Skip to content

fix(metric_aggregation): Fix duplicate metrics registration #15142

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 4 commits into from
Nov 27, 2024
Merged

fix(metric_aggregation): Fix duplicate metrics registration #15142

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
9dc5bed
Fix aggregation metrics duplicate registration
shantanualsi Nov 27, 2024
ee454bb
Use single instance of prometheus metric in pattern ingester
shantanualsi Nov 27, 2024
84acbe2
Convert agg metrics to singleton
shantanualsi Nov 27, 2024
5f54957
Fix log messages
shantanualsi Nov 27, 2024
File filter

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Prev Previous commit
Next Next commit
Convert agg metrics to singleton
  • Loading branch information
@shantanualsi
shantanualsi committed Nov 27, 2024
commit 84acbe25276f0c43ec526f315419a6abe982992f
155 changes: 83 additions & 72 deletions pkg/pattern/aggregation/metrics.go
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,12 +1,19 @@
package aggregation

import (
"sync"

"github.com/prometheus/client_golang/prometheus"
"github.com/prometheus/client_golang/prometheus/promauto"

"github.com/grafana/loki/v3/pkg/util/constants"
)

var (
aggMetrics *Metrics
metricsOnce sync.Once
)

type Metrics struct {
reg prometheus.Registerer

Expand All @@ -28,76 +35,80 @@ type Metrics struct {
}

func NewMetrics(r prometheus.Registerer) *Metrics {
return &Metrics{
chunks: promauto.With(r).NewGaugeVec(prometheus.GaugeOpts{
Namespace: constants.Loki,
Subsystem: "pattern_ingester",
Name: "metric_chunks",
Help: "The total number of chunks in memory.",
}, []string{"service_name"}),
samples: promauto.With(r).NewCounterVec(prometheus.CounterOpts{
Namespace: constants.Loki,
Subsystem: "pattern_ingester",
Name: "metric_samples",
Help: "The total number of samples in memory.",
}, []string{"service_name"}),
pushErrors: promauto.With(r).NewCounterVec(prometheus.CounterOpts{
Namespace: constants.Loki,
Subsystem: "pattern_ingester",
Name: "push_errors_total",
Help: "Total number of errors when pushing metrics to Loki.",
}, []string{"tenant_id", "error_type"}),

pushRetries: promauto.With(r).NewCounterVec(prometheus.CounterOpts{
Namespace: constants.Loki,
Subsystem: "pattern_ingester",
Name: "push_retries_total",
Help: "Total number of retries when pushing metrics to Loki.",
}, []string{"tenant_id"}),

pushSuccesses: promauto.With(r).NewCounterVec(prometheus.CounterOpts{
Namespace: constants.Loki,
Subsystem: "pattern_ingester",
Name: "push_successes_total",
Help: "Total number of successful pushes to Loki.",
}, []string{"tenant_id"}),

// Batch metrics
payloadSize: promauto.With(r).NewHistogramVec(prometheus.HistogramOpts{
Namespace: constants.Loki,
Subsystem: "pattern_ingester",
Name: "push_payload_bytes",
Help: "Size of push payloads in bytes.",
Buckets: []float64{1024, 4096, 16384, 65536, 262144, 1048576},
}, []string{"tenant_id"}),

streamsPerPush: promauto.With(r).NewHistogramVec(prometheus.HistogramOpts{
Namespace: constants.Loki,
Subsystem: "pattern_ingester",
Name: "streams_per_push",
Help: "Number of streams in each push request.",
Buckets: []float64{1, 5, 10, 25, 50, 100, 250, 500, 1000},
}, []string{"tenant_id"}),

entriesPerPush: promauto.With(r).NewHistogramVec(prometheus.HistogramOpts{
Namespace: constants.Loki,
Subsystem: "pattern_ingester",
Name: "entries_per_push",
Help: "Number of entries in each push request.",
Buckets: []float64{10, 50, 100, 500, 1000, 5000, 10000},
}, []string{"tenant_id"}),

servicesTracked: promauto.With(r).NewGaugeVec(prometheus.GaugeOpts{
Namespace: constants.Loki,
Subsystem: "pattern_ingester",
Name: "services_tracked",
Help: "Number of unique services being tracked.",
}, []string{"tenant_id"}),
writeTimeout: promauto.With(r).NewCounterVec(prometheus.CounterOpts{
Namespace: constants.Loki,
Subsystem: "pattern_ingester",
Name: "write_timeouts_total",
Help: "Total number of write timeouts.",
}, []string{"tenant_id"}),
}
metricsOnce.Do(func() {
aggMetrics = &Metrics{
chunks: promauto.With(r).NewGaugeVec(prometheus.GaugeOpts{
Namespace: constants.Loki,
Subsystem: "pattern_ingester",
Name: "metric_chunks",
Help: "The total number of chunks in memory.",
}, []string{"service_name"}),
samples: promauto.With(r).NewCounterVec(prometheus.CounterOpts{
Namespace: constants.Loki,
Subsystem: "pattern_ingester",
Name: "metric_samples",
Help: "The total number of samples in memory.",
}, []string{"service_name"}),
pushErrors: promauto.With(r).NewCounterVec(prometheus.CounterOpts{
Namespace: constants.Loki,
Subsystem: "pattern_ingester",
Name: "push_errors_total",
Help: "Total number of errors when pushing metrics to Loki.",
}, []string{"tenant_id", "error_type"}),

pushRetries: promauto.With(r).NewCounterVec(prometheus.CounterOpts{
Namespace: constants.Loki,
Subsystem: "pattern_ingester",
Name: "push_retries_total",
Help: "Total number of retries when pushing metrics to Loki.",
}, []string{"tenant_id"}),

pushSuccesses: promauto.With(r).NewCounterVec(prometheus.CounterOpts{
Namespace: constants.Loki,
Subsystem: "pattern_ingester",
Name: "push_successes_total",
Help: "Total number of successful pushes to Loki.",
}, []string{"tenant_id"}),

// Batch metrics
payloadSize: promauto.With(r).NewHistogramVec(prometheus.HistogramOpts{
Namespace: constants.Loki,
Subsystem: "pattern_ingester",
Name: "push_payload_bytes",
Help: "Size of push payloads in bytes.",
Buckets: []float64{1024, 4096, 16384, 65536, 262144, 1048576},
}, []string{"tenant_id"}),

streamsPerPush: promauto.With(r).NewHistogramVec(prometheus.HistogramOpts{
Namespace: constants.Loki,
Subsystem: "pattern_ingester",
Name: "streams_per_push",
Help: "Number of streams in each push request.",
Buckets: []float64{1, 5, 10, 25, 50, 100, 250, 500, 1000},
}, []string{"tenant_id"}),

entriesPerPush: promauto.With(r).NewHistogramVec(prometheus.HistogramOpts{
Namespace: constants.Loki,
Subsystem: "pattern_ingester",
Name: "entries_per_push",
Help: "Number of entries in each push request.",
Buckets: []float64{10, 50, 100, 500, 1000, 5000, 10000},
}, []string{"tenant_id"}),

servicesTracked: promauto.With(r).NewGaugeVec(prometheus.GaugeOpts{
Namespace: constants.Loki,
Subsystem: "pattern_ingester",
Name: "services_tracked",
Help: "Number of unique services being tracked.",
}, []string{"tenant_id"}),
writeTimeout: promauto.With(r).NewCounterVec(prometheus.CounterOpts{
Namespace: constants.Loki,
Subsystem: "pattern_ingester",
Name: "write_timeouts_total",
Help: "Total number of write timeouts.",
}, []string{"tenant_id"}),
}
})

return aggMetrics
}
2 changes: 1 addition & 1 deletion pkg/pattern/aggregation/push_test.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -83,7 +83,7 @@ func Test_Push(t *testing.T) {
"user", "secret",
false,
&backoff,
log.NewNopLogger(), nil,
log.NewNopLogger(), NewMetrics(nil),
)
require.NoError(t, err)
ts, payload := testPayload()
Expand Down
Loading