Metrics now have expiration if not updated

metrics/pve_node_container_collector.go

@@ -1,155 +1,184 @@
 package metrics
 
 import (
+	"time"
+
 	"github.com/prometheus/client_golang/prometheus"
-	"github.com/prometheus/client_golang/prometheus/promauto"
 	"lostak.dev/pve-exporter/proxmox"
 )
 
 // PVE container collector.
 type PveContainerCollector struct {
 	apiClient *proxmox.PveApiClient // PVE API client instance.
+	registry  *TTLRegistry          // TTL metrics registry.
 
-	state  *prometheus.GaugeVec // Container state prometheus gauge.
-	uptime *prometheus.GaugeVec // Container uptime prometheus gauge.
+	state  *TTLGaugeVec // Container state prometheus gauge.
+	uptime *TTLGaugeVec // Container uptime prometheus gauge.
 
-	cpu      *prometheus.GaugeVec // Container count of CPUs prometheus gauge.
-	cpuUsage *prometheus.GaugeVec // Container CPU usage % prometheus gauge.
+	cpu      *TTLGaugeVec // Container count of CPUs prometheus gauge.
+	cpuUsage *TTLGaugeVec // Container CPU usage % prometheus gauge.
 
-	memBytes     *prometheus.GaugeVec // Container memory in bytes prometheus gauge.
-	memBytesUsed *prometheus.GaugeVec // Container memory usage in bytes prometheus gauge.
+	memBytes     *TTLGaugeVec // Container memory in bytes prometheus gauge.
+	memBytesUsed *TTLGaugeVec // Container memory usage in bytes prometheus gauge.
 
-	netReceive  *prometheus.GaugeVec // Container network RX in bytes prometheus gauge.
-	netTransmit *prometheus.GaugeVec // Container network TX in bytes prometheus gauge.
+	netReceive  *TTLGaugeVec // Container network RX in bytes prometheus gauge.
+	netTransmit *TTLGaugeVec // Container network TX in bytes prometheus gauge.
 
-	diskWrite *prometheus.GaugeVec // Container disk written in bytes prometheus gauge.
-	diskRead  *prometheus.GaugeVec // Container disk read in bytes prometheus gauge.
+	diskWrite *TTLGaugeVec // Container disk written in bytes prometheus gauge.
+	diskRead  *TTLGaugeVec // Container disk read in bytes prometheus gauge.
 
-	disk    *prometheus.GaugeVec // Container disk space usage in bytes prometheus gauge.
-	diskMax *prometheus.GaugeVec // Container disk size in bytes prometheus gauge.
-	swap    *prometheus.GaugeVec // Container swap usage in bytes prometheus gauge.
+	disk    *TTLGaugeVec // Container disk space usage in bytes prometheus gauge.
+	diskMax *TTLGaugeVec // Container disk size in bytes prometheus gauge.
+	swap    *TTLGaugeVec // Container swap usage in bytes prometheus gauge.
 }
 
 // Create new instance of PVE container collector.
-func NewPveContainerCollector(apiClient *proxmox.PveApiClient) *PveContainerCollector {
+func NewPveContainerCollector(apiClient *proxmox.PveApiClient, registry *TTLRegistry) *PveContainerCollector {
 	c := PveContainerCollector{apiClient: apiClient}
+	c.registry = registry
 
 	// Container state.
-	c.state = promauto.NewGaugeVec(
+	c.state = NewTTLGaugeVec(
 		prometheus.GaugeOpts{
 			Name: "pve_ct_state",
 			Help: "Container state.",
 		},
 		[]string{"cluster", "node", "vmid", "name"},
+		1*time.Minute,
 	)
+	c.registry.Register(c.state)
 
 	// Container uptime.
-	c.uptime = promauto.NewGaugeVec(
+	c.uptime = NewTTLGaugeVec(
 		prometheus.GaugeOpts{
 			Name: "pve_ct_uptime",
 			Help: "Container uptime.",
 		},
 		[]string{"cluster", "node", "vmid", "name"},
+		1*time.Minute,
 	)
+	c.registry.Register(c.uptime)
 
 	// Container CPU count.
-	c.cpu = promauto.NewGaugeVec(
+	c.cpu = NewTTLGaugeVec(
 		prometheus.GaugeOpts{
 			Name: "pve_ct_cpu_count",
 			Help: "Container CPU count.",
 		},
 		[]string{"cluster", "node", "vmid", "name"},
+		1*time.Minute,
 	)
+	c.registry.Register(c.cpu)
 
 	// Container CPU usage.
-	c.cpuUsage = promauto.NewGaugeVec(
+	c.cpuUsage = NewTTLGaugeVec(
 		prometheus.GaugeOpts{
 			Name: "pve_ct_cpu_usage",
 			Help: "Container CPU usage.",
 		},
 		[]string{"cluster", "node", "vmid", "name"},
+		1*time.Minute,
 	)
+	c.registry.Register(c.cpuUsage)
 
 	// Container memory total.
-	c.memBytes = promauto.NewGaugeVec(
+	c.memBytes = NewTTLGaugeVec(
 		prometheus.GaugeOpts{
 			Name: "pve_ct_mem_total_bytes",
 			Help: "Container total memory in bytes.",
 		},
 		[]string{"cluster", "node", "vmid", "name"},
+		1*time.Minute,
 	)
+	c.registry.Register(c.memBytes)
 
 	// Container memory usage.
-	c.memBytesUsed = promauto.NewGaugeVec(
+	c.memBytesUsed = NewTTLGaugeVec(
 		prometheus.GaugeOpts{
 			Name: "pve_ct_mem_used_bytes",
 			Help: "Container used memory in bytes.",
 		},
 		[]string{"cluster", "node", "vmid", "name"},
+		1*time.Minute,
 	)
+	c.registry.Register(c.memBytesUsed)
 
 	// Container network RX.
-	c.netReceive = promauto.NewGaugeVec(
+	c.netReceive = NewTTLGaugeVec(
 		prometheus.GaugeOpts{
 			Name: "pve_ct_network_in_bytes",
 			Help: "Container network RX bytes.",
 		},
 		[]string{"cluster", "node", "vmid", "name"},
+		1*time.Minute,
 	)
+	c.registry.Register(c.netReceive)
 
 	// Container network TX.
-	c.netTransmit = promauto.NewGaugeVec(
+	c.netTransmit = NewTTLGaugeVec(
 		prometheus.GaugeOpts{
 			Name: "pve_ct_network_out_bytes",
 			Help: "Container network TX bytes.",
 		},
 		[]string{"cluster", "node", "vmid", "name"},
+		1*time.Minute,
 	)
+	c.registry.Register(c.netTransmit)
 
 	// Container disk written.
-	c.diskWrite = promauto.NewGaugeVec(
+	c.diskWrite = NewTTLGaugeVec(
 		prometheus.GaugeOpts{
 			Name: "pve_ct_disk_wr_bytes",
 			Help: "Container disk written bytes.",
 		},
 		[]string{"cluster", "node", "vmid", "name"},
+		1*time.Minute,
 	)
+	c.registry.Register(c.diskWrite)
 
 	// Container disk read.
-	c.diskRead = promauto.NewGaugeVec(
+	c.diskRead = NewTTLGaugeVec(
 		prometheus.GaugeOpts{
 			Name: "pve_ct_disk_rd_bytes",
 			Help: "Container disk read bytes.",
 		},
 		[]string{"cluster", "node", "vmid", "name"},
+		1*time.Minute,
 	)
+	c.registry.Register(c.diskRead)
 
 	// Container disk size.
-	c.disk = promauto.NewGaugeVec(
+	c.disk = NewTTLGaugeVec(
 		prometheus.GaugeOpts{
 			Name: "pve_ct_disk_usage_bytes",
 			Help: "Container disk read bytes.",
 		},
 		[]string{"cluster", "node", "vmid", "name"},
+		1*time.Minute,
 	)
+	c.registry.Register(c.disk)
 
 	// Container disk size.
-	c.diskMax = promauto.NewGaugeVec(
+	c.diskMax = NewTTLGaugeVec(
 		prometheus.GaugeOpts{
 			Name: "pve_ct_disk_size_bytes",
 			Help: "Container disk size bytes.",
 		},
 		[]string{"cluster", "node", "vmid", "name"},
+		1*time.Minute,
 	)
+	c.registry.Register(c.diskMax)
 
 	// Container swap usage.
-	c.swap = promauto.NewGaugeVec(
+	c.swap = NewTTLGaugeVec(
 		prometheus.GaugeOpts{
 			Name: "pve_ct_swap_used_bytes",
 			Help: "Container swap usage bytes.",
 		},
 		[]string{"cluster", "node", "vmid", "name"},
+		1*time.Minute,
 	)
+	c.registry.Register(c.swap)
 
 	return &c
 }
@@ -161,20 +190,6 @@ func (c *PveContainerCollector) CollectMetrics() error {
 		return err
 	}
 
-	c.state.Reset()
-	c.cpu.Reset()
-	c.memBytes.Reset()
-	c.diskMax.Reset()
-	c.uptime.Reset()
-	c.cpuUsage.Reset()
-	c.memBytesUsed.Reset()
-	c.netReceive.Reset()
-	c.netTransmit.Reset()
-	c.diskRead.Reset()
-	c.diskWrite.Reset()
-	c.disk.Reset()
-	c.swap.Reset()
-
 	for _, node := range cluster.NodeStatuses {
 		containers, err := c.apiClient.GetNodeContainerList(node.Name)
 		if err != nil {