client_golang/prometheus/telemetry.go

48 lines
1.7 KiB
Go

// Copyright (c) 2013, Prometheus Team
// All rights reserved.
//
// Use of this source code is governed by a BSD-style license that can be found
// in the LICENSE file.
//
package prometheus
import (
"time"
)
// Boilerplate metrics about the metrics reporting subservice. These are only
// exposed if the DefaultRegistry's exporter is hooked into the HTTP request
// handler.
var (
marshalErrorCount = NewCounter()
dumpErrorCount = NewCounter()
requestCount = NewCounter()
requestLatencyBuckets = LogarithmicSizedBucketsFor(0, 1000)
requestLatency = NewHistogram(&HistogramSpecification{
Starts: requestLatencyBuckets,
BucketBuilder: AccumulatingBucketBuilder(EvictAndReplaceWith(50, AverageReducer), 1000),
ReportablePercentiles: []float64{0.01, 0.05, 0.5, 0.9, 0.99},
})
startTime = NewGauge()
)
func init() {
startTime.Set(nil, float64(time.Now().Unix()))
DefaultRegistry.Register("telemetry_requests_metrics_total", "A counter of the total requests made against the telemetry system.", NilLabels, requestCount)
DefaultRegistry.Register("telemetry_requests_metrics_latency_microseconds", "A histogram of the response latency for requests made against the telemetry system.", NilLabels, requestLatency)
DefaultRegistry.Register("instance_start_time_seconds", "The time at which the current instance started (UTC).", NilLabels, startTime)
}
// This callback accumulates the microsecond duration of the reporting
// framework's overhead such that it can be reported.
var requestLatencyAccumulator = func(began time.Time) {
microseconds := float64(time.Since(began) / time.Microsecond)
requestLatency.Add(nil, microseconds)
}