install_guide_aks.en.md

Installation Guide of RoboticBase Core (Microsoft Azure AKS)

Requirements

	version
OS	macOS Mojave 10.14.6 Ubuntu 16.04
pyenv	1.2.16
pipenv	2018.11.26
kubectl	1.17.3
helm	3.1.1
openssl	2.6.5
azure cli	2.1.0

Preparation

Install tools

Install pyenv and pipenv

macOS

$ brew install pyenv
$ echo 'export PYENV_ROOT="$HOME/.pyenv"' >> ~/.bashrc
$ echo 'export PATH="$PYENV_ROOT/bin:$PATH"' >> ~/.bashrc
$ echo 'eval "$(pyenv init -)"' >> ~/.bashrc
$ source ~/.bashrc
$ brew install pipenv

ubuntu

$ sudo apt install -y build-essential libffi-dev libssl-dev zlib1g-dev libbz2-dev libreadline-dev libsqlite3-dev git python3-pip
$ git clone https://github.com/pyenv/pyenv.git ~/.pyenv
$ echo 'export PYENV_ROOT="$HOME/.pyenv"' >> ~/.bashrc
$ echo 'export PATH="$PYENV_ROOT/bin:$PATH"' >> ~/.bashrc
$ echo 'eval "$(pyenv init -)"' >> ~/.bashrc
$ source ~/.bashrc
$ pip3 install pipenv

Install kubectl

macOS

$ curl -LO "https://storage.googleapis.com/kubernetes-release/release/v1.17.3/bin/darwin/amd64/kubectl"
$ chmod +x ./kubectl
$ sudo mv ./kubectl /usr/local/bin/kubectl

ubuntu

$ curl -LO "https://storage.googleapis.com/kubernetes-release/release/v1.17.3/bin/linux/amd64/kubectl"
$ chmod +x ./kubectl
$ sudo mv ./kubectl /usr/local/bin/kubectl

Install helm

macOS

$ curl -LO "https://get.helm.sh/helm-v3.1.1-darwin-amd64.tar.gz"
$ tar xvfz helm-v3.1.1-darwin-amd64.tar.gz
$ sudo mv darwin-amd64/helm /usr/local/bin/helm

ubuntu

$ curl -LO "https://get.helm.sh/helm-v3.1.1-linux-amd64.tar.gz"
$ tar xvfz helm-v3.1.1-linux-amd64.tar.gz
$ sudo mv linux-amd64/helm /usr/local/bin/helm

Install openssl

macOS

$ brew install openssl

ubuntu

$ sudo apt install -y openssl

Install Azure CLI

macOS

$ brew install azure-cli

ubuntu

$ curl -sL https://aka.ms/InstallAzureCLIDeb | sudo bash

Prepare ansible

Clone RoboticBase/core

$ git clone https://github.com/RoboticBase/core.git
$ cd core

Install ansible and related libraries

$ cd ansible
$ pipenv install

Installation of RoboticBase/core

Set your variables (mandatory)

1. Set the password of MQTT user (iotagent) in the following yaml file:
   • group_vars/all.yml
     • target: mqtt.users[?name==`iotagent`].password
2. Change the domain name defined in the following yaml file to your own domain:
   • inventories/aks/group_vars/aks.yml
     • target: dns.domain
3. Change the email address defined in the following yaml file to your own email:
   • inventories/aks/group_vars/aks.yml
     • target: dns.email
4. Change the SSH key path defined in the following yaml file to your own public key:
   • inventories/aks/host_vars/azure.yml
     • target: resources.ssh_key_file_path

Update your variables (optional)

1. If necessary, update the variables in the following yaml file such as the number of worker nodes or the size of VM:
   • inventories/aks/group_vars/aks.yml
2. If necessary, update the variables in the following yaml file such as the number of pod replicas:
   • inventories/aks/group_vars/aks.yml

Generate Azure Credentials

1. Generate Azure Credentials which are used to operate Azure from ansible by executing following shell script:

   $ ./tools/generate_azure_credentials.sh
   

Start RoboticBase/core

1. Start "pipenv shell"

   $ pipenv shell
   

2. Start RoboticBase/core on Azure AKS using ansible

   $ ansible-playbook -i inventories/aks --extra-vars="ansible_python_interpreter=$(which python)" aks.yml
   

3. After starting ansible-playbook, you will be prompted to log in Azure

Configure grafana

1. Access grafana (https://grafana.{{ your domain }})
2. Input "admin" to email or username and "prom-operator" to password, and push Log In
3. Select Change Password from Preferences, and change the password of admin
4. Confirm the dashboards which were installed for monitoring the resources of Azure AKS

Configure kibana

1. Confirm the username and password of kibana

   $ ./tools/show_kibana_credentials.py
   

2. Access kibana (https://kibana.{{ your domain }})

3. Input the username and password of kibana, and Log in kibana

4. Show Home view by clicking Explore on my own

5. Show Management view by clicking Management

6. Click Index Patterns

7. Click Create Index Patterns

8. Input "logstash-*" to Index pattern, and click Next Step

9. Select "@timestamp" at Time Filter field name, click Create Index pattern

10. The Index of log messages is created automatically

11. Click Discover, and confirm the log messages gathered from Pods