QI3IX Global CDN & Traffic Management Infrastructure
Azure Front Door is Microsoft's modern cloud Content Delivery Network (CDN) that provides fast, reliable, and secure access to our global web applications. It acts as the single global entry point for all 54 of our custom domains, routing traffic to the appropriate backend services based on intelligent rules and health monitoring.
Think of Front Door as a sophisticated traffic director at a massive intersection. When users visit any of our domains (like www.drantech.ai or www.celljourney.com), Front Door receives that request first, determines the best backend server to handle it, ensures that server is healthy, and then routes the traffic accordingly — all in milliseconds.
Front Door Profile
Custom Domains
Origin Groups
Production Apps
www.drantech.ai
SSL Termination
Health Check
Route Matching
drantechai-backend
Dranschak-East2-App
(Flask Application)
If a backend's /api/health endpoint returns anything other than HTTP 200 OK, Front Door marks that origin as UNHEALTHY and stops routing traffic to it. This causes 503 Service Unavailable errors for all domains routed to that origin group. Every Flask app MUST have a functioning /api/health endpoint.
DranTechAI Developer Hub
CellJourney Biotech Platform
EnableYou QSPECTIVE Portal
Laughing Uphill Adventure Hub
QSPECTIVE AI Platform
BlockifyAI Services
Family Platform
The following domains are registered and configured in Front Door but serve a professional "Coming Soon" holding
page. They route to the holding-page-origin-group origin group.
Strategic reserve
Strategic reserve
Strategic reserve
Strategic reserve
Strategic reserve
Strategic reserve
See AZURE_FRONTDOOR_INVENTORY.md for complete list
Origin groups are logical collections of backend servers. Front Door routes traffic to origin groups, which then distribute requests across their configured origins (App Services or Static Web Apps). Each origin group includes health probe settings to monitor backend availability.
These origin groups serve Azure Static Web Apps. Most are internal development environments or specialized
deployment targets. They typically use / as the health probe path (default for static sites).
| Origin Group Name | Static App | Health Probe | Purpose |
|---|---|---|---|
| sarcastic-mushroom | sarcastic-mushroom-*.azurestaticapps.net | /api/health | Deployment environment |
| red-beach | red-beach-*.azurestaticapps.net | /api/health | Deployment environment |
| red-mushroom | red-mushroom-*.azurestaticapps.net | /api/health | Deployment environment |
| novadyme | novadyme-*.azurestaticapps.net | /api/health | NovaDyme platform |
| happy-dune | happy-dune-*.azurestaticapps.net | / (needs update) | Deployment environment |
| a-lucky-fig | a-lucky-fig-*.azurestaticapps.net | / (needs update) | Deployment environment |
| five-why | five-why-*.azurestaticapps.net | / (needs update) | 5Why platform |
| gray-dune | gray-dune-*.azurestaticapps.net | / (needs update) | Temporary holding page host |
| green-rock | green-rock-*.azurestaticapps.net | / (needs update) | Deployment environment |
Six origin groups (happy-dune, a-lucky-fig, five-why, gray-dune, green-rock, white-sky) currently use
/ as the health probe path. These should be updated to /api/health for consistency
and better monitoring. The configuration script
scripts/azure/configure_frontdoor_complete.ps1 handles these updates automatically.
QI3IX uses a one route per domain architecture. Each of the 54 custom domains has a dedicated route that maps the domain to its designated origin group. This provides clear separation, easier debugging, and straightforward configuration management.
Route Name Pattern: [domain-name]-route
Example: drantechai-route
Route Settings:
├─ Patterns to match: www.drantech.ai/*
├─ Origin group: drantechai-backend
├─ Forwarding protocol: HTTPS only
├─ Query string caching: Ignore query string
├─ Dynamic compression: Enabled
└─ HTTPS redirect: Enabled (301 permanent redirect)
| Route Name | Domain Pattern | Origin Group | Backend Type |
|---|---|---|---|
| drantechai-route | www.drantech.ai/* | drantechai-backend | Flask App Service |
| celljourney-route | www.celljourney.com/* | celljourney-backend | Flask App Service |
| enableyou-route | www.enableyou.ai/* | enableyou-backend | Flask App Service |
| laughinguphill-route | www.laughinguphill.com/* | laughinguphill-backend | Flask App Service |
| landjglobal-route | landjglobal.com/* | holding-page-origin-group | Static Web App |
| + 49 more routes (see complete inventory) | |||
Azure Front Door continuously monitors backend health to ensure traffic is only routed to healthy servers. Every 30 seconds, Front Door sends HTTP GET requests to the configured health probe path. Backends must respond with HTTP 200 OK within the timeout period to be considered healthy.
Health Probe Settings (All Origin Groups):
├─ Path: /api/health
├─ Protocol: HTTP (not HTTPS)
├─ Method: GET or HEAD
├─ Interval: 30 seconds
├─ Timeout: 5 seconds
├─ Successful samples required: 2-3
└─ Unhealthy threshold: 2 consecutive failures
Why HTTP instead of HTTPS?
• Front Door terminates SSL at the edge
• Backend probes use internal Azure network (no SSL needed)
• Faster probe responses (no SSL handshake)
• Microsoft official recommendation for App Service origins
@app.route("/api/health", methods=["GET", "HEAD"])
def api_health():
"""
Front Door health probe endpoint.
CRITICAL: Must always return 200 OK for Front Door to route traffic.
- Do NOT add authentication
- Do NOT add database checks
- Do NOT add heavy logic
- Must be fast (< 500ms)
"""
return "OK", 200
Front Door will serve 503 Service Unavailable if:
# Check origin health status via Azure CLI
az afd origin show \
--resource-group Corporate \
--profile-name qi3ix-global-frontdoor \
--origin-group-name drantechai-backend \
--origin-name dranschak-east2-app \
--query "healthProbeSettings"
# Expected response for healthy origin:
{
"probePath": "/api/health",
"probeProtocol": "Http",
"probeRequestType": "GET",
"probeIntervalInSeconds": 30
}
QI3IX maintains automated PowerShell scripts for Front Door configuration and auditing:
This script handles all Front Door configuration in three phases: (1) updates 6 origin groups to use /api/health, (2) creates holding-page-origin-group, (3) creates 54 individual routes with domain mappings.
Root Cause: Front Door marked origin unhealthy
Resolution Steps:
curl https://[app].azurewebsites.net/api/healthRoot Cause: Front Door hasn't completed successful health probe yet
Resolution: Wait 30-60 seconds for Front Door probe interval, then verify origin health in Azure Portal
Azure Front Door operates from 100+ edge locations worldwide. When a user requests content, Front Door routes the request to the nearest edge location, significantly reducing latency. Static assets (images, CSS, JavaScript) are cached at the edge, reducing load on backend servers and improving response times.
Edge Locations Worldwide
Average Edge Latency
Cache Hit Ratio
SLA Uptime Guarantee
Front Door manages SSL certificates for all 54 custom domains automatically. Certificates are provisioned via Azure's partnership with DigiCert and auto-renew before expiration. This eliminates manual certificate management overhead.
Azure Front Door Premium SKU includes Azure DDoS Protection Standard at no additional cost. This provides always-on traffic monitoring and automatic mitigation of common network-layer attacks (SYN floods, UDP floods, reflection attacks).
The Premium SKU supports Azure WAF for application-layer protection (SQL injection, XSS, etc.). This can be enabled in future phases as traffic patterns and security requirements evolve.
| Component | Pricing Model | Estimated Monthly Cost |
|---|---|---|
| Front Door Profile (Premium) | Fixed monthly fee | ~$35 base |
| Data Transfer (Outbound) | Per GB transferred to internet | Varies by traffic volume |
| Custom Domains (54 total) | Included (no per-domain fee) | $0 |
| SSL Certificates (54 total) | Included (Azure-managed certs) | $0 |
| Health Probes | Included in base fee | $0 |
Complete inventory of 54 domains and 14 origin groups
Detailed implementation guide with verification steps
Quick reference guide for common operations
Deployment rules and health probe requirements