homelab-deploy/internal/metrics/metrics.go

// Package metrics provides Prometheus metrics for the homelab-deploy listener.
package metrics

import (
	"code.t-juice.club/torjus/homelab-deploy/internal/messages"
	"github.com/prometheus/client_golang/prometheus"
)

// Collector holds all Prometheus metrics for the listener.
type Collector struct {
	deploymentsTotal     *prometheus.CounterVec
	deploymentDuration   *prometheus.HistogramVec
	deploymentInProgress prometheus.Gauge
	info                 *prometheus.GaugeVec
}

// NewCollector creates a new metrics collector and registers it with the given registerer.
func NewCollector(reg prometheus.Registerer) *Collector {
	c := &Collector{
		deploymentsTotal: prometheus.NewCounterVec(
			prometheus.CounterOpts{
				Name: "homelab_deploy_deployments_total",
				Help: "Total deployment requests processed",
			},
			[]string{"status", "action", "error_code"},
		),
		deploymentDuration: prometheus.NewHistogramVec(
			prometheus.HistogramOpts{
				Name: "homelab_deploy_deployment_duration_seconds",
				Help: "Deployment execution time",
				// Bucket boundaries for typical NixOS build times
				Buckets: []float64{30, 60, 120, 300, 600, 900, 1200, 1800},
			},
			[]string{"action", "success"},
		),
		deploymentInProgress: prometheus.NewGauge(
			prometheus.GaugeOpts{
				Name: "homelab_deploy_deployment_in_progress",
				Help: "1 if deployment running, 0 otherwise",
			},
		),
		info: prometheus.NewGaugeVec(
			prometheus.GaugeOpts{
				Name: "homelab_deploy_info",
				Help: "Static instance metadata",
			},
			[]string{"hostname", "tier", "role", "version"},
		),
	}

	reg.MustRegister(c.deploymentsTotal)
	reg.MustRegister(c.deploymentDuration)
	reg.MustRegister(c.deploymentInProgress)
	reg.MustRegister(c.info)

	c.initMetrics()

	return c
}

// initMetrics initializes all metric label combinations with zero values.
// This ensures metrics appear in Prometheus scrapes before any deployments occur.
func (c *Collector) initMetrics() {
	actions := []messages.Action{
		messages.ActionSwitch,
		messages.ActionBoot,
		messages.ActionTest,
		messages.ActionDryActivate,
	}

	// Initialize deployment counter for common status/action combinations
	for _, action := range actions {
		// Successful completions (no error code)
		c.deploymentsTotal.WithLabelValues("completed", string(action), "")
		// Failed deployments (no error code - from RecordDeploymentEnd)
		c.deploymentsTotal.WithLabelValues("failed", string(action), "")
	}

	// Initialize histogram for all action/success combinations
	for _, action := range actions {
		c.deploymentDuration.WithLabelValues(string(action), "true")
		c.deploymentDuration.WithLabelValues(string(action), "false")
	}
}

// SetInfo sets the static instance metadata.
func (c *Collector) SetInfo(hostname, tier, role, version string) {
	c.info.WithLabelValues(hostname, tier, role, version).Set(1)
}

// RecordDeploymentStart marks the start of a deployment.
func (c *Collector) RecordDeploymentStart() {
	c.deploymentInProgress.Set(1)
}

// RecordDeploymentEnd records the completion of a deployment.
func (c *Collector) RecordDeploymentEnd(action messages.Action, success bool, durationSeconds float64) {
	c.deploymentInProgress.Set(0)

	successLabel := "false"
	if success {
		successLabel = "true"
	}

	c.deploymentDuration.WithLabelValues(string(action), successLabel).Observe(durationSeconds)

	status := "completed"
	if !success {
		status = "failed"
	}

	c.deploymentsTotal.WithLabelValues(status, string(action), "").Inc()
}

// RecordDeploymentFailure records a deployment failure with an error code.
func (c *Collector) RecordDeploymentFailure(action messages.Action, errorCode messages.ErrorCode, durationSeconds float64) {
	c.deploymentInProgress.Set(0)
	c.deploymentDuration.WithLabelValues(string(action), "false").Observe(durationSeconds)
	c.deploymentsTotal.WithLabelValues("failed", string(action), string(errorCode)).Inc()
}

// RecordRejection records a rejected deployment request.
func (c *Collector) RecordRejection(action messages.Action, errorCode messages.ErrorCode) {
	c.deploymentsTotal.WithLabelValues("rejected", string(action), string(errorCode)).Inc()
}