Initial commit
This commit is contained in:
Jan Lošťák
311
metrics/pve_node_virtual_machine_collector.go
Normal file
311
metrics/pve_node_virtual_machine_collector.go
Normal file
@@ -0,0 +1,311 @@
|
||||
package metrics
|
||||
|
||||
import (
|
||||
"strconv"
|
||||
|
||||
"github.com/prometheus/client_golang/prometheus"
|
||||
"github.com/prometheus/client_golang/prometheus/promauto"
|
||||
"lostak.dev/pve-exporter/proxmox"
|
||||
)
|
||||
|
||||
// PVE virtual machine collector.
|
||||
type PveVirtualMachineCollector struct {
|
||||
apiClient *proxmox.PveApiClient // PVE API client instance.
|
||||
|
||||
state *prometheus.GaugeVec // Virtual machine state prometheus gauge.
|
||||
uptime *prometheus.GaugeVec // Virtual machine uptime prometheus gauge.
|
||||
|
||||
cpu *prometheus.GaugeVec // Virtual machine count of CPUs prometheus gauge.
|
||||
cpuUsage *prometheus.GaugeVec // Virtual machine CPU usage % prometheus gauge.
|
||||
|
||||
memBytes *prometheus.GaugeVec // Virtual machine memory in bytes prometheus gauge.
|
||||
memBytesUsed *prometheus.GaugeVec // Virtual machine memory usage in bytes prometheus gauge.
|
||||
|
||||
disk *prometheus.GaugeVec // Virtual machine disk space usage in bytes prometheus gauge.
|
||||
diskMax *prometheus.GaugeVec // Virtual machine disk size in bytes prometheus gauge.
|
||||
swap *prometheus.GaugeVec // Virtual machine swap usage in bytes prometheus gauge.
|
||||
|
||||
netReceive *prometheus.GaugeVec // Virtual machine network receive in bytes prometheus gauge.
|
||||
netTransmit *prometheus.GaugeVec // Virtual machine network transmit in bytes prometheus gauge.
|
||||
|
||||
diskReadOps *prometheus.GaugeVec // Virtual machine disk read ops prometheus gauge.
|
||||
diskWriteOps *prometheus.GaugeVec // Virtual machine disk write ops prometheus gauge.
|
||||
|
||||
diskReadBytes *prometheus.GaugeVec // Virtual machine disk read bytes prometheus gauge.
|
||||
diskWriteBytes *prometheus.GaugeVec // Virtual machine disk write bytes prometheus gauge.
|
||||
|
||||
diskReadTimeNs *prometheus.GaugeVec // Virtual machine disk read time total prometheus gauge.
|
||||
diskWriteTimeNs *prometheus.GaugeVec // Virtual machine disk write time total prometheus gauge.
|
||||
|
||||
diskFailedReadOps *prometheus.GaugeVec // Virtual machine disk failed read ops prometheus gauge.
|
||||
diskFailedWriteOps *prometheus.GaugeVec // Virtual machine disk failed write ops prometheus gauge.
|
||||
|
||||
agent *prometheus.GaugeVec // Virtual machine agent enabled prometheus gauge.
|
||||
}
|
||||
|
||||
// Create new instance of PVE virtual machine collector.
|
||||
func NewPveVirtualMachineCollector(apiClient *proxmox.PveApiClient) *PveVirtualMachineCollector {
|
||||
c := PveVirtualMachineCollector{apiClient: apiClient}
|
||||
|
||||
// Virtual machine state.
|
||||
c.state = promauto.NewGaugeVec(
|
||||
prometheus.GaugeOpts{
|
||||
Name: "pve_vm_state",
|
||||
Help: "Virtual machine state.",
|
||||
},
|
||||
[]string{"cluster", "node", "vmid", "name"},
|
||||
)
|
||||
|
||||
// Virtual machine uptime.
|
||||
c.uptime = promauto.NewGaugeVec(
|
||||
prometheus.GaugeOpts{
|
||||
Name: "pve_vm_uptime",
|
||||
Help: "Virtual machine uptime.",
|
||||
},
|
||||
[]string{"cluster", "node", "vmid", "name"},
|
||||
)
|
||||
|
||||
// Virtual machine agent state.
|
||||
c.agent = promauto.NewGaugeVec(
|
||||
prometheus.GaugeOpts{
|
||||
Name: "pve_vm_agent",
|
||||
Help: "Virtual machine agent state.",
|
||||
},
|
||||
[]string{"cluster", "node", "vmid", "name"},
|
||||
)
|
||||
|
||||
// Virtual machine CPU count.
|
||||
c.cpu = promauto.NewGaugeVec(
|
||||
prometheus.GaugeOpts{
|
||||
Name: "pve_vm_cpu_count",
|
||||
Help: "Virtual machine CPU count.",
|
||||
},
|
||||
[]string{"cluster", "node", "vmid", "name"},
|
||||
)
|
||||
|
||||
// Virtual machine CPU usage.
|
||||
c.cpuUsage = promauto.NewGaugeVec(
|
||||
prometheus.GaugeOpts{
|
||||
Name: "pve_vm_cpu_usage",
|
||||
Help: "Virtual machine CPU usage.",
|
||||
},
|
||||
[]string{"cluster", "node", "vmid", "name"},
|
||||
)
|
||||
|
||||
// Virtual machine memory total.
|
||||
c.memBytes = promauto.NewGaugeVec(
|
||||
prometheus.GaugeOpts{
|
||||
Name: "pve_vm_mem_total_bytes",
|
||||
Help: "Virtual machine total memory in bytes.",
|
||||
},
|
||||
[]string{"cluster", "node", "vmid", "name"},
|
||||
)
|
||||
|
||||
// Virtual machine memory usage.
|
||||
c.memBytesUsed = promauto.NewGaugeVec(
|
||||
prometheus.GaugeOpts{
|
||||
Name: "pve_vm_mem_used_bytes",
|
||||
Help: "Virtual machine used memory in bytes.",
|
||||
},
|
||||
[]string{"cluster", "node", "vmid", "name"},
|
||||
)
|
||||
|
||||
// Virtual machine disk size.
|
||||
c.disk = promauto.NewGaugeVec(
|
||||
prometheus.GaugeOpts{
|
||||
Name: "pve_vm_disk_usage_bytes",
|
||||
Help: "Virtual machine disk read bytes.",
|
||||
},
|
||||
[]string{"cluster", "node", "vmid", "name"},
|
||||
)
|
||||
|
||||
// Virtual machine disk size.
|
||||
c.diskMax = promauto.NewGaugeVec(
|
||||
prometheus.GaugeOpts{
|
||||
Name: "pve_vm_disk_size_bytes",
|
||||
Help: "Virtual machine disk size bytes.",
|
||||
},
|
||||
[]string{"cluster", "node", "vmid", "name"},
|
||||
)
|
||||
|
||||
// Virtual machine network receive bytes.
|
||||
c.netReceive = promauto.NewGaugeVec(
|
||||
prometheus.GaugeOpts{
|
||||
Name: "pve_vm_network_in_bytes",
|
||||
Help: "Virtual machine network receive in bytes.",
|
||||
},
|
||||
[]string{"cluster", "node", "vmid", "name", "interface"},
|
||||
)
|
||||
|
||||
// Virtual machine network transmit bytes.
|
||||
c.netTransmit = promauto.NewGaugeVec(
|
||||
prometheus.GaugeOpts{
|
||||
Name: "pve_vm_network_out_bytes",
|
||||
Help: "Virtual machine network transmit in bytes.",
|
||||
},
|
||||
[]string{"cluster", "node", "vmid", "name", "interface"},
|
||||
)
|
||||
|
||||
// Virtual machine disk read ops.
|
||||
c.diskReadOps = promauto.NewGaugeVec(
|
||||
prometheus.GaugeOpts{
|
||||
Name: "pve_vm_disk_rd_operations",
|
||||
Help: "Virtual machine disk read ops.",
|
||||
},
|
||||
[]string{"cluster", "node", "vmid", "name", "device"},
|
||||
)
|
||||
|
||||
// Virtual machine disk write ops.
|
||||
c.diskWriteOps = promauto.NewGaugeVec(
|
||||
prometheus.GaugeOpts{
|
||||
Name: "pve_vm_disk_wr_operations",
|
||||
Help: "Virtual machine disk write ops.",
|
||||
},
|
||||
[]string{"cluster", "node", "vmid", "name", "device"},
|
||||
)
|
||||
|
||||
// Virtual machine disk read bytes.
|
||||
c.diskReadBytes = promauto.NewGaugeVec(
|
||||
prometheus.GaugeOpts{
|
||||
Name: "pve_vm_disk_rd_bytes",
|
||||
Help: "Virtual machine disk read bytes.",
|
||||
},
|
||||
[]string{"cluster", "node", "vmid", "name", "device"},
|
||||
)
|
||||
|
||||
// Virtual machine disk write bytes.
|
||||
c.diskWriteBytes = promauto.NewGaugeVec(
|
||||
prometheus.GaugeOpts{
|
||||
Name: "pve_vm_disk_wr_bytes",
|
||||
Help: "Virtual machine disk write bytes.",
|
||||
},
|
||||
[]string{"cluster", "node", "vmid", "name", "device"},
|
||||
)
|
||||
|
||||
// Virtual machine failed disk read ops.
|
||||
c.diskFailedReadOps = promauto.NewGaugeVec(
|
||||
prometheus.GaugeOpts{
|
||||
Name: "pve_vm_disk_failed_rd_ops",
|
||||
Help: "Virtual machine failed disk read ops.",
|
||||
},
|
||||
[]string{"cluster", "node", "vmid", "name", "device"},
|
||||
)
|
||||
|
||||
// Virtual machine failed disk write ops.
|
||||
c.diskFailedWriteOps = promauto.NewGaugeVec(
|
||||
prometheus.GaugeOpts{
|
||||
Name: "pve_vm_disk_failed_wr_ops",
|
||||
Help: "Virtual machine failed disk write ops.",
|
||||
},
|
||||
[]string{"cluster", "node", "vmid", "name", "device"},
|
||||
)
|
||||
|
||||
// Virtual machine disk read time total nanoseconds.
|
||||
c.diskReadTimeNs = promauto.NewGaugeVec(
|
||||
prometheus.GaugeOpts{
|
||||
Name: "pve_vm_disk_rd_time_total_ns",
|
||||
Help: "Virtual machine disk read time total in nanoseconds.",
|
||||
},
|
||||
[]string{"cluster", "node", "vmid", "name", "device"},
|
||||
)
|
||||
|
||||
// Virtual machine disk write time total nanoseconds.
|
||||
c.diskWriteTimeNs = promauto.NewGaugeVec(
|
||||
prometheus.GaugeOpts{
|
||||
Name: "pve_vm_disk_wr_time_total_ns",
|
||||
Help: "Virtual machine disk write time total in nanoseconds.",
|
||||
},
|
||||
[]string{"cluster", "node", "vmid", "name", "device"},
|
||||
)
|
||||
|
||||
return &c
|
||||
}
|
||||
|
||||
// PveMetricsCollector interface implementation.
|
||||
func (c *PveVirtualMachineCollector) CollectMetrics() error {
|
||||
cluster, err := c.apiClient.GetClusterStatus()
|
||||
if err != nil {
|
||||
return err
|
||||
}
|
||||
|
||||
for _, node := range cluster.NodeStatuses {
|
||||
qemus, err := c.apiClient.GetNodeQemuList(node.Name)
|
||||
if err != nil {
|
||||
return err
|
||||
}
|
||||
|
||||
for _, qemu := range *qemus {
|
||||
// Skip templates because they are always offline.
|
||||
if qemu.Template == 1 {
|
||||
continue
|
||||
}
|
||||
|
||||
labels := prometheus.Labels{
|
||||
"cluster": cluster.GetClusterName(),
|
||||
"node": node.Name,
|
||||
"vmid": strconv.Itoa(qemu.VMID),
|
||||
"name": qemu.Name,
|
||||
}
|
||||
|
||||
c.state.With(labels).Set(qemu.GetStatusNumeric())
|
||||
c.cpu.With(labels).Set(float64(qemu.CPUs))
|
||||
c.memBytes.With(labels).Set(float64(qemu.MaxMem))
|
||||
c.diskMax.With(labels).Set(float64(qemu.MaxDisk))
|
||||
|
||||
// Metrics only on running virtual machines.
|
||||
if qemu.IsRunning() {
|
||||
c.uptime.With(labels).Set(float64(qemu.Uptime))
|
||||
c.cpuUsage.With(labels).Set(float64(qemu.CPU))
|
||||
c.memBytesUsed.With(labels).Set(float64(qemu.Mem))
|
||||
|
||||
detail, err := c.apiClient.GetNodeQemu(node.Name, strconv.Itoa(qemu.VMID))
|
||||
if err != nil {
|
||||
return err
|
||||
}
|
||||
|
||||
c.agent.With(labels).Set(float64(detail.Agent))
|
||||
|
||||
for iface, value := range detail.Nics {
|
||||
labels := prometheus.Labels{
|
||||
"cluster": cluster.GetClusterName(),
|
||||
"node": node.Name,
|
||||
"vmid": strconv.Itoa(qemu.VMID),
|
||||
"name": qemu.Name,
|
||||
"interface": iface,
|
||||
}
|
||||
|
||||
c.netReceive.With(labels).Set(float64(value.NetIn))
|
||||
c.netTransmit.With(labels).Set(float64(value.NetOut))
|
||||
}
|
||||
|
||||
for device, value := range detail.BlockStat {
|
||||
labels := prometheus.Labels{
|
||||
"cluster": cluster.GetClusterName(),
|
||||
"node": node.Name,
|
||||
"vmid": strconv.Itoa(qemu.VMID),
|
||||
"name": qemu.Name,
|
||||
"device": device,
|
||||
}
|
||||
|
||||
c.diskReadOps.With(labels).Set(float64(value.RdOperations))
|
||||
c.diskWriteOps.With(labels).Set(float64(value.WrOperations))
|
||||
|
||||
c.diskReadBytes.With(labels).Set(float64(value.RdBytes))
|
||||
c.diskWriteBytes.With(labels).Set(float64(value.WrBytes))
|
||||
|
||||
c.diskFailedReadOps.With(labels).Set(float64(value.FailedRdOperations))
|
||||
c.diskFailedWriteOps.With(labels).Set(float64(value.FailedWrOperations))
|
||||
|
||||
c.diskReadTimeNs.With(labels).Set(float64(value.RdTotalTimeNs))
|
||||
c.diskWriteTimeNs.With(labels).Set(float64(value.WrTotalTimeNs))
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
return nil
|
||||
}
|
||||
|
||||
// PveMetricsCollector interface implementation.
|
||||
func (c *PveVirtualMachineCollector) GetName() string {
|
||||
return "Virtual Machine"
|
||||
}
|
||||
Reference in New Issue
Block a user