Monitoring and Alerts

Metrics and Dashboards

• Prometheus: scrape API metrics on port 9090 (full stack compose maps host 9091)
• Grafana: dashboards under deploy/monitoring/grafana-dashboards/
• API: /api/stats, /api/metrics/* for runtime counters; protected when API auth is enabled

Example query:

curl 'http://localhost:9091/api/v1/query?query=proxywhirl_available_proxies'

Alert Thresholds (starting points)

Tune thresholds per environment and traffic profile.

Logging

ProxyWhirl uses Loguru. Configure level and output via environment:

PROXYWHIRL_LOG_LEVEL=INFO
PROXYWHIRL_LOG_FILE=/var/log/proxywhirl.log

Aggregate logs with your platform standard (CloudWatch, Loki, ELK). Avoid shipping credential-bearing URLs.

Capacity

• Scale horizontally via Kubernetes HPA (deploy/kubernetes/hpa.yaml, Helm HPA template)
• Vertical scaling: adjust CPU/memory requests in deployment manifests
• Load-test before raising concurrency limits; see expanded pytest stress markers for regression probes

Health Scripts

./deploy/scripts/health-check.sh
PROXYWHIRL_API_URL=https://api.example.com PROXYWHIRL_API_KEY=<key> ./deploy/scripts/health-check.sh

See Operations for smoke-test commands.