collector/cpu: split cpu freq metrics into separate collector (#1253)
The cpu frequency information is not always needed and/or available. This change allows the cpu frequency metrics to be enabled/disabled separately from the other cpu metrics, and also prevents a frequency metric failure (such as a parse error) from failing the main cpu collector. Fixes #1241 Signed-off-by: Paul Gier <pgier@redhat.com>
This commit is contained in:
parent
f028b81615
commit
cc847f2f44
9 changed files with 241 additions and 175 deletions
|
|
@ -23,18 +23,11 @@ import (
|
|||
"github.com/prometheus/client_golang/prometheus"
|
||||
"github.com/prometheus/common/log"
|
||||
"github.com/prometheus/procfs"
|
||||
"github.com/prometheus/procfs/sysfs"
|
||||
)
|
||||
|
||||
type cpuCollector struct {
|
||||
cpu *prometheus.Desc
|
||||
cpuGuest *prometheus.Desc
|
||||
cpuFreq *prometheus.Desc
|
||||
cpuFreqMin *prometheus.Desc
|
||||
cpuFreqMax *prometheus.Desc
|
||||
scalingFreq *prometheus.Desc
|
||||
scalingFreqMin *prometheus.Desc
|
||||
scalingFreqMax *prometheus.Desc
|
||||
cpuCoreThrottle *prometheus.Desc
|
||||
cpuPackageThrottle *prometheus.Desc
|
||||
}
|
||||
|
|
@ -52,36 +45,6 @@ func NewCPUCollector() (Collector, error) {
|
|||
"Seconds the cpus spent in guests (VMs) for each mode.",
|
||||
[]string{"cpu", "mode"}, nil,
|
||||
),
|
||||
cpuFreq: prometheus.NewDesc(
|
||||
prometheus.BuildFQName(namespace, cpuCollectorSubsystem, "frequency_hertz"),
|
||||
"Current cpu thread frequency in hertz.",
|
||||
[]string{"cpu"}, nil,
|
||||
),
|
||||
cpuFreqMin: prometheus.NewDesc(
|
||||
prometheus.BuildFQName(namespace, cpuCollectorSubsystem, "frequency_min_hertz"),
|
||||
"Minimum cpu thread frequency in hertz.",
|
||||
[]string{"cpu"}, nil,
|
||||
),
|
||||
cpuFreqMax: prometheus.NewDesc(
|
||||
prometheus.BuildFQName(namespace, cpuCollectorSubsystem, "frequency_max_hertz"),
|
||||
"Maximum cpu thread frequency in hertz.",
|
||||
[]string{"cpu"}, nil,
|
||||
),
|
||||
scalingFreq: prometheus.NewDesc(
|
||||
prometheus.BuildFQName(namespace, cpuCollectorSubsystem, "scaling_frequency_hertz"),
|
||||
"Current scaled cpu thread frequency in hertz.",
|
||||
[]string{"cpu"}, nil,
|
||||
),
|
||||
scalingFreqMin: prometheus.NewDesc(
|
||||
prometheus.BuildFQName(namespace, cpuCollectorSubsystem, "scaling_frequency_min_hrts"),
|
||||
"Minimum scaled cpu thread frequency in hertz.",
|
||||
[]string{"cpu"}, nil,
|
||||
),
|
||||
scalingFreqMax: prometheus.NewDesc(
|
||||
prometheus.BuildFQName(namespace, cpuCollectorSubsystem, "scaling_frequency_max_hrts"),
|
||||
"Maximum scaled cpu thread frequency in hertz.",
|
||||
[]string{"cpu"}, nil,
|
||||
),
|
||||
cpuCoreThrottle: prometheus.NewDesc(
|
||||
prometheus.BuildFQName(namespace, cpuCollectorSubsystem, "core_throttles_total"),
|
||||
"Number of times this cpu core has been throttled.",
|
||||
|
|
@ -100,82 +63,12 @@ func (c *cpuCollector) Update(ch chan<- prometheus.Metric) error {
|
|||
if err := c.updateStat(ch); err != nil {
|
||||
return err
|
||||
}
|
||||
if err := c.updateCPUfreq(ch); err != nil {
|
||||
return err
|
||||
}
|
||||
if err := c.updateThermalThrottle(ch); err != nil {
|
||||
return err
|
||||
}
|
||||
return nil
|
||||
}
|
||||
|
||||
// updateCPUfreq reads /sys/devices/system/cpu/cpu* and expose cpu frequency statistics.
|
||||
func (c *cpuCollector) updateCPUfreq(ch chan<- prometheus.Metric) error {
|
||||
fs, err := sysfs.NewFS(*sysPath)
|
||||
if err != nil {
|
||||
return fmt.Errorf("failed to open sysfs: %v", err)
|
||||
}
|
||||
|
||||
cpuFreqs, err := fs.NewSystemCpufreq()
|
||||
if err != nil {
|
||||
return err
|
||||
}
|
||||
|
||||
// sysfs cpufreq values are kHz, thus multiply by 1000 to export base units (hz).
|
||||
// See https://www.kernel.org/doc/Documentation/cpu-freq/user-guide.txt
|
||||
for _, stats := range cpuFreqs {
|
||||
if stats.CpuinfoCurrentFrequency != nil {
|
||||
ch <- prometheus.MustNewConstMetric(
|
||||
c.cpuFreq,
|
||||
prometheus.GaugeValue,
|
||||
float64(*stats.CpuinfoCurrentFrequency)*1000.0,
|
||||
stats.Name,
|
||||
)
|
||||
}
|
||||
if stats.CpuinfoMinimumFrequency != nil {
|
||||
ch <- prometheus.MustNewConstMetric(
|
||||
c.cpuFreqMin,
|
||||
prometheus.GaugeValue,
|
||||
float64(*stats.CpuinfoMinimumFrequency)*1000.0,
|
||||
stats.Name,
|
||||
)
|
||||
}
|
||||
if stats.CpuinfoMaximumFrequency != nil {
|
||||
ch <- prometheus.MustNewConstMetric(
|
||||
c.cpuFreqMax,
|
||||
prometheus.GaugeValue,
|
||||
float64(*stats.CpuinfoMaximumFrequency)*1000.0,
|
||||
stats.Name,
|
||||
)
|
||||
}
|
||||
if stats.ScalingCurrentFrequency != nil {
|
||||
ch <- prometheus.MustNewConstMetric(
|
||||
c.scalingFreq,
|
||||
prometheus.GaugeValue,
|
||||
float64(*stats.ScalingCurrentFrequency)*1000.0,
|
||||
stats.Name,
|
||||
)
|
||||
}
|
||||
if stats.ScalingMinimumFrequency != nil {
|
||||
ch <- prometheus.MustNewConstMetric(
|
||||
c.scalingFreqMin,
|
||||
prometheus.GaugeValue,
|
||||
float64(*stats.ScalingMinimumFrequency)*1000.0,
|
||||
stats.Name,
|
||||
)
|
||||
}
|
||||
if stats.ScalingMaximumFrequency != nil {
|
||||
ch <- prometheus.MustNewConstMetric(
|
||||
c.scalingFreqMax,
|
||||
prometheus.GaugeValue,
|
||||
float64(*stats.ScalingMaximumFrequency)*1000.0,
|
||||
stats.Name,
|
||||
)
|
||||
}
|
||||
}
|
||||
return nil
|
||||
}
|
||||
|
||||
// updateThermalThrottle reads /sys/devices/system/cpu/cpu* and expose thermal throttle statistics.
|
||||
func (c *cpuCollector) updateThermalThrottle(ch chan<- prometheus.Metric) error {
|
||||
cpus, err := filepath.Glob(sysFilePath("devices/system/cpu/cpu[0-9]*"))
|
||||
|
|
|
|||
Loading…
Add table
Add a link
Reference in a new issue