lessandro-ugulino / grafana-prometheus-for-fargate Goto Github PK

• Introduction
• Diagram
• Requirements
• Ansible
  • Playbook Command
• AWS
  • Network
  • ECS Cluster
  • AWS EFS
  • ECS Tasks
  • Domain
• Application
  • Web Console
  • ECS Auto Scaling
• Destroy
  • Via Ansible
  • Via CloudFormation

This pipeline via automation will deploy Prometheus and Grafana on an AWS Fargate also the core infrastructure to support it.

Grafana - Grafana is a multi-platform open source analytics and interactive visualization web application.

Prometheus - Prometheus is a free software application used for event monitoring and alerting.

The below diagram shows the AWS resources.

1. These tools were installed:

• AWS Cli
• Ansible

2. Update the file group_vars/all/common.yml accordingly.

---
cloudformation:
  check_mode_override: false
  region: ap-southeast-2
  create_changeset: true
  state: present

Route53:
  HostedZoneName: "ugulino.com."
  RecordName: "monitor.ugulino.com"
  CertificateId: "5eaa197e" # your ACM Id

The file deploy-role.yml will deploy each role required to deploy this project.

---
- hosts: localhost
  roles:
    - vpc-baseline
    - ecs-cluster
    - ecs-monitor
    - route53

The below command will execute the tasks defined on the above file.

ansible-playbook deploy-role.yml -vvv

Output

You can verify the deployment via ansible output or via AWS CloudFormation console.

• Ansible Output:

• AWS CloudFormation Console:

An 'A' record was created pointing to Application Load Balancer via automation.

You can access and setup Grafana accessing the URL https://monitor.ugulino.com (be aware that this is my domain, probably you've setup a different one)

Default username and password is admin

Setup the Data Source for prometheus.

EC2 REQUIREMENTS:

• The EC2 instances taged as below
• The application node-exporter installed on the EC2.
• Allow Inbound and Outbound port 9100.

Grafana Dashboard Id 1860

The below prometheus.yml was used for this environment. Upload this file to AWS EFS.

################-----//-----################

# USE THIS FILE AS AN EXAMPLE.

################-----//-----################

# my global config
global:
  scrape_interval: 15s # Set the scrape interval to every 15 seconds. Default is every 1 minute.
  evaluation_interval: 15s # Evaluate rules every 15 seconds. The default is every 1 minute.

scrape_configs:
  - job_name: "ec2-nodes"
    ec2_sd_configs:
      - region: ap-southeast-2
        port: 9100

    relabel_configs:
      - source_labels: [__meta_ec2_tag_prometheus]
        regex: true.*
        action: keep

      - source_labels: [__meta_ec2_tag_Name]
        target_label: instance

Basically, to destroy this deployment you can either do it via ansible command or AWS CloudFormation.

Change the file group_vars/all/common.yml

cloudformation:
  check_mode_override: false
  region: ap-southeast-2
  create_changeset: true
  state: absent # this is the value you need to change to destroy the stacks

Execute the command:

ansible-playbook deploy-role.yml -vvv

Select the CloudFormation stack, click delete, and then Delete stack.