Caddy Integration
Automatic HTTPS monitoring, request metrics, and reverse proxy latency tracking for Caddy. Monitor certificate renewals, upstream health, and request throughput with AI anomaly detection.
How It Works
Enable Caddy Admin API
Enable the Caddy admin API in your Caddyfile (admin localhost:2019). The TigerOps agent reads Caddy metrics and configuration state from the admin endpoint.
Configure Prometheus Metrics
Add the prometheus metrics handler to your Caddyfile global block. This exposes /metrics on port 2019 with Caddy request, TLS, and runtime metrics in Prometheus format.
Install TigerOps Agent
Deploy the TigerOps agent on your Caddy host. It scrapes Caddy Prometheus metrics, reads access logs, and monitors certificate expiry for all configured domains automatically.
Alert on Upstreams and Certs
Set alerts for upstream health failures, p99 proxy latency, and certificate expiry windows. TigerOps correlates Caddy upstream errors with backend service anomalies automatically.
What You Get Out of the Box
Automatic HTTPS Certificate Tracking
Monitor ACME certificate issuance, renewal attempts, and expiry for all domains managed by Caddy. TigerOps alerts 30 and 7 days before expiry and tracks renewal failure rates.
Reverse Proxy Upstream Latency
Track p50, p95, and p99 upstream response times per backend. TigerOps detects slow upstream regressions in real time and correlates them with Caddy error rate spikes.
Request Rate and Status Code Distribution
Monitor requests per second, HTTP status code distribution (2xx/3xx/4xx/5xx), and error rate trends per site block. AI anomaly detection fires on unusual error rate patterns.
TLS Handshake Metrics
Track TLS 1.2 vs. TLS 1.3 handshake rates, cipher suite distribution, and handshake latency. Detect TLS configuration issues before they impact user experience.
Load Balancer Health Checks
When using Caddy as a load balancer, TigerOps tracks upstream health check pass/fail rates, active connection counts per upstream, and lb_policy effectiveness.
Caddyfile Configuration Change Tracking
TigerOps records Caddy configuration reload events from the admin API. Correlate configuration changes with traffic pattern shifts and error rate changes on a unified timeline.
Caddyfile with Prometheus Metrics
Enable Prometheus metrics and the admin API in your Caddyfile for TigerOps to scrape.
{
# Enable admin API for TigerOps agent
admin localhost:2019
# Enable Prometheus metrics
servers {
metrics
}
}
# Your site blocks
example.com {
reverse_proxy /api/* backend:8080 {
health_uri /healthz
health_interval 10s
health_timeout 5s
}
# Access log for TigerOps log ingestion
log {
output file /var/log/caddy/access.log {
roll_size 100mb
roll_keep 10
}
format json
}
}
# TigerOps agent scraping Caddy metrics:
# Prometheus endpoint: http://localhost:2019/metrics
# Admin API: http://localhost:2019/config/
#
# tigerops-agent.yaml:
# integrations:
# caddy:
# admin_url: http://localhost:2019
# scrape_interval: 15s
# log_files:
# - /var/log/caddy/access.logCommon Questions
How does TigerOps collect Caddy metrics?
TigerOps uses two sources: the Caddy Prometheus metrics endpoint (/metrics on the admin port) for runtime metrics, and access log parsing for request-level latency and status codes. Enable both for full coverage.
Does TigerOps monitor Caddy ACME certificate renewals?
Yes. TigerOps reads certificate metadata from the Caddy admin API (/config/apps/tls/certificates) and tracks expiry dates, renewal timestamps, and ACME challenge success/failure rates for all managed domains.
Can TigerOps alert when a Caddy upstream becomes unavailable?
Yes. TigerOps monitors the caddy_reverse_proxy_upstreams_healthy metric and fires an alert when any upstream drops to unhealthy state. You can configure per-upstream and aggregate health thresholds.
How do I monitor multiple Caddy instances across servers?
Install the TigerOps agent on each Caddy host with the same workspace API key. TigerOps groups metrics by host label and creates fleet-wide dashboards with per-instance drill-down.
Does TigerOps work with Caddy running inside Docker or Kubernetes?
Yes. Use the TigerOps agent as a sidecar or DaemonSet. For Kubernetes, expose the Caddy admin port as a service and annotate the pod with prometheus.io/scrape=true for automatic discovery.
Full Caddy Observability in Under 5 Minutes
Certificate tracking, upstream health monitoring, and AI anomaly detection for every Caddy instance you run.