Uptime Monitoring

​

What is Uptime Monitoring?

• HTTP checks: Verify your site responds correctly
• Response time tracking: Monitor TTFB and total request time
• SSL certificate monitoring: Track expiration dates
• Status code validation: Ensure proper HTTP responses
• Geographic probing: Checks from multiple regions
• Downtime detection: Alert on failures and outages

​

How It Works

1. Schedule creation: You define check frequency and URL
2. Distributed execution: QStash triggers checks from edge locations
3. HTTP probe: Databuddy’s uptime service fetches your URL
4. Data collection: Records response time, status, SSL info
5. Event streaming: Sends results to Kafka → ClickHouse
6. Analytics: View uptime metrics in dashboard

​

Setting Up Monitors

​

Via Dashboard

1. Navigate to Uptime in the sidebar
2. Click Create Monitor
3. Enter the URL to monitor
4. Configure check settings
5. Click Create

​

Monitor Configuration

​

Example: Basic Monitor

{
  "url": "https://example.com",
  "checkInterval": 5,
  "timeout": 30000,
  "cacheBust": false
}

​

Example: Advanced Monitor

{
  "url": "https://api.example.com/health",
  "checkInterval": 1,
  "timeout": 10000,
  "cacheBust": true,
  "websiteId": "website-123"
}

​

Check Intervals

• 1 minute: Critical services, APIs
• 5 minutes: Standard websites (recommended)
• 15 minutes: Low-priority sites
• 30 minutes: Backup monitoring
• 1 hour: Cost-effective monitoring

​

Timeout Configuration

• 10 seconds: Fast APIs, CDN-backed sites
• 30 seconds: Default (recommended)
• 60 seconds: Slow backends, heavy pages

​

Cache Busting

• Disabled (default): Standard checks
• Enabled: Adds ?_cb=<random> to URL

• CDN-cached pages
• Aggressive browser caching
• Testing origin server health

• Standard monitoring
• APIs that don’t cache
• Conserving bandwidth

​

Monitored Metrics

​

Response Metrics

• HTTP Status Code: 200, 404, 500, etc.
• TTFB (Time to First Byte): Server response time in ms
• Total Time: Complete request duration in ms
• Response Size: Bytes received
• Content Hash: SHA-256 hash for change detection

​

SSL Metrics

• Certificate Expiry: Unix timestamp of expiration
• Certificate Valid: Boolean indicating validity

​

Network Metrics

• Redirect Count: Number of HTTP redirects followed
• Probe Region: Geographic location of check
• Probe IP: IP address of monitoring probe

​

Status

• UP: Successful check (2xx or 3xx status)
• DOWN: Failed check (4xx, 5xx, timeout, or network error)

​

SSL Certificate Monitoring

​

Certificate Checks

• Expiration date: When certificate expires
• Validity: Whether certificate is currently valid
• Issuer: Certificate authority (extracted from cert)

​

Expiration Alerts

• 30 days before expiration
• 14 days before expiration
• 7 days before expiration
• Day of expiration

​

HTTP Status Handling

​

Success (UP)

• 2xx: OK (200, 201, 204, etc.)
• 3xx: Redirects (301, 302, 307, 308)

​

Failure (DOWN)

• 4xx: Client errors (404, 403, etc.)
• 5xx: Server errors (500, 502, 503, etc.)
• 0: Timeout or network error

​

Viewing Uptime Data

​

Monitor Dashboard

• Current status (UP/DOWN)
• Last check time
• Uptime percentage (24h, 7d, 30d)
• Average response time
• Recent incidents

​

Monitor Details

• Uptime graph: Visual timeline of status
• Response time chart: TTFB and total time trends
• Incident history: Downtime events with duration
• SSL status: Certificate expiration countdown
• Geographic performance: Response times by region

​

Metrics Over Time

• Hourly: Last 24-48 hours
• Daily: Last 30-90 days
• Weekly: Last 6-12 months

​

Content Change Detection

• Content hash: SHA-256 hash of response body
• Change detection: Alerts when hash changes
• False positive filtering: Ignore dynamic timestamps, ads

• Detect defacement
• Monitor for unauthorized changes
• Track deployment success

​

JSON Response Parsing

​

JSON Parsing Config

{
  "jsonParsingConfig": {
    "enabled": true,
    "path": "$.data.status",
    "expectedValue": "healthy"
  }
}

​

Use Cases

• API health endpoints
• Microservice status pages
• Custom validation logic
• Extract metrics from responses

​

Failure Handling

​

Retry Logic

1. Initial check fails
2. Wait 30 seconds
3. Retry (up to 3 times)
4. Mark as DOWN if all retries fail

​

Failure Streaks

• Streak count: Number of consecutive DOWN checks
• Alert threshold: Notify after 2-3 consecutive failures
• Recovery: Streak resets on first UP check

​

Infrastructure

• Runtime: Bun + Elysia
• Scheduler: Upstash QStash (distributed cron)
• Database: PostgreSQL (monitor configs)
• Analytics: Kafka + ClickHouse (check results)
• Observability: OpenTelemetry → Axiom

​

Request Flow

QStash schedule trigger
  ↓
Uptime service receives webhook
  ↓
Verify QStash signature
  ↓
Lookup monitor config (PostgreSQL)
  ↓
Perform HTTP check with timeout
  ↓
Check SSL certificate (if HTTPS)
  ↓
Get probe metadata (IP, region)
  ↓
Send event to Kafka
  ↓
ClickHouse ingestion
  ↓
Dashboard analytics

​

Security

​

Webhook Verification

const isValid = await receiver.verify({
  body,
  signature,
  url: process.env.UPTIME_URL
})

​

Request Headers

User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) ...
Accept: text/html,application/xhtml+xml,...
Accept-Encoding: gzip, deflate, br
Cache-Control: no-cache

​

Compression Support

• Tries gzip, deflate, br first
• Falls back to gzip, deflate on encoding errors
• Handles corrupt compression gracefully

​

Best Practices

​

Monitor Critical Pages

1. Homepage
2. Login/signup pages
3. Checkout/payment flows
4. API endpoints
5. Critical content pages

​

Choose Appropriate Intervals

• Mission-critical: 1-minute checks
• Production websites: 5-minute checks
• Staging/dev: 15-minute checks
• Backup monitors: 30-minute checks

​

Set Realistic Timeouts

• Fast sites: 5-10 seconds
• Average sites: 30 seconds (default)
• Slow backends: 60 seconds

​

Monitor SSL Expiration

• Enable certificate monitoring
• Set up 30-day expiration alerts
• Use Let’s Encrypt auto-renewal where possible

​

Use JSON Parsing for APIs

• Check response structure
• Validate specific fields
• Monitor API contract compliance

​

Alerts & Notifications

​

Alert Channels

• Email
• Slack
• Discord
• Webhook (custom integrations)
• SMS (via Twilio)

​

Alert Conditions

• Site goes DOWN (after retry logic)
• Site recovers (returns UP)
• SSL certificate expires in less than 30 days
• Response time exceeds threshold
• Content hash changes unexpectedly

​

Uptime SLA Calculation

Uptime % = (Successful Checks / Total Checks) × 100

​

Example

• Total checks: 288 (24h × 12 checks/hour)
• Failed checks: 3
• Uptime: (285 / 288) × 100 = 98.96%

​

Industry Standards

• 99.9% (“three nines”): 43 minutes downtime/month
• 99.95%: 22 minutes downtime/month
• 99.99% (“four nines”): 4 minutes downtime/month

​

Exporting Uptime Data

1. Open monitor details
2. Select date range
3. Click Export
4. Choose format (CSV or JSON)
5. Download data

​

Next Steps