.txt" 2>&1

## virtual machine for rancher-agent
./install-rancher-agent \
  "$agent_version" \
  "$server_ip" \
  "$server_internal_port" \
  "$server_internal_https_port" \
>> "${project_root}/logs/install-rancher-agent.txt" 2>&1

Note: when running the above install scripts, it is assumed that docker, and rancher-cli dependencies have been accounted.