Camping microservices orchestration and automation with xOpera and Ansible

Table of Contents

• Purpose
• Functionality
• Quick test and deploy
• Running with xOpera
  • Setting inputs
  • Deploy
  • Undeploy

Purpose

The repository implements the orchestration and automation of camping microservices on managed Microsoft Azure Kubernetes instance. The orchestration is done using xOpera TOSCA orchestrator and Ansible is used for the automation.

The solution does the following:

• login to the Azure account
• create a new Azure Resource Group
• create a new Azure Kubernetes Service (AKS) cluster
• deploy camping microservices on AKS
• create an ingress controller on AKS
• establish monitoring with Datadog

Functionality

The main functionality of image-resize is to create thumbnails from the source image. Source image must be uploaded into source bucket and then three thumbnails will be created and saved to another bucket.

Quick test and deploy

If you want to test this solution immediately run the following commands:

# clone this repo
git clone [email protected]:camping-rso/orchestration.git
cd orchestration

# install prerequisites into Python virtualenv
python3 -m venv .venv && . .venv/bin/activate
pip install --upgrade pip
pip install -r requirements.txt

# run xOpera deployment (make sure to setup the inputs in inputs.yml file)
opera deploy -i inputs.yml service_template.yml

# run undeployment if you wish to tear down the deployed services
opera undeploy

Running with xOpera

This topic explains orchestration of the prepared TOSCA YAML artifact using xOpera.

Inputs

The artifact gets deployed from an existing virtual env that should be created before. To begin the orchestration you have to install xOpera to your virtualenv with pip. Parameters are passed to xOpera via inputs.yml file that can be modified. Currently the following values can be modified:

Variable	Description	Example
host_ip	Host ip for the deployment	localhost
azure_credentials	Azure credentials, is of type map which should include both Azure username and password	{ username: user, password: secret }
location	The location for the Azure resources	westeurope
resource_group_name	The name of the testing resource group	CampingResourceGroup
aks_name	The name for the new AKS cluster	camping
camping_overview_ms_db_connection_string	Connection string to PostgreSQL	"Host=localhost;Port=5432;Database=camping;Username=postgres;Password=potgres;Pooling=true;"
camping_reservations_ms_db_connection_string	Connection string to PostgreSQL	"Host=localhost;Port=5432;Database=camping;Username=postgres;Password=potgres;Pooling=true;"
camping_opinions_ms_db_connection_string	Connection string to PostgreSQL	"Host=localhost;Port=5432;Database=camping;Username=postgres;Password=potgres;Pooling=true;"
camping_images_ms_postgres_credentials	PostgreSQL credentials	{ username: postgres, password: postgres }
camping_catalog_ms_influx_credentials	PostgreSQL credentials	{ host: "influx.com", port: 12345, user: adim, password: secret, database: test }
datadog_api_key	Datadog API key	dkwferf4g89r4g89rg4r4t84

Deploy

To deploy the prepared blueprint run opera deploy -i inputs.yml service_template.yml If the deploy has finished successfully your AKS was created and is prepared for further usage.

Undeploy

Run opera undeploy to undeploy the solution. This will remove the created Azure Kubernetes service from the Azure portal along with prepared environment (resource group and other resources).