如何使用Ansible在AWS创建Network Load Balancer？能否借助Boto3实现？

Absolutely! Even though Ansible doesn’t have a dedicated Network Load Balancer (NLB) module at the moment, you can fully implement NLB creation in AWS using Boto3—AWS’s official Python SDK—right within your Ansible playbooks. Below are two straightforward approaches to get this done:

Approach 1: Use the script Module to Run an External Boto3 Script #

This method involves writing a standalone Python script that uses Boto3 to create the NLB, then calling that script from your playbook.

Step 1: Write the Boto3 Script (create_nlb.py) #

import boto3

def create_network_load_balancer():
    # Initialize the ELBv2 client (NLBs use the ELBv2 API)
    elbv2_client = boto3.client('elbv2', region_name='us-east-1')  # Update your region here

    # Create the NLB
    nlb_response = elbv2_client.create_load_balancer(
        Name='my-production-nlb',
        Subnets=['subnet-0123456789abcdef0', 'subnet-0abcdef1234567890'],  # Replace with your subnets
        Scheme='internet-facing',  # Use 'internal' for private NLBs
        Type='network',
        IpAddressType='ipv4'
    )

    # Create a target group for the NLB
    target_group_response = elbv2_client.create_target_group(
        Name='my-nlb-target-group',
        Protocol='TCP',
        Port=80,
        VpcId='vpc-0123456789abcdef0',  # Replace with your VPC ID
        TargetType='instance'  # Can also be 'ip' or 'lambda'
    )

    # Attach a listener to forward traffic to the target group
    elbv2_client.create_listener(
        LoadBalancerArn=nlb_response['LoadBalancers'][0]['LoadBalancerArn'],
        Protocol='TCP',
        Port=80,
        DefaultActions=[
            {
                'Type': 'forward',
                'TargetGroupArn': target_group_response['TargetGroups'][0]['TargetGroupArn']
            }
        ]
    )

    print(f"NLB created successfully! ARN: {nlb_response['LoadBalancers'][0]['LoadBalancerArn']}")

if __name__ == "__main__":
    create_network_load_balancer()

Step 2: Call the Script from Your Ansible Playbook #

- name: Deploy AWS NLB using Boto3 script
  hosts: localhost
  gather_facts: false
  vars:
    aws_access_key: "{{ your_aws_access_key }}"
    aws_secret_key: "{{ your_aws_secret_key }}"
    aws_region: us-east-1
  tasks:
    - name: Execute NLB creation script
      ansible.builtin.script: create_nlb.py
      environment:
        AWS_ACCESS_KEY_ID: "{{ aws_access_key }}"
        AWS_SECRET_ACCESS_KEY: "{{ aws_secret_key }}"
        AWS_DEFAULT_REGION: "{{ aws_region }}"

Approach 2: Use the python Module for Inline Boto3 Code #

If you prefer to keep everything within the playbook without external files, you can run Boto3 code directly using Ansible’s python module:

- name: Create AWS NLB with inline Boto3 code
  hosts: localhost
  gather_facts: false
  vars:
    aws_region: us-east-1
    nlb_name: my-inline-nlb
    subnet_ids: ['subnet-0123456789abcdef0', 'subnet-0abcdef1234567890']
    vpc_id: vpc-0123456789abcdef0
  tasks:
    - name: Run inline Boto3 logic to create NLB
      ansible.builtin.python:
        code: |
          import boto3
          import json

          # Initialize ELBv2 client
          elbv2 = boto3.client('elbv2', region_name='{{ aws_region }}')

          # Create NLB
          nlb = elbv2.create_load_balancer(
              Name='{{ nlb_name }}',
              Subnets={{ subnet_ids }},
              Scheme='internet-facing',
              Type='network',
              IpAddressType='ipv4'
          )

          # Create target group
          target_group = elbv2.create_target_group(
              Name='{{ nlb_name }}-tg',
              Protocol='TCP',
              Port=80,
              VpcId='{{ vpc_id }}',
              TargetType='instance'
          )

          # Attach listener
          elbv2.create_listener(
              LoadBalancerArn=nlb['LoadBalancers'][0]['LoadBalancerArn'],
              Protocol='TCP',
              Port=80,
              DefaultActions=[{'Type': 'forward', 'TargetGroupArn': target_group['TargetGroups'][0]['TargetGroupArn']}]
          )

          # Return structured result
          output = {
              'changed': True,
              'nlb_arn': nlb['LoadBalancers'][0]['LoadBalancerArn'],
              'target_group_arn': target_group['TargetGroups'][0]['TargetGroupArn']
          }
          print(json.dumps(output))
      register: nlb_output

    - name: Print created NLB details
      ansible.builtin.debug:
        msg: |
          NLB ARN: {{ (nlb_output.stdout | from_json).nlb_arn }}
          Target Group ARN: {{ (nlb_output.stdout | from_json).target_group_arn }}

Key Notes #

• Prerequisites: Ensure the machine running Ansible has Boto3 installed (pip install boto3). You’ll also need valid AWS credentials configured (either via environment variables, ~/.aws/credentials, or IAM roles if running on EC2).
• Permissions: The AWS credentials used need permissions like elasticloadbalancing:CreateLoadBalancer, elasticloadbalancing:CreateTargetGroup, and elasticloadbalancing:CreateListener.
• Customization: Adjust parameters like subnet IDs, VPC ID, ports, protocols, and NLB scheme (public/private) to match your infrastructure needs.

内容的提问来源于stack exchange，提问作者Elieyal