Python

Python is a top choice for DevOps and SRE engineers working across AWS, Azure, and GCP. Its versatility, rich ecosystem, and strong cloud SDK support make it ideal for automating infrastructure, integrating with CI/CD, and building cloud-native solutions.

Why DevOps & SREs Should Learn Python

  1. Cloud Automation: Python SDKs for AWS (boto3), Azure, and GCP enable engineers to automate provisioning, scaling, and management of cloud resources.

  2. IaC & Configuration: Python is used in tools like Ansible, Pulumi, and custom Terraform modules for advanced automation.

  3. CI/CD Integration: Python scripts are common in GitHub Actions, Azure Pipelines, and GitLab CI/CD for deployment, testing, and monitoring.

  4. Observability & Monitoring: Python powers log analysis, custom Prometheus exporters, and alerting integrations.

  5. LLM & AI Integration: Python is the primary language for integrating Large Language Models (LLMs) into DevOps workflows (e.g., using OpenAI, Azure OpenAI, or Hugging Face APIs).

Real-Life Examples

1. AWS EC2 Instance Automation (boto3)

import boto3

ec2 = boto3.resource('ec2')
instance = ec2.create_instances(
    ImageId='ami-0abcdef1234567890',
    MinCount=1,
    MaxCount=1,
    InstanceType='t3.micro',
    KeyName='my-keypair',
    TagSpecifications=[{'ResourceType': 'instance', 'Tags': [{'Key': 'Owner', 'Value': 'devops'}]}]
)
print(f'Launched instance: {instance[0].id}')

2. Azure VM Creation (Azure SDK)

from azure.identity import DefaultAzureCredential
from azure.mgmt.compute import ComputeManagementClient
from azure.mgmt.compute.models import HardwareProfile, NetworkInterfaceReference, \
    OSDisk, StorageAccountTypes, StorageProfile, VirtualHardDisk, VirtualMachine, \
    VirtualMachineSizeTypes, OSProfile, LinuxConfiguration, SshConfiguration, SshPublicKey

credential = DefaultAzureCredential()

compute_client = ComputeManagementClient(
    credential=credential,
    subscription_id='<your-subscription-id>'
)

vm_name = 'my-vm'
location = 'westus2'
rg_name = 'my-resource-group'
vnet_name = 'my-vnet'
subnet_name = 'my-subnet'
username = 'my-username'
password = 'my-password'
ssh_key = 'ssh-rsa AAAAB3NzaC1yc2EAAAADAQABAAACAQCx...'

vm_parameters = VirtualMachine(
    location=location,
    hardware_profile=HardwareProfile(vm_size=VirtualMachineSizeTypes.standard_b1s),
    storage_profile=StorageProfile(
        image_reference={
            'publisher': 'Canonical',
            'offer': 'UbuntuServer',
            'sku': '20.04-LTS',
            'version': 'latest'
        },
        os_disk=OSDisk(
            create_option='fromImage',
            managed_disk=VirtualHardDisk(storage_account_type=StorageAccountTypes.standard_lrs)
        )
    ),
    os_profile=OSProfile(
        computer_name=vm_name,
        admin_username=username,
        admin_password=password,
        linux_configuration=LinuxConfiguration(
            disable_password_authentication=True,
            ssh=SshConfiguration(
                public_keys=[
                    SshPublicKey(
                        path='/home/{}/.ssh/authorized_keys'.format(username),
                        key_data=ssh_key
                    )
                ]
            )
        )
    ),
    network_profile={
        'network_interfaces': [
            {
                'id': compute_client.network_interfaces.get(
                    resource_group_name=rg_name,
                    network_interface_name='<your-network-interface-name>'
                ).id,
                'properties': {
                    'primary': True
                }
            }
        ]
    }
)

compute_client.virtual_machines.create_or_update(
    resource_group_name=rg_name,
    vm_name=vm_name,
    parameters=vm_parameters
)

3. GCP Cloud Storage Bucket Creation (google-cloud-storage)

from google.cloud import storage

client = storage.Client()
bucket = client.create_bucket('my-devops-bucket-2025')
print(f'Created bucket: {bucket.name}')

4. Integrating LLMs for Automated Change Summaries

import openai

openai.api_key = 'sk-...'
response = openai.ChatCompletion.create(
    model='gpt-4',
    messages=[
        {"role": "system", "content": "Summarize this deployment change log for SREs."},
        {"role": "user", "content": open('changelog.txt').read()}
    ]
)
print(response['choices'][0]['message']['content'])

5. CI/CD Pipeline Step (GitHub Actions)

- name: Run Python deployment script
  run: |
    python deploy.py --env=prod

Best Practices (2025)

  • Use virtual environments (venv, pipenv) for dependency management

  • Store secrets securely (AWS Secrets Manager, Azure Key Vault, GCP Secret Manager)

  • Write idempotent scripts for infrastructure changes

  • Integrate Python scripts with CI/CD for repeatable automation

  • Use logging and exception handling for observability

  • Test automation code with pytest or unittest

Common Pitfalls

  • Hardcoding credentials in scripts

  • Not handling API rate limits or errors

  • Ignoring dependency pinning (requirements.txt)

  • Lack of logging and monitoring

  • Not using version control for automation scripts

References

Python Joke: Why did the DevOps engineer love Python scripts? Because they always passed the test—unless they were indented wrong!

PreviousYQ the toolNextGo

Last updated