[Q2-P2] Add synthetic monitoring with continuous SDK health checks

Problem

We only test SDK when:

Someone runs tests manually
CI runs on commits
Customers use it in production

Gap: No continuous validation that SDK works in production

Solution

Add synthetic monitoring that runs realistic SDK queries continuously:

# monitoring/synthetic/sdk_health_check.py

import schedule
import time
from oilpriceapi import OilPriceAPI

def health_check():
    """Run realistic SDK queries."""
    client = OilPriceAPI(api_key=MONITOR_KEY)

    checks = [
        ('current_price', lambda: client.prices.get('WTI_USD')),
        ('1_week_historical', lambda: client.historical.get(
            'WTI_USD',
            (datetime.now() - timedelta(days=7)).date(),
            datetime.now().date()
        )),
        ('1_month_historical', lambda: client.historical.get(
            'WTI_USD',
            (datetime.now() - timedelta(days=30)).date(),
            datetime.now().date()
        )),
    ]

    for name, check in checks:
        try:
            start = time.time()
            result = check()
            duration = time.time() - start

            send_metric(f'synthetic.{name}.success', 1)
            send_metric(f'synthetic.{name}.duration', duration)
        except Exception as e:
            send_metric(f'synthetic.{name}.failure', 1)
            send_alert(f'Synthetic check failed: {name}: {e}')

# Run every 5 minutes
schedule.every(5).minutes.do(health_check)

Deploy

# Deploy as cron job or lambda
apiVersion: batch/v1
kind: CronJob
metadata:
  name: sdk-synthetic-monitoring
spec:
  schedule: "*/5 * * * *"  # Every 5 minutes
  jobTemplate:
    spec:
      template:
        spec:
          containers:
          - name: health-check
            image: oilpriceapi/sdk-monitor:latest
            env:
            - name: MONITOR_API_KEY
              valueFrom:
                secretKeyRef:
                  name: monitoring-secrets
                  key: api-key

Alerts

ALERT: Synthetic check failure rate >10% over 15min
ALERT: Synthetic check duration >2x baseline
ALERT: Synthetic check not running (missing data)

Acceptance Criteria

Synthetic monitoring script created
Deployed and running every 5 minutes
Alerts configured
Dashboard shows synthetic check results
Runbook for responding to failures

Estimated Effort

Time: 4 hours

📝 CodeRabbit Plan Mode

Generate an implementation plan and prompts that you can use with your favorite coding agent.

Create Plan

Examples

_{You can configure auto-planning by selecting labels in the issue_enrichment configuration.}

🔗 Similar Issues

Related Issues

👤 Suggested Assignees

karlwaldman

🧪 Issue enrichment is currently in open beta.

To disable automatic issue enrichment, add the following to your .coderabbit.yaml:

issue_enrichment:
  auto_enrich:
    enabled: false

💬 Have feedback or questions? Drop into our discord or schedule a call!

feat: Complete Q1 & Q2 QA improvements (#20-28)

Completion receipt for PR #74 / c2f6a882b1aa483815b9191e8c83a1a25a7c224f:

replaced the perpetually failing full integration/stress schedule with an hourly, bounded latest-price + five-record history synthetic
missing monitor credentials now fail loudly instead of silently skipping
receipts contain structural assertions and durations but no keys, response values, URLs, bodies, or exception messages
removed the undeployed/broken Docker-Prometheus example that referenced missing configs and printed part of the key
local: 401 non-integration tests, 4 synthetic unit tests, keyless production contract, authenticated synthetic, package build, and docs build passed
main CI: Python 3.8–3.12, Ruff, mypy, storefront validation, unit tests, live API tests, and Pages are green
first real manual run: https://github.com/OilpriceAPI/python-sdk/actions/runs/30169188492 — completed in 21s; latest 0.462s, history 0.417s, both pass
30-day artifact sdk-health-30169188492 exists and expires 2026-08-24
runbook is live at https://oilpriceapi.github.io/python-sdk/SYNTHETIC_MONITORING/

The schedule is hourly rather than every five minutes: this keeps detection bounded while avoiding roughly 8,640 CI starts/month and unnecessary API quota. Push/PR live checks cover changes between scheduled runs.

Navigation Menu

Repository navigation

[Q2-P2] Add synthetic monitoring with continuous SDK health checks #28

Problem

Solution

Deploy

Alerts

Acceptance Criteria

Estimated Effort

📝 CodeRabbit Plan Mode

🔗 Similar Issues

👤 Suggested Assignees

Metadata

Assignees

Labels

Type

Projects

Milestone

Relationships

Development

[Q2-P2] Add synthetic monitoring with continuous SDK health checks #28

Description

Problem

Solution

Deploy

Alerts

Acceptance Criteria

Estimated Effort

Activity

coderabbitai commented on Dec 17, 2025

📝 CodeRabbit Plan Mode

🔗 Similar Issues

👤 Suggested Assignees

karlwaldman commented on Jul 25, 2026

Metadata

Metadata

Assignees

Labels

Type

Projects

Milestone

Relationships

Development

Issue actions