From 848bec6d245591d133050237e2109d7c1f686b1e Mon Sep 17 00:00:00 2001 From: Chris Lu Date: Wed, 26 Nov 2025 15:51:38 -0800 Subject: [PATCH] Metrics: Add Prometheus metrics for concurrent upload tracking (#7555) * metrics: add Prometheus metrics for concurrent upload tracking Add Prometheus metrics to monitor concurrent upload activity for both filer and S3 servers. This provides visibility into the upload limiting feature added in the previous PR. New Metrics: - SeaweedFS_filer_in_flight_upload_bytes: Current bytes being uploaded to filer - SeaweedFS_filer_in_flight_upload_count: Current number of uploads to filer - SeaweedFS_s3_in_flight_upload_bytes: Current bytes being uploaded to S3 - SeaweedFS_s3_in_flight_upload_count: Current number of uploads to S3 The metrics are updated atomically whenever uploads start or complete, providing real-time visibility into upload concurrency levels. This helps operators: - Monitor upload concurrency in real-time - Set appropriate limits based on actual usage patterns - Detect potential bottlenecks or capacity issues - Track the effectiveness of upload limiting configuration * grafana: add dashboard panels for concurrent upload metrics Add 4 new panels to the Grafana dashboard to visualize the concurrent upload metrics added in this PR: Filer Section: - Filer Concurrent Uploads: Shows current number of concurrent uploads - Filer Concurrent Upload Bytes: Shows current bytes being uploaded S3 Gateway Section: - S3 Concurrent Uploads: Shows current number of concurrent uploads - S3 Concurrent Upload Bytes: Shows current bytes being uploaded These panels help operators monitor upload concurrency in real-time and tune the upload limiting configuration based on actual usage patterns. * more efficient --- other/metrics/grafana_seaweedfs.json | 333 ++++++++++++++++++++++++++- weed/s3api/s3api_circuit_breaker.go | 22 +- weed/server/filer_server_handlers.go | 14 +- weed/stats/metrics.go | 36 +++ 4 files changed, 392 insertions(+), 13 deletions(-) diff --git a/other/metrics/grafana_seaweedfs.json b/other/metrics/grafana_seaweedfs.json index e8c98ae21..147677294 100644 --- a/other/metrics/grafana_seaweedfs.json +++ b/other/metrics/grafana_seaweedfs.json @@ -459,6 +459,170 @@ "show": true } ] + }, + { + "aliasColors": {}, + "bars": false, + "dashLength": 10, + "dashes": false, + "datasource": "Prometheus", + "fill": 1, + "gridPos": { + "h": 6, + "w": 12, + "x": 0, + "y": 25 + }, + "id": 86, + "legend": { + "avg": false, + "current": true, + "max": false, + "min": false, + "show": true, + "total": false, + "values": true + }, + "lines": true, + "linewidth": 1, + "links": [], + "nullPointMode": "null", + "percentage": false, + "pointradius": 5, + "points": false, + "renderer": "flot", + "seriesOverrides": [], + "spaceLength": 10, + "stack": false, + "steppedLine": false, + "targets": [ + { + "expr": "SeaweedFS_filer_in_flight_upload_count", + "format": "time_series", + "intervalFactor": 2, + "legendFormat": "uploads", + "refId": "A", + "step": 30 + } + ], + "thresholds": [], + "timeFrom": null, + "timeShift": null, + "title": "Filer Concurrent Uploads", + "tooltip": { + "msResolution": true, + "shared": true, + "sort": 2, + "value_type": "individual" + }, + "type": "graph", + "xaxis": { + "buckets": null, + "mode": "time", + "name": null, + "show": true, + "values": [] + }, + "yaxes": [ + { + "format": "short", + "label": null, + "logBase": 1, + "max": null, + "min": "0", + "show": true + }, + { + "format": "short", + "label": null, + "logBase": 1, + "max": null, + "min": null, + "show": true + } + ] + }, + { + "aliasColors": {}, + "bars": false, + "dashLength": 10, + "dashes": false, + "datasource": "Prometheus", + "fill": 1, + "gridPos": { + "h": 6, + "w": 12, + "x": 12, + "y": 25 + }, + "id": 87, + "legend": { + "avg": false, + "current": true, + "max": false, + "min": false, + "show": true, + "total": false, + "values": true + }, + "lines": true, + "linewidth": 1, + "links": [], + "nullPointMode": "null", + "percentage": false, + "pointradius": 5, + "points": false, + "renderer": "flot", + "seriesOverrides": [], + "spaceLength": 10, + "stack": false, + "steppedLine": false, + "targets": [ + { + "expr": "SeaweedFS_filer_in_flight_upload_bytes", + "format": "time_series", + "intervalFactor": 2, + "legendFormat": "bytes", + "refId": "A", + "step": 30 + } + ], + "thresholds": [], + "timeFrom": null, + "timeShift": null, + "title": "Filer Concurrent Upload Bytes", + "tooltip": { + "msResolution": true, + "shared": true, + "sort": 2, + "value_type": "individual" + }, + "type": "graph", + "xaxis": { + "buckets": null, + "mode": "time", + "name": null, + "show": true, + "values": [] + }, + "yaxes": [ + { + "format": "bytes", + "label": null, + "logBase": 1, + "max": null, + "min": "0", + "show": true + }, + { + "format": "short", + "label": null, + "logBase": 1, + "max": null, + "min": null, + "show": true + } + ] } ], "repeat": null, @@ -479,7 +643,172 @@ "y": 1 }, "id": 70, - "panels": [], + "panels": [ + { + "aliasColors": {}, + "bars": false, + "dashLength": 10, + "dashes": false, + "datasource": "Prometheus", + "fill": 1, + "gridPos": { + "h": 6, + "w": 12, + "x": 0, + "y": 31 + }, + "id": 88, + "legend": { + "avg": false, + "current": true, + "max": false, + "min": false, + "show": true, + "total": false, + "values": true + }, + "lines": true, + "linewidth": 1, + "links": [], + "nullPointMode": "null", + "percentage": false, + "pointradius": 5, + "points": false, + "renderer": "flot", + "seriesOverrides": [], + "spaceLength": 10, + "stack": false, + "steppedLine": false, + "targets": [ + { + "expr": "SeaweedFS_s3_in_flight_upload_count", + "format": "time_series", + "intervalFactor": 2, + "legendFormat": "uploads", + "refId": "A", + "step": 30 + } + ], + "thresholds": [], + "timeFrom": null, + "timeShift": null, + "title": "S3 Concurrent Uploads", + "tooltip": { + "msResolution": true, + "shared": true, + "sort": 2, + "value_type": "individual" + }, + "type": "graph", + "xaxis": { + "buckets": null, + "mode": "time", + "name": null, + "show": true, + "values": [] + }, + "yaxes": [ + { + "format": "short", + "label": null, + "logBase": 1, + "max": null, + "min": "0", + "show": true + }, + { + "format": "short", + "label": null, + "logBase": 1, + "max": null, + "min": null, + "show": true + } + ] + }, + { + "aliasColors": {}, + "bars": false, + "dashLength": 10, + "dashes": false, + "datasource": "Prometheus", + "fill": 1, + "gridPos": { + "h": 6, + "w": 12, + "x": 12, + "y": 31 + }, + "id": 89, + "legend": { + "avg": false, + "current": true, + "max": false, + "min": false, + "show": true, + "total": false, + "values": true + }, + "lines": true, + "linewidth": 1, + "links": [], + "nullPointMode": "null", + "percentage": false, + "pointradius": 5, + "points": false, + "renderer": "flot", + "seriesOverrides": [], + "spaceLength": 10, + "stack": false, + "steppedLine": false, + "targets": [ + { + "expr": "SeaweedFS_s3_in_flight_upload_bytes", + "format": "time_series", + "intervalFactor": 2, + "legendFormat": "bytes", + "refId": "A", + "step": 30 + } + ], + "thresholds": [], + "timeFrom": null, + "timeShift": null, + "title": "S3 Concurrent Upload Bytes", + "tooltip": { + "msResolution": true, + "shared": true, + "sort": 2, + "value_type": "individual" + }, + "type": "graph", + "xaxis": { + "buckets": null, + "mode": "time", + "name": null, + "show": true, + "values": [] + }, + "yaxes": [ + { + "format": "bytes", + "label": null, + "logBase": 1, + "max": null, + "min": "0", + "show": true + }, + { + "format": "short", + "label": null, + "logBase": 1, + "max": null, + "min": null, + "show": true + } + ] + } + ], "repeat": null, "title": "S3 Gateway", "type": "row" @@ -2693,4 +3022,4 @@ "title": "SeaweedFS", "uid": "nh02dOVnz", "version": 2 -} +} \ No newline at end of file diff --git a/weed/s3api/s3api_circuit_breaker.go b/weed/s3api/s3api_circuit_breaker.go index 3c4f55a23..e19caf91b 100644 --- a/weed/s3api/s3api_circuit_breaker.go +++ b/weed/s3api/s3api_circuit_breaker.go @@ -3,6 +3,10 @@ package s3api import ( "errors" "fmt" + "net/http" + "sync" + "sync/atomic" + "github.com/gorilla/mux" "github.com/seaweedfs/seaweedfs/weed/filer" "github.com/seaweedfs/seaweedfs/weed/glog" @@ -11,9 +15,7 @@ import ( "github.com/seaweedfs/seaweedfs/weed/pb/s3_pb" "github.com/seaweedfs/seaweedfs/weed/s3api/s3_constants" "github.com/seaweedfs/seaweedfs/weed/s3api/s3err" - "net/http" - "sync" - "sync/atomic" + "github.com/seaweedfs/seaweedfs/weed/stats" ) type CircuitBreaker struct { @@ -121,12 +123,18 @@ func (cb *CircuitBreaker) Limit(f func(w http.ResponseWriter, r *http.Request), cb.s3a.inFlightDataLimitCond.L.Unlock() // Increment counters - atomic.AddInt64(&cb.s3a.inFlightUploads, 1) - atomic.AddInt64(&cb.s3a.inFlightDataSize, contentLength) + newUploads := atomic.AddInt64(&cb.s3a.inFlightUploads, 1) + newSize := atomic.AddInt64(&cb.s3a.inFlightDataSize, contentLength) + // Update metrics + stats.S3InFlightUploadCountGauge.Set(float64(newUploads)) + stats.S3InFlightUploadBytesGauge.Set(float64(newSize)) defer func() { // Decrement counters - atomic.AddInt64(&cb.s3a.inFlightUploads, -1) - atomic.AddInt64(&cb.s3a.inFlightDataSize, -contentLength) + newUploads := atomic.AddInt64(&cb.s3a.inFlightUploads, -1) + newSize := atomic.AddInt64(&cb.s3a.inFlightDataSize, -contentLength) + // Update metrics + stats.S3InFlightUploadCountGauge.Set(float64(newUploads)) + stats.S3InFlightUploadBytesGauge.Set(float64(newSize)) cb.s3a.inFlightDataLimitCond.Signal() }() } diff --git a/weed/server/filer_server_handlers.go b/weed/server/filer_server_handlers.go index a2eab9365..f068579ec 100644 --- a/weed/server/filer_server_handlers.go +++ b/weed/server/filer_server_handlers.go @@ -112,12 +112,18 @@ func (fs *FilerServer) filerHandler(w http.ResponseWriter, r *http.Request) { fs.inFlightDataLimitCond.L.Unlock() // Increment counters - atomic.AddInt64(&fs.inFlightUploads, 1) - atomic.AddInt64(&fs.inFlightDataSize, contentLength) + newUploads := atomic.AddInt64(&fs.inFlightUploads, 1) + newSize := atomic.AddInt64(&fs.inFlightDataSize, contentLength) + // Update metrics + stats.FilerInFlightUploadCountGauge.Set(float64(newUploads)) + stats.FilerInFlightUploadBytesGauge.Set(float64(newSize)) defer func() { // Decrement counters - atomic.AddInt64(&fs.inFlightUploads, -1) - atomic.AddInt64(&fs.inFlightDataSize, -contentLength) + newUploads := atomic.AddInt64(&fs.inFlightUploads, -1) + newSize := atomic.AddInt64(&fs.inFlightDataSize, -contentLength) + // Update metrics + stats.FilerInFlightUploadCountGauge.Set(float64(newUploads)) + stats.FilerInFlightUploadBytesGauge.Set(float64(newSize)) fs.inFlightDataLimitCond.Signal() }() diff --git a/weed/stats/metrics.go b/weed/stats/metrics.go index 95491abf1..914d59d28 100644 --- a/weed/stats/metrics.go +++ b/weed/stats/metrics.go @@ -167,6 +167,22 @@ var ( Help: "Current number of in-flight requests being handled by filer.", }, []string{"type"}) + FilerInFlightUploadBytesGauge = prometheus.NewGauge( + prometheus.GaugeOpts{ + Namespace: Namespace, + Subsystem: "filer", + Name: "in_flight_upload_bytes", + Help: "Current number of bytes being uploaded to filer.", + }) + + FilerInFlightUploadCountGauge = prometheus.NewGauge( + prometheus.GaugeOpts{ + Namespace: Namespace, + Subsystem: "filer", + Name: "in_flight_upload_count", + Help: "Current number of uploads in progress to filer.", + }) + FilerServerLastSendTsOfSubscribeGauge = prometheus.NewGaugeVec( prometheus.GaugeOpts{ Namespace: Namespace, @@ -371,6 +387,22 @@ var ( Help: "Current number of in-flight requests being handled by s3.", }, []string{"type"}) + S3InFlightUploadBytesGauge = prometheus.NewGauge( + prometheus.GaugeOpts{ + Namespace: Namespace, + Subsystem: "s3", + Name: "in_flight_upload_bytes", + Help: "Current number of bytes being uploaded to S3.", + }) + + S3InFlightUploadCountGauge = prometheus.NewGauge( + prometheus.GaugeOpts{ + Namespace: Namespace, + Subsystem: "s3", + Name: "in_flight_upload_count", + Help: "Current number of uploads in progress to S3.", + }) + S3BucketTrafficReceivedBytesCounter = prometheus.NewCounterVec( prometheus.CounterOpts{ Namespace: Namespace, @@ -422,6 +454,8 @@ func init() { Gather.MustRegister(FilerHandlerCounter) Gather.MustRegister(FilerRequestHistogram) Gather.MustRegister(FilerInFlightRequestsGauge) + Gather.MustRegister(FilerInFlightUploadBytesGauge) + Gather.MustRegister(FilerInFlightUploadCountGauge) Gather.MustRegister(FilerStoreCounter) Gather.MustRegister(FilerStoreHistogram) Gather.MustRegister(FilerSyncOffsetGauge) @@ -450,6 +484,8 @@ func init() { Gather.MustRegister(S3HandlerCounter) Gather.MustRegister(S3RequestHistogram) Gather.MustRegister(S3InFlightRequestsGauge) + Gather.MustRegister(S3InFlightUploadBytesGauge) + Gather.MustRegister(S3InFlightUploadCountGauge) Gather.MustRegister(S3TimeToFirstByteHistogram) Gather.MustRegister(S3BucketTrafficReceivedBytesCounter) Gather.MustRegister(S3BucketTrafficSentBytesCounter)