A Terraform module to deploy and run MigratoryData on Microsoft Azure using Ansible.
This guide provides a step-by-step process to create an infrastructure on Azure using Terraform, and deploy the MigratoryData server cluster using Ansible. It also covers how to add a monitoring module, using Ansible, that installs Prometheus and Grafana for real-time statistics. Additionally, the guide includes instructions on how to update the MigratoryData server across all instances.
Prerequisites
Before deploying MigratoryData using Terraform, ensure that you have a Microsoft Azure account and have installed the following tools:
Login to Azure
Login to Azure with the following command and follow the instructions on the screen to configure your Azure credentials:
az login
For other authentication methods, see Terraform’s Azure Provider documentation.
Configure the deployment
Clone the MigratoryData’s repository with terraform configuration files:
git clone https://github.com/migratorydata/terraform-azure-migratorydata.git
cd terraform-azure-migratorydata/deploy
Update if necessary the configuration files from the deploy/configs
directory. See the Configuration guide for more details. If you developed custom extensions, add them to the the deploy/extensions
directory.
Update deploy/terraform.tfvars
file to match your configuration. The following variables are required:
region
- The Azure region where the resources will be deployed.namespace
- The namespace for the resources.address_space
- The address space for the virtual network.num_instances
- The number of nodes to start the MigratoryData cluster.max_num_instances
- The maximum number of instances of MigratoryData Nodes to scale the deployment when necessary.instance_type
- The type of the virtual machines to be deployed.ssh_private_key
- The path to the private key used to access the virtual machines.enable_monitoring
- A boolean value to enable or disable monitoring and logging.
Your terraform.tfvars
file should be similar to the following:
region = "eastus"
namespace = "migratorydata"
additional_tags = {}
address_space = "10.0.0.0/16"
num_instances = 3
max_num_instances = 5
instance_type = "Standard_F2s_v2"
ssh_private_key = "~/.ssh/id_rsa"
enable_monitoring = true
To update the version of MigratoryData server, update the ansible/vars.yaml
file. The following variables are required:
package_url
- The URL where the MigratoryData package can be downloaded.package_name
- The name of the MigratoryData package.
SSH keys
In order to establish a secure connection to the virtual machines, Ansible requires both public and private SSH keys. These keys can be generated on your local machine using the ssh-keygen command. Once generated, the path to the private key can be specified in the terraform.tfvars
file using the ssh_private_key
variable. Follow the steps below to generate a new SSH key pair:
ssh-keygen -t rsa -b 4096 -C "your_email@example.com"
Executing this command will create a new RSA key pair with a key size of 4096 bits. The keys will be stored in the .ssh
directory of your home folder. The public key will be saved as ~/.ssh/id_rsa.pub
and the private key as ~/.ssh/id_rsa
.
Deploy the infrastructure
Before Terraform can manage any resources, it needs to be initialized. Initialization prepares Terraform for use by downloading the necessary provider plugins:
terraform init
It’s always a good practice to check the deployment plan before applying it. This will give you an overview of the changes that will be made to your infrastructure:
terraform plan
Once you’ve reviewed and are satisfied with the deployment plan, you can apply it:
terraform apply
Installing Ansible Collections
The provided commands are used to install necessary Ansible collections. Ansible collections are a distribution format for Ansible content that can include playbooks, roles, modules, and plugins.
# This command installs the community.general collection, which includes many of the most commonly used modules and other resources for Ansible.
ansible-galaxy collection install community.general
# This collection is designed to interact with Prometheus, a powerful open-source monitoring and alerting toolkit.
ansible-galaxy collection install prometheus.prometheus
# This collection provides modules to interact with Grafana, a popular open-source platform for visualizing metrics, which complements Prometheus.
ansible-galaxy collection install grafana.grafana
By running these commands, you ensure that your Ansible environment has the necessary collections to manage and monitor your infrastructure effectively.
Install MigratoryData server
By running these commands, you initiate the installation of the MigratoryData server on your infrastructure. The ansible/install.yaml
playbook automates the installation process, ensuring a consistent setup across all your servers.
# used to disable SSH host key checking
export ANSIBLE_HOST_KEY_CHECKING=False
ansible-playbook ansible/install.yaml -i artifacts/hosts.ini
Monitoring setup
If the enable_monitoring
variable is set to true, then you can install grafana and prometheus on the monitor virtual machine using the following command:
# For mac you need tar and gnu-tar
brew install gnu-tar
# Fix for MAC OS error `ERROR! A worker was found in a dead state`
export OBJC_DISABLE_INITIALIZE_FORK_SAFETY=YES
# used to disable SSH host key checking
export ANSIBLE_HOST_KEY_CHECKING=False
ansible-playbook ansible/monitor.yaml -i artifacts/hosts.ini
Verify deployment
- You can access the MigratoryData cluster using the public IP address of the load balancer. You can find it running the following command, and find the value of
loadbalancer_ip
in the output:
terraform output
- You can check the health of the cluster by accessing the load balancer public ip on port
80
with path/health
.
http://${loadbalancer_public_ip}/health
- You can also verify the deployment by SSH-ing into the virtual machines. You can do this using the command
ssh admin@machine_public_ip
orssh -i ssh_private_key admin@machine_public_ip
if you need to specify a private key. Once logged in, you can check the status of the MigratoryData service and inspect the logs to ensure everything is running as expected.
sudo su
# Check the status of the MigratoryData service
systemctl status migratorydata
# Inspect the logs
nano /var/log/migratorydata/all/out.log
- To access the monitoring tools, you can use the public IP address of the monitor virtual machine. You can find it running the following command, and find the value of
monitor-public-ip-grafana-access
in the output. By default the username and password for Grafana areadmin
andupdate_password
respectively. To access the MigratoryData dashboard, go to Dashboard -> General in the Grafana UI. For logging, go to Explore and select theLoki
data source.
terraform output
# Access Grafana
http://${monitor-public-ip-grafana-access}:3000
Update MigratoryData server
The package_url
and package_name
variables in the ansible/vars.yaml
file should be updated to point to the new version of the MigratoryData server. The package_url
is the URL where the new version can be downloaded, and the package_name
is the name of the package file.
Update MigratoryData server running the following commands:
# used to disable SSH host key checking
export ANSIBLE_HOST_KEY_CHECKING=False
ansible-playbook ansible/update.yaml -i artifacts/hosts.ini
The ansible/update.yaml
playbook contains tasks that update the MigratoryData server on the machines specified in the hosts.ini
inventory file. The playbook is designed to sequentially update the MigratoryData server on each server in your infrastructure. The playbook operates on one host to ensure service availability during the update process. The tasks executed on each host include stopping the MigratoryData service, downloading the new MigratoryData server package, updating the package, and restarting the service. After updating each server, Ansible pauses for a period of time before moving to the next server, allowing the updated server to fully restart and rejoin the cluster before the next server is updated.
Scale
To scale the deployment, update the num_instances
variable in the terraform.tfvars
file. For example, to scale the deployment to 5 instances, update the num_instances
variable to 5 and run the following commands:
terraform plan
terraform apply
After the infrastructure is created and the new virtual machines are running, you can install the MigratoryData server on the new virtual machines using the following command:
# used to disable SSH host key checking
export ANSIBLE_HOST_KEY_CHECKING=False
ansible-playbook ansible/install.yaml -i artifacts/hosts.ini
If the monitoring is enabled we need to update the monitor virtual machine with the new hosts for the prometheus configuration. Run the following command:
# used to disable SSH host key checking
export ANSIBLE_HOST_KEY_CHECKING=False
ansible-playbook ansible/monitor.yaml -i artifacts/hosts.ini
Uninstall
To destroy the deployment run the following command:
terraform destroy
Build realtime apps
Use any of the MigratoryData’s client APIs to develop real-time applications for communication with this MigratoryData cluster.