mirror of
https://github.com/prometheus/prometheus
synced 2026-04-20 14:31:49 +08:00
b70a871988
Building off config-specific Prometheus refresh metrics from an earlier PR (https://github.com/prometheus/prometheus/pull/17138), this deletes refresh metrics like `prometheus_sd_refresh_duration_seconds` and `prometheus_sd_refresh_failures_total` when the underlying scrape job configuration is removed on reload. This reduces un-needed cardinality from scrape job specific metrics while still preserving metrics that indicate overall health of a service discovery engine. For example, `prometheus_sd_refresh_failures_total{config="linode-servers",mechanism="linode"} 1` will no longer be exported by Prometheus when the `linode-servers` scrape job for the Linode service provider is removed. The generic, service discovery specific `prometheus_sd_linode_failures_total` metric will persist however. * fix: add targetsMtx lock for targets access * test: validate refresh/discover metrics are gone * ref: combine sdMetrics and refreshMetrics Good idea from @bboreham to combine sdMetrics and refreshMetrics! They're always passed around together and don't have much of a reason not to be combined. mechanismMetrics makes it clear what kind of metrics this is used for (service discovery mechanisms). --------- Signed-off-by: Will Bollock <wbollock@linode.com>
97 lines
3.4 KiB
Go
97 lines
3.4 KiB
Go
// Copyright The Prometheus Authors
|
|
// Licensed under the Apache License, Version 2.0 (the "License");
|
|
// you may not use this file except in compliance with the License.
|
|
// You may obtain a copy of the License at
|
|
//
|
|
// http://www.apache.org/licenses/LICENSE-2.0
|
|
//
|
|
// Unless required by applicable law or agreed to in writing, software
|
|
// distributed under the License is distributed on an "AS IS" BASIS,
|
|
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
|
// See the License for the specific language governing permissions and
|
|
// limitations under the License.
|
|
|
|
package discovery
|
|
|
|
import (
|
|
"time"
|
|
|
|
"github.com/prometheus/client_golang/prometheus"
|
|
)
|
|
|
|
// RefreshMetricsVecs are metric vectors for the "refresh" package.
|
|
// We define them here in the "discovery" package in order to avoid a cyclic dependency between
|
|
// "discovery" and "refresh".
|
|
type RefreshMetricsVecs struct {
|
|
failuresVec *prometheus.CounterVec
|
|
durationVec *prometheus.SummaryVec
|
|
durationHistVec *prometheus.HistogramVec
|
|
|
|
metricRegisterer MetricRegisterer
|
|
}
|
|
|
|
var _ RefreshMetricsManager = (*RefreshMetricsVecs)(nil)
|
|
|
|
func NewRefreshMetrics(reg prometheus.Registerer) RefreshMetricsManager {
|
|
m := &RefreshMetricsVecs{
|
|
failuresVec: prometheus.NewCounterVec(
|
|
prometheus.CounterOpts{
|
|
Name: "prometheus_sd_refresh_failures_total",
|
|
Help: "Number of refresh failures for the given SD mechanism.",
|
|
},
|
|
[]string{"mechanism", "config"}),
|
|
durationVec: prometheus.NewSummaryVec(
|
|
prometheus.SummaryOpts{
|
|
Name: "prometheus_sd_refresh_duration_seconds",
|
|
Help: "The duration of a refresh in seconds for the given SD mechanism.",
|
|
Objectives: map[float64]float64{0.5: 0.05, 0.9: 0.01, 0.99: 0.001},
|
|
},
|
|
[]string{"mechanism", "config"}),
|
|
durationHistVec: prometheus.NewHistogramVec(
|
|
prometheus.HistogramOpts{
|
|
Name: "prometheus_sd_refresh_duration_histogram_seconds",
|
|
Help: "The duration of a refresh for the given SD mechanism.",
|
|
Buckets: []float64{.01, .1, 1, 10},
|
|
NativeHistogramBucketFactor: 1.1,
|
|
NativeHistogramMaxBucketNumber: 100,
|
|
NativeHistogramMinResetDuration: 1 * time.Hour,
|
|
},
|
|
[]string{"mechanism"}),
|
|
}
|
|
|
|
// The reason we register metric vectors instead of metrics is so that
|
|
// the metrics are not visible until they are recorded.
|
|
m.metricRegisterer = NewMetricRegisterer(reg, []prometheus.Collector{
|
|
m.failuresVec,
|
|
m.durationVec,
|
|
m.durationHistVec,
|
|
})
|
|
|
|
return m
|
|
}
|
|
|
|
// Instantiate returns metrics out of metric vectors for a given mechanism and config.
|
|
func (m *RefreshMetricsVecs) Instantiate(mech, config string) *RefreshMetrics {
|
|
return &RefreshMetrics{
|
|
Failures: m.failuresVec.WithLabelValues(mech, config),
|
|
Duration: m.durationVec.WithLabelValues(mech, config),
|
|
DurationHistogram: m.durationHistVec.WithLabelValues(mech),
|
|
}
|
|
}
|
|
|
|
// Register implements discovery.DiscovererMetrics.
|
|
func (m *RefreshMetricsVecs) Register() error {
|
|
return m.metricRegisterer.RegisterMetrics()
|
|
}
|
|
|
|
// Unregister implements discovery.DiscovererMetrics.
|
|
func (m *RefreshMetricsVecs) Unregister() {
|
|
m.metricRegisterer.UnregisterMetrics()
|
|
}
|
|
|
|
// DeleteLabelValues deletes refresh metrics for a specific mechanism and config. Smart to use this when a scrape job is removed.
|
|
func (m *RefreshMetricsVecs) DeleteLabelValues(mech, config string) {
|
|
m.failuresVec.DeleteLabelValues(mech, config)
|
|
m.durationVec.DeleteLabelValues(mech, config)
|
|
}
|