1 min read
Azure Security: A Year in Review
Major Security Releases in 2022
Microsoft Defender for Cloud
defender_cloud_2022:
new_features:
- Defender CSPM (Cloud Security Posture Management)
- Attack path analysis
- Cloud security graph
- Agentless scanning for VMs
- Defender for DevOps
- Governance capabilities
enhanced_workload_protection:
- Defender for Containers improvements
- Defender for Storage malware scanning
- Defender for Azure Cosmos DB
- Defender for Resource Manager
compliance:
- New regulatory compliance standards
- Custom recommendations
- Continuous export to Event Hub
Zero Trust Implementation
from dataclasses import dataclass
from typing import List, Dict
from enum import Enum
class TrustLevel(Enum):
NONE = 0
LOW = 1
MEDIUM = 2
HIGH = 3
@dataclass
class ZeroTrustAssessment:
identity_score: float
device_score: float
network_score: float
application_score: float
data_score: float
@property
def overall_score(self) -> float:
weights = {
"identity": 0.25,
"device": 0.20,
"network": 0.15,
"application": 0.20,
"data": 0.20
}
return (
self.identity_score * weights["identity"] +
self.device_score * weights["device"] +
self.network_score * weights["network"] +
self.application_score * weights["application"] +
self.data_score * weights["data"]
)
def get_recommendations(self) -> List[str]:
recommendations = []
if self.identity_score < 0.7:
recommendations.extend([
"Enable MFA for all users",
"Implement Conditional Access policies",
"Deploy Privileged Identity Management",
"Enable risk-based sign-in policies"
])
if self.device_score < 0.7:
recommendations.extend([
"Require device compliance",
"Implement Microsoft Intune",
"Enable device health attestation",
"Deploy endpoint detection and response"
])
if self.network_score < 0.7:
recommendations.extend([
"Implement network segmentation",
"Deploy Azure Firewall Premium",
"Enable DDoS Protection",
"Use Private Endpoints"
])
if self.application_score < 0.7:
recommendations.extend([
"Implement API Management",
"Use Azure Front Door with WAF",
"Enable application-level encryption",
"Deploy Defender for App Service"
])
if self.data_score < 0.7:
recommendations.extend([
"Enable Azure Information Protection",
"Implement data classification",
"Use customer-managed keys",
"Enable soft delete and versioning"
])
return recommendations
def assess_zero_trust_maturity(subscription_id: str) -> ZeroTrustAssessment:
"""Assess Zero Trust implementation maturity."""
# In production, these would query actual Azure resources
assessment = ZeroTrustAssessment(
identity_score=evaluate_identity_controls(subscription_id),
device_score=evaluate_device_controls(subscription_id),
network_score=evaluate_network_controls(subscription_id),
application_score=evaluate_application_controls(subscription_id),
data_score=evaluate_data_controls(subscription_id)
)
return assessment
Security Automation
from azure.mgmt.security import SecurityCenter
from azure.identity import DefaultAzureCredential
class SecurityAutomation:
def __init__(self, subscription_id: str):
credential = DefaultAzureCredential()
self.client = SecurityCenter(credential, subscription_id)
self.subscription_id = subscription_id
def create_security_automation(
self,
name: str,
resource_group: str,
trigger_severity: str,
action_type: str,
action_target: str
):
"""Create automated response to security alerts."""
automation = {
"location": "global",
"properties": {
"description": f"Auto-respond to {trigger_severity} alerts",
"isEnabled": True,
"scopes": [
{
"scopePath": f"/subscriptions/{self.subscription_id}"
}
],
"sources": [
{
"eventSource": "Alerts",
"ruleSets": [
{
"rules": [
{
"propertyJPath": "Severity",
"propertyType": "String",
"expectedValue": trigger_severity,
"operator": "Equals"
}
]
}
]
}
],
"actions": [
{
"actionType": action_type,
"logicAppResourceId": action_target if action_type == "LogicApp" else None,
"uri": action_target if action_type == "EventHub" else None
}
]
}
}
return self.client.automations.create_or_update(
resource_group,
name,
automation
)
def get_security_score(self) -> Dict:
"""Get current security score and recommendations."""
secure_scores = list(self.client.secure_scores.list())
results = []
for score in secure_scores:
controls = list(
self.client.secure_score_controls.list_by_secure_score(
score.name
)
)
results.append({
"name": score.name,
"current_score": score.current_score,
"max_score": score.max_score,
"percentage": score.percentage,
"controls": [
{
"name": c.display_name,
"current": c.current_score,
"max": c.max_score,
"healthy_resources": c.healthy_resource_count,
"unhealthy_resources": c.unhealthy_resource_count
}
for c in controls
]
})
return results
Key Vault Best Practices
// Secure Key Vault deployment
resource keyVault 'Microsoft.KeyVault/vaults@2022-07-01' = {
name: 'kv-${environment}-${uniqueString(resourceGroup().id)}'
location: location
properties: {
sku: {
family: 'A'
name: 'premium' // HSM-backed keys
}
tenantId: tenant().tenantId
// Network security
networkAcls: {
defaultAction: 'Deny'
bypass: 'AzureServices'
ipRules: [
{
value: allowedIpRange
}
]
virtualNetworkRules: [
{
id: subnet.id
}
]
}
// Access configuration
enableRbacAuthorization: true // Use RBAC instead of access policies
enableSoftDelete: true
softDeleteRetentionInDays: 90
enablePurgeProtection: true
// Security features
enabledForDeployment: false
enabledForDiskEncryption: true
enabledForTemplateDeployment: false
publicNetworkAccess: 'Disabled' // Private endpoint only
}
}
// Private endpoint for Key Vault
resource privateEndpoint 'Microsoft.Network/privateEndpoints@2022-07-01' = {
name: 'pe-${keyVault.name}'
location: location
properties: {
subnet: {
id: privateEndpointSubnet.id
}
privateLinkServiceConnections: [
{
name: 'kv-connection'
properties: {
privateLinkServiceId: keyVault.id
groupIds: ['vault']
}
}
]
}
}
// Diagnostic settings
resource diagnostics 'Microsoft.Insights/diagnosticSettings@2021-05-01-preview' = {
name: 'kv-diagnostics'
scope: keyVault
properties: {
workspaceId: logAnalyticsWorkspace.id
logs: [
{
category: 'AuditEvent'
enabled: true
retentionPolicy: {
enabled: true
days: 365
}
}
]
metrics: [
{
category: 'AllMetrics'
enabled: true
retentionPolicy: {
enabled: true
days: 365
}
}
]
}
}
Security Lessons from 2022
security_lessons_2022:
supply_chain_security:
lesson: "Software supply chain attacks increased significantly"
actions:
- Implement SBOM (Software Bill of Materials)
- Use signed container images
- Enable GitHub Advanced Security
- Scan dependencies with Dependabot
identity_attacks:
lesson: "Identity remains the primary attack vector"
actions:
- Require phishing-resistant MFA
- Implement Conditional Access
- Monitor for impossible travel
- Enable Identity Protection
api_security:
lesson: "API vulnerabilities are increasingly exploited"
actions:
- Implement API Management
- Use OAuth 2.0 properly
- Rate limit all endpoints
- Validate all inputs
ransomware:
lesson: "Ransomware continues to evolve"
actions:
- Implement immutable backups
- Use Azure Backup with soft delete
- Enable Defender for Storage
- Practice recovery procedures
Security Checklist for 2023
security_checklist_2023 = {
"identity": [
"Passwordless authentication rollout",
"Phishing-resistant MFA (FIDO2, Windows Hello)",
"Continuous access evaluation",
"Workload identity for service accounts"
],
"data": [
"Data classification and labeling",
"Customer-managed keys for sensitive data",
"Confidential computing for high-security workloads",
"Data loss prevention policies"
],
"network": [
"Private endpoints for all PaaS services",
"Azure Firewall Premium with IDPS",
"Web Application Firewall 2.0",
"DDoS Protection Standard"
],
"applications": [
"Defender for DevOps integration",
"Container image signing",
"API security scanning",
"Runtime application self-protection"
],
"operations": [
"Security automation with Logic Apps",
"SOAR integration",
"Threat intelligence feeds",
"Regular penetration testing"
]
}
Conclusion
2022 brought significant security advancements in Azure, from Defender for Cloud improvements to enhanced Zero Trust capabilities. As we move into 2023, focus on identity-first security, assume breach mentality, and automate security responses. The threat landscape will continue to evolve, but so will our defenses.