Add detailed scanner metrics (#15161)

cmd/admin-handlers.go

@@ -28,6 +28,7 @@ import (
 	"hash/crc32"
 	"io"
 	"io/ioutil"
+	"math"
 	"math/rand"
 	"net/http"
 	"net/url"
@@ -50,6 +51,7 @@ import (
 	"github.com/minio/minio/internal/kms"
 	"github.com/minio/minio/internal/logger"
 	"github.com/minio/minio/internal/logger/message/log"
+	"github.com/minio/minio/internal/pubsub"
 	iampolicy "github.com/minio/pkg/iam/policy"
 	xnet "github.com/minio/pkg/net"
 	"github.com/secure-io/sio-go"
@@ -343,6 +345,92 @@ func (a adminAPIHandlers) StorageInfoHandler(w http.ResponseWriter, r *http.Requ
 	writeSuccessResponseJSON(w, jsonBytes)
 }
 
+// MetricsHandler - GET /minio/admin/v3/metrics
+// ----------
+// Get realtime server metrics
+func (a adminAPIHandlers) MetricsHandler(w http.ResponseWriter, r *http.Request) {
+	ctx := newContext(r, w, "Metrics")
+
+	defer logger.AuditLog(ctx, w, r, mustGetClaimsFromToken(r))
+
+	objectAPI, _ := validateAdminReq(ctx, w, r, iampolicy.ServerInfoAdminAction)
+	if objectAPI == nil {
+		return
+	}
+	const defaultMetricsInterval = time.Second
+
+	interval, err := time.ParseDuration(r.Form.Get("interval"))
+	if err != nil || interval < time.Second {
+		interval = defaultMetricsInterval
+	}
+	n, err := strconv.Atoi(r.Form.Get("n"))
+	if err != nil || n <= 0 {
+		n = math.MaxInt32
+	}
+	var types madmin.MetricType
+	if t, _ := strconv.ParseUint(r.Form.Get("types"), 10, 64); t != 0 {
+		types = madmin.MetricType(t)
+	} else {
+		types = madmin.MetricsAll
+	}
+	hosts := strings.Split(r.Form.Get("hosts"), ",")
+	byhost := strings.EqualFold(r.Form.Get("by-host"), "true")
+	var hostMap map[string]struct{}
+	if len(hosts) > 0 && hosts[0] != "" {
+		hostMap = make(map[string]struct{}, len(hosts))
+		for _, k := range hosts {
+			if k != "" {
+				hostMap[k] = struct{}{}
+			}
+		}
+	}
+	done := ctx.Done()
+	ticker := time.NewTicker(interval)
+	defer ticker.Stop()
+	w.Header().Set(xhttp.ContentType, string(mimeJSON))
+
+	for n > 0 {
+		var m madmin.RealtimeMetrics
+		mLocal := collectLocalMetrics(types, hostMap)
+		m.Merge(&mLocal)
+
+		// Allow half the interval for collecting remote...
+		cctx, cancel := context.WithTimeout(ctx, interval/2)
+		mRemote := collectRemoteMetrics(cctx, types, hostMap)
+		cancel()
+		m.Merge(&mRemote)
+		if !byhost {
+			m.ByHost = nil
+		}
+
+		m.Final = n <= 1
+		// Marshal API response
+		jsonBytes, err := json.Marshal(m)
+		if err != nil {
+			writeErrorResponseJSON(ctx, w, toAdminAPIErr(ctx, err), r.URL)
+			return
+		}
+		_, err = w.Write(jsonBytes)
+		if err != nil {
+			n = 0
+		}
+
+		n--
+		if n <= 0 {
+			break
+		}
+
+		// Flush before waiting for next...
+		w.(http.Flusher).Flush()
+
+		select {
+		case <-ticker.C:
+		case <-done:
+			return
+		}
+	}
+}
+
 // DataUsageInfoHandler - GET /minio/admin/v3/datausage
 // ----------
 // Get server/cluster data usage info
@@ -1300,72 +1388,45 @@ const (
 // - input entry is not of the type *madmin.TraceInfo*
 // - errOnly entries are to be traced, not status code 2xx, 3xx.
 // - madmin.TraceInfo type is asked by opts
-func mustTrace(entry interface{}, opts madmin.ServiceTraceOpts) (shouldTrace bool) {
+func shouldTrace(trcInfo madmin.TraceInfo, opts madmin.ServiceTraceOpts) (shouldTrace bool) {
-	trcInfo, ok := entry.(madmin.TraceInfo)
+	// Reject all unwanted types.
-	if !ok {
+	want := opts.TraceTypes()
+	if !want.Contains(trcInfo.TraceType) {
 		return false
 	}
 
-	// Override shouldTrace decision with errOnly filtering
+	isHTTP := trcInfo.TraceType.Overlaps(madmin.TraceInternal|madmin.TraceS3) && trcInfo.HTTP != nil
-	defer func() {
-		if shouldTrace && opts.OnlyErrors {
-			shouldTrace = trcInfo.RespInfo.StatusCode >= http.StatusBadRequest
-		}
-	}()
 
-	if opts.Threshold > 0 {
+	// Check latency...
-		var latency time.Duration
+	if opts.Threshold > 0 && trcInfo.Duration < opts.Threshold {
-		switch trcInfo.TraceType {
-		case madmin.TraceOS:
-			latency = trcInfo.OSStats.Duration
-		case madmin.TraceStorage:
-			latency = trcInfo.StorageStats.Duration
-		case madmin.TraceHTTP:
-			latency = trcInfo.CallStats.Latency
-		}
-		if latency < opts.Threshold {
 		return false
 	}
+
+	// Check internal path
+	isInternal := isHTTP && HasPrefix(trcInfo.HTTP.ReqInfo.Path, minioReservedBucketPath+SlashSeparator)
+	if isInternal && !opts.Internal {
+		return false
+	}
+
+	// Filter non-errors.
+	if isHTTP && opts.OnlyErrors && trcInfo.HTTP.RespInfo.StatusCode < http.StatusBadRequest {
+		return false
 	}
 
-	if opts.Internal && trcInfo.TraceType == madmin.TraceHTTP && HasPrefix(trcInfo.ReqInfo.Path, minioReservedBucketPath+SlashSeparator) {
 	return true
-	}
-
-	if opts.S3 && trcInfo.TraceType == madmin.TraceHTTP && !HasPrefix(trcInfo.ReqInfo.Path, minioReservedBucketPath+SlashSeparator) {
-		return true
-	}
-
-	if opts.Storage && trcInfo.TraceType == madmin.TraceStorage {
-		return true
-	}
-
-	return opts.OS && trcInfo.TraceType == madmin.TraceOS
 }
 
 func extractTraceOptions(r *http.Request) (opts madmin.ServiceTraceOpts, err error) {
-	q := r.Form
+	if err := opts.ParseParams(r); err != nil {
-
+		return opts, err
-	opts.OnlyErrors = q.Get("err") == "true"
+	}
-	opts.S3 = q.Get("s3") == "true"
-	opts.Internal = q.Get("internal") == "true"
-	opts.Storage = q.Get("storage") == "true"
-	opts.OS = q.Get("os") == "true"
-
 	// Support deprecated 'all' query
-	if q.Get("all") == "true" {
+	if r.Form.Get("all") == "true" {
 		opts.S3 = true
 		opts.Internal = true
 		opts.Storage = true
 		opts.OS = true
-	}
+		// Older mc - cannot deal with more types...
-
-	if t := q.Get("threshold"); t != "" {
-		d, err := time.ParseDuration(t)
-		if err != nil {
-			return opts, err
-		}
-		opts.Threshold = d
 	}
 	return
 }
@@ -1388,20 +1449,21 @@ func (a adminAPIHandlers) TraceHandler(w http.ResponseWriter, r *http.Request) {
 		writeErrorResponseJSON(ctx, w, errorCodes.ToAPIErr(ErrInvalidRequest), r.URL)
 		return
 	}
-
 	setEventStreamHeaders(w)
 
 	// Trace Publisher and peer-trace-client uses nonblocking send and hence does not wait for slow receivers.
 	// Use buffered channel to take care of burst sends or slow w.Write()
-	traceCh := make(chan interface{}, 4000)
+	traceCh := make(chan pubsub.Maskable, 4000)
 
 	peers, _ := newPeerRestClients(globalEndpoints)
+	mask := pubsub.MaskFromMaskable(traceOpts.TraceTypes())
 
-	traceFn := func(entry interface{}) bool {
+	err = globalTrace.Subscribe(mask, traceCh, ctx.Done(), func(entry pubsub.Maskable) bool {
-		return mustTrace(entry, traceOpts)
+		if e, ok := entry.(madmin.TraceInfo); ok {
+			return shouldTrace(e, traceOpts)
 		}
-
+		return false
-	err = globalTrace.Subscribe(traceCh, ctx.Done(), traceFn)
+	})
 	if err != nil {
 		writeErrorResponseJSON(ctx, w, errorCodes.ToAPIErr(ErrSlowDown), r.URL)
 		return
@@ -1442,7 +1504,7 @@ func (a adminAPIHandlers) TraceHandler(w http.ResponseWriter, r *http.Request) {
 	}
 }
 
-// The handler sends console logs to the connected HTTP client.
+// The ConsoleLogHandler handler sends console logs to the connected HTTP client.
 func (a adminAPIHandlers) ConsoleLogHandler(w http.ResponseWriter, r *http.Request) {
 	ctx := newContext(r, w, "ConsoleLog")
 
@@ -1460,11 +1522,10 @@ func (a adminAPIHandlers) ConsoleLogHandler(w http.ResponseWriter, r *http.Reque
 		limitLines = 10
 	}
 
-	logKind := r.Form.Get("logType")
+	logKind := madmin.LogKind(strings.ToUpper(r.Form.Get("logType"))).LogMask()
-	if logKind == "" {
+	if logKind == 0 {
-		logKind = string(logger.All)
+		logKind = madmin.LogMaskAll
 	}
-	logKind = strings.ToUpper(logKind)
 
 	// Avoid reusing tcp connection if read timeout is hit
 	// This is needed to make r.Context().Done() work as
@@ -1473,7 +1534,7 @@ func (a adminAPIHandlers) ConsoleLogHandler(w http.ResponseWriter, r *http.Reque
 
 	setEventStreamHeaders(w)
 
-	logCh := make(chan interface{}, 4000)
+	logCh := make(chan pubsub.Maskable, 4000)
 
 	peers, _ := newPeerRestClients(globalEndpoints)
 

cmd/admin-router.go

@@ -66,6 +66,8 @@ func registerAdminRouter(router *mux.Router, enableConfigOps bool) {
 		adminRouter.Methods(http.MethodGet).Path(adminVersion + "/storageinfo").HandlerFunc(gz(httpTraceAll(adminAPI.StorageInfoHandler)))
 		// DataUsageInfo operations
 		adminRouter.Methods(http.MethodGet).Path(adminVersion + "/datausageinfo").HandlerFunc(gz(httpTraceAll(adminAPI.DataUsageInfoHandler)))
+		// Metrics operation
+		adminRouter.Methods(http.MethodGet).Path(adminVersion + "/metrics").HandlerFunc(gz(httpTraceAll(adminAPI.MetricsHandler)))
 
 		if globalIsDistErasure || globalIsErasure {
 			// Heal operations

cmd/background-heal-ops.go

@@ -22,6 +22,7 @@ import (
 	"runtime"
 
 	"github.com/minio/madmin-go"
+	"github.com/minio/minio/internal/pubsub"
 )
 
 // healTask represents what to heal along with options
@@ -49,10 +50,10 @@ type healRoutine struct {
 	workers int
 }
 
-func systemIO() int {
+func activeListeners() int {
 	// Bucket notification and http trace are not costly, it is okay to ignore them
 	// while counting the number of concurrent connections
-	return int(globalHTTPListen.NumSubscribers()) + int(globalTrace.NumSubscribers())
+	return int(globalHTTPListen.NumSubscribers(pubsub.MaskAll)) + int(globalTrace.NumSubscribers(pubsub.MaskAll))
 }
 
 func waitForLowHTTPReq() {
@@ -61,7 +62,7 @@ func waitForLowHTTPReq() {
 		currentIO = httpServer.GetRequestCount
 	}
 
-	globalHealConfig.Wait(currentIO, systemIO)
+	globalHealConfig.Wait(currentIO, activeListeners)
 }
 
 func initBackgroundHealing(ctx context.Context, objAPI ObjectLayer) {

cmd/bucket-stats.go

@@ -26,7 +26,7 @@ import (
 // ReplicationLatency holds information of bucket operations latency, such us uploads
 type ReplicationLatency struct {
 	// Single & Multipart PUTs latency
-	UploadHistogram LastMinuteLatencies
+	UploadHistogram LastMinuteHistogram
 }
 
 // Merge two replication latency into a new one
@@ -41,7 +41,7 @@ func (rl ReplicationLatency) getUploadLatency() (ret map[string]uint64) {
 	avg := rl.UploadHistogram.GetAvgData()
 	for k, v := range avg {
 		// Convert nanoseconds to milliseconds
-		ret[sizeTagToString(k)] = v.avg() / uint64(time.Millisecond)
+		ret[sizeTagToString(k)] = uint64(v.avg() / time.Millisecond)
 	}
 	return
 }

cmd/consolelogger.go

@@ -18,10 +18,11 @@
 package cmd
 
 import (
-	ring "container/ring"
+	"container/ring"
 	"context"
 	"sync"
 
+	"github.com/minio/madmin-go"
 	"github.com/minio/minio/internal/logger"
 	"github.com/minio/minio/internal/logger/message/log"
 	"github.com/minio/minio/internal/logger/target/console"
@@ -71,11 +72,11 @@ func (sys *HTTPConsoleLoggerSys) SetNodeName(nodeName string) {
 // HasLogListeners returns true if console log listeners are registered
 // for this node or peers
 func (sys *HTTPConsoleLoggerSys) HasLogListeners() bool {
-	return sys != nil && sys.pubsub.NumSubscribers() > 0
+	return sys != nil && sys.pubsub.NumSubscribers(madmin.LogMaskAll) > 0
 }
 
 // Subscribe starts console logging for this node.
-func (sys *HTTPConsoleLoggerSys) Subscribe(subCh chan interface{}, doneCh <-chan struct{}, node string, last int, logKind string, filter func(entry interface{}) bool) error {
+func (sys *HTTPConsoleLoggerSys) Subscribe(subCh chan pubsub.Maskable, doneCh <-chan struct{}, node string, last int, logKind madmin.LogMask, filter func(entry pubsub.Maskable) bool) error {
 	// Enable console logging for remote client.
 	if !sys.HasLogListeners() {
 		logger.AddSystemTarget(sys)
@@ -115,8 +116,7 @@ func (sys *HTTPConsoleLoggerSys) Subscribe(subCh chan interface{}, doneCh <-chan
 			}
 		}
 	}
-
+	return sys.pubsub.Subscribe(pubsub.MaskFromMaskable(madmin.LogMaskAll), subCh, doneCh, filter)
-	return sys.pubsub.Subscribe(subCh, doneCh, filter)
 }
 
 // Init if HTTPConsoleLoggerSys is valid, always returns nil right now
@@ -163,9 +163,9 @@ func (sys *HTTPConsoleLoggerSys) Type() types.TargetType {
 
 // Send log message 'e' to console and publish to console
 // log pubsub system
-func (sys *HTTPConsoleLoggerSys) Send(e interface{}, logKind string) error {
+func (sys *HTTPConsoleLoggerSys) Send(entry interface{}) error {
 	var lg log.Info
-	switch e := e.(type) {
+	switch e := entry.(type) {
 	case log.Entry:
 		lg = log.Info{Entry: e, NodeName: sys.nodeName}
 	case string:
@@ -179,5 +179,5 @@ func (sys *HTTPConsoleLoggerSys) Send(e interface{}, logKind string) error {
 	sys.logBuf = sys.logBuf.Next()
 	sys.Unlock()
 
-	return sys.console.Send(e, string(logger.All))
+	return sys.console.Send(entry, string(logger.All))
 }

cmd/data-scanner-metric.go

@@ -0,0 +1,315 @@
+package cmd
+
+import (
+	"strings"
+	"sync"
+	"sync/atomic"
+	"time"
+	"unsafe"
+
+	"github.com/minio/madmin-go"
+	"github.com/minio/minio/internal/bucket/lifecycle"
+)
+
+//go:generate stringer -type=scannerMetric -trimprefix=scannerMetric $GOFILE
+
+type scannerMetric uint8
+
+type scannerMetrics struct {
+	// All fields must be accessed atomically and aligned.
+	operations [scannerMetricLast]uint64
+	latency    [scannerMetricLastRealtime]lockedLastMinuteLatency
+
+	// actions records actions performed.
+	actions        [lifecycle.ActionCount]uint64
+	actionsLatency [lifecycle.ActionCount]lockedLastMinuteLatency
+
+	// currentPaths contains (string,*currentPathTracker) for each disk processing.
+	// Alignment not required.
+	currentPaths sync.Map
+
+	cycleInfoMu sync.Mutex
+	cycleInfo   *currentScannerCycle
+}
+
+var globalScannerMetrics scannerMetrics
+
+const (
+	// START Realtime metrics, that only to records
+	// last minute latencies and total operation count.
+	scannerMetricReadMetadata scannerMetric = iota
+	scannerMetricCheckMissing
+	scannerMetricSaveUsage
+	scannerMetricApplyAll
+	scannerMetricApplyVersion
+	scannerMetricTierObjSweep
+	scannerMetricHealCheck
+	scannerMetricILM
+	scannerMetricCheckReplication
+	scannerMetricYield
+
+	// START Trace metrics:
+	scannerMetricStartTrace
+	scannerMetricScanObject // Scan object. All operations included.
+
+	// END realtime metrics:
+	scannerMetricLastRealtime
+
+	// Trace only metrics:
+	scannerMetricScanFolder     // Scan a folder on disk, recursively.
+	scannerMetricScanCycle      // Full cycle, cluster global
+	scannerMetricScanBucketDisk // Single bucket on one disk
+
+	// Must be last:
+	scannerMetricLast
+)
+
+// log scanner action.
+// Use for s > scannerMetricStartTrace
+func (p *scannerMetrics) log(s scannerMetric, paths ...string) func() {
+	startTime := time.Now()
+	return func() {
+		duration := time.Since(startTime)
+
+		atomic.AddUint64(&p.operations[s], 1)
+		if s < scannerMetricLastRealtime {
+			p.latency[s].add(duration)
+		}
+
+		if s > scannerMetricStartTrace && globalTrace.NumSubscribers(madmin.TraceScanner) > 0 {
+			globalTrace.Publish(scannerTrace(s, startTime, duration, strings.Join(paths, " ")))
+		}
+	}
+}
+
+// time a scanner action.
+// Use for s < scannerMetricLastRealtime
+func (p *scannerMetrics) time(s scannerMetric) func() {
+	startTime := time.Now()
+	return func() {
+		duration := time.Since(startTime)
+
+		atomic.AddUint64(&p.operations[s], 1)
+		if s < scannerMetricLastRealtime {
+			p.latency[s].add(duration)
+		}
+	}
+}
+
+// timeSize add time and size of a scanner action.
+// Use for s < scannerMetricLastRealtime
+func (p *scannerMetrics) timeSize(s scannerMetric) func(sz int) {
+	startTime := time.Now()
+	return func(sz int) {
+		duration := time.Since(startTime)
+
+		atomic.AddUint64(&p.operations[s], 1)
+		if s < scannerMetricLastRealtime {
+			p.latency[s].addSize(duration, int64(sz))
+		}
+	}
+}
+
+// incTime will increment time on metric s with a specific duration.
+// Use for s < scannerMetricLastRealtime
+func (p *scannerMetrics) incTime(s scannerMetric, d time.Duration) {
+	atomic.AddUint64(&p.operations[s], 1)
+	if s < scannerMetricLastRealtime {
+		p.latency[s].add(d)
+	}
+}
+
+func (p *scannerMetrics) incNoTime(s scannerMetric) {
+	atomic.AddUint64(&p.operations[s], 1)
+	if s < scannerMetricLastRealtime {
+		p.latency[s].add(0)
+	}
+}
+
+// timeILM times an ILM action.
+// lifecycle.NoneAction is ignored.
+// Use for s < scannerMetricLastRealtime
+func (p *scannerMetrics) timeILM(a lifecycle.Action) func() {
+	if a == lifecycle.NoneAction || a >= lifecycle.ActionCount {
+		return func() {}
+	}
+	startTime := time.Now()
+	return func() {
+		duration := time.Since(startTime)
+		atomic.AddUint64(&p.actions[a], 1)
+		p.actionsLatency[a].add(duration)
+	}
+}
+
+type currentPathTracker struct {
+	name *unsafe.Pointer // contains atomically accessed *string
+}
+
+// currentPathUpdater provides a lightweight update function for keeping track of
+// current objects for each disk.
+// Returns a function that can be used to update the current object
+// and a function to call to when processing finished.
+func (p *scannerMetrics) currentPathUpdater(disk, initial string) (update func(path string), done func()) {
+	initialPtr := unsafe.Pointer(&initial)
+	tracker := &currentPathTracker{
+		name: &initialPtr,
+	}
+
+	p.currentPaths.Store(disk, tracker)
+	return func(path string) {
+			atomic.StorePointer(tracker.name, unsafe.Pointer(&path))
+		}, func() {
+			p.currentPaths.Delete(disk)
+		}
+}
+
+// getCurrentPaths returns the paths currently being processed.
+func (p *scannerMetrics) getCurrentPaths() []string {
+	var res []string
+	prefix := globalMinioAddr + "/"
+	p.currentPaths.Range(func(key, value interface{}) bool {
+		// We are a bit paranoid, but better miss an entry than crash.
+		name, ok := key.(string)
+		if !ok {
+			return true
+		}
+		obj, ok := value.(*currentPathTracker)
+		if !ok {
+			return true
+		}
+		strptr := (*string)(atomic.LoadPointer(obj.name))
+		if strptr != nil {
+			res = append(res, prefix+name+"/"+*strptr)
+		}
+		return true
+	})
+	return res
+}
+
+// activeDisks returns the number of currently active disks.
+// (since this is concurrent it may not be 100% reliable)
+func (p *scannerMetrics) activeDisks() int {
+	var i int
+	p.currentPaths.Range(func(k, v interface{}) bool {
+		i++
+		return true
+	})
+	return i
+}
+
+// lifetime returns the lifetime count of the specified metric.
+func (p *scannerMetrics) lifetime(m scannerMetric) uint64 {
+	if m < 0 || m >= scannerMetricLast {
+		return 0
+	}
+	val := atomic.LoadUint64(&p.operations[m])
+	return val
+}
+
+// lastMinute returns the last minute statistics of a metric.
+// m should be < scannerMetricLastRealtime
+func (p *scannerMetrics) lastMinute(m scannerMetric) AccElem {
+	if m < 0 || m >= scannerMetricLastRealtime {
+		return AccElem{}
+	}
+	val := p.latency[m].total()
+	return val
+}
+
+// lifetimeActions returns the lifetime count of the specified ilm metric.
+func (p *scannerMetrics) lifetimeActions(a lifecycle.Action) uint64 {
+	if a == lifecycle.NoneAction || a >= lifecycle.ActionCount {
+		return 0
+	}
+	val := atomic.LoadUint64(&p.actions[a])
+	return val
+}
+
+// lastMinuteActions returns the last minute statistics of an ilm metric.
+func (p *scannerMetrics) lastMinuteActions(a lifecycle.Action) AccElem {
+	if a == lifecycle.NoneAction || a >= lifecycle.ActionCount {
+		return AccElem{}
+	}
+	val := p.actionsLatency[a].total()
+	return val
+}
+
+// setCycle updates the current cycle metrics.
+func (p *scannerMetrics) setCycle(c *currentScannerCycle) {
+	if c != nil {
+		c2 := c.clone()
+		c = &c2
+	}
+	p.cycleInfoMu.Lock()
+	p.cycleInfo = c
+	p.cycleInfoMu.Unlock()
+}
+
+// getCycle returns the current cycle metrics.
+// If not nil, the returned value can safely be modified.
+func (p *scannerMetrics) getCycle() *currentScannerCycle {
+	p.cycleInfoMu.Lock()
+	defer p.cycleInfoMu.Unlock()
+	if p.cycleInfo == nil {
+		return nil
+	}
+	c := p.cycleInfo.clone()
+	return &c
+}
+
+func (p *scannerMetrics) report() madmin.ScannerMetrics {
+	var m madmin.ScannerMetrics
+	cycle := p.getCycle()
+	if cycle != nil {
+		m.CurrentCycle = cycle.current
+		m.CyclesCompletedAt = cycle.cycleCompleted
+		m.CurrentStarted = cycle.started
+	}
+	m.CollectedAt = time.Now()
+	m.ActivePaths = p.getCurrentPaths()
+	m.LifeTimeOps = make(map[string]uint64, scannerMetricLast)
+	for i := scannerMetric(0); i < scannerMetricLast; i++ {
+		if n := atomic.LoadUint64(&p.operations[i]); n > 0 {
+			m.LifeTimeOps[i.String()] = n
+		}
+	}
+	if len(m.LifeTimeOps) == 0 {
+		m.LifeTimeOps = nil
+	}
+
+	m.LastMinute.Actions = make(map[string]madmin.TimedAction, scannerMetricLastRealtime)
+	for i := scannerMetric(0); i < scannerMetricLastRealtime; i++ {
+		lm := p.lastMinute(i)
+		if lm.N > 0 {
+			m.LastMinute.Actions[i.String()] = lm.asTimedAction()
+		}
+	}
+	if len(m.LastMinute.Actions) == 0 {
+		m.LastMinute.Actions = nil
+	}
+
+	// ILM
+	m.LifeTimeILM = make(map[string]uint64)
+	for i := lifecycle.NoneAction + 1; i < lifecycle.ActionCount; i++ {
+		if n := atomic.LoadUint64(&p.actions[i]); n > 0 {
+			m.LifeTimeILM[i.String()] = n
+		}
+	}
+	if len(m.LifeTimeILM) == 0 {
+		m.LifeTimeILM = nil
+	}
+
+	if len(m.LifeTimeILM) > 0 {
+		m.LastMinute.ILM = make(map[string]madmin.TimedAction, len(m.LifeTimeILM))
+		for i := lifecycle.NoneAction + 1; i < lifecycle.ActionCount; i++ {
+			lm := p.lastMinuteActions(i)
+			if lm.N > 0 {
+				m.LastMinute.ILM[i.String()] = madmin.TimedAction{Count: uint64(lm.N), AccTime: uint64(lm.Total)}
+			}
+		}
+		if len(m.LastMinute.ILM) == 0 {
+			m.LastMinute.ILM = nil
+		}
+	}
+	return m
+}

cmd/data-scanner.go

@@ -31,7 +31,6 @@ import (
 	"path"
 	"strings"
 	"sync"
-	"sync/atomic"
 	"time"
 
 	"github.com/bits-and-blooms/bloom/v3"
@@ -66,7 +65,7 @@ var (
 
 	dataScannerLeaderLockTimeout = newDynamicTimeout(30*time.Second, 10*time.Second)
 	// Sleeper values are updated when config is loaded.
-	scannerSleeper = newDynamicSleeper(10, 10*time.Second)
+	scannerSleeper = newDynamicSleeper(10, 10*time.Second, true)
 	scannerCycle   = uatomic.NewDuration(dataScannerStartDelay)
 )
 
@@ -162,35 +161,45 @@ func runDataScanner(pctx context.Context, objAPI ObjectLayer) {
 	// No unlock for "leader" lock.
 
 	// Load current bloom cycle
-	nextBloomCycle := intDataUpdateTracker.current() + 1
+	var cycleInfo currentScannerCycle
+	cycleInfo.next = intDataUpdateTracker.current() + 1
 
 	buf, _ := readConfig(ctx, objAPI, dataUsageBloomNamePath)
-	if len(buf) >= 8 {
+	if len(buf) == 8 {
-		if err = binary.Read(bytes.NewReader(buf), binary.LittleEndian, &nextBloomCycle); err != nil {
+		cycleInfo.next = binary.LittleEndian.Uint64(buf)
+	} else if len(buf) > 8 {
+		cycleInfo.next = binary.LittleEndian.Uint64(buf[:8])
+		buf = buf[8:]
+		_, err = cycleInfo.UnmarshalMsg(buf)
 		logger.LogIf(ctx, err)
 	}
-	}
 
 	scannerTimer := time.NewTimer(scannerCycle.Load())
 	defer scannerTimer.Stop()
+	defer globalScannerMetrics.setCycle(nil)
 
 	for {
 		select {
 		case <-ctx.Done():
 			return
 		case <-scannerTimer.C:
-			if intDataUpdateTracker.debug {
+			// Reset the timer for next cycle.
-				console.Debugln("starting scanner cycle")
+			// If scanner takes longer we start at once.
-			}
+			scannerTimer.Reset(scannerCycle.Load())
+
+			stopFn := globalScannerMetrics.log(scannerMetricScanCycle)
+			cycleInfo.current = cycleInfo.next
+			cycleInfo.started = time.Now()
+			globalScannerMetrics.setCycle(&cycleInfo)
 
 			bgHealInfo := readBackgroundHealInfo(ctx, objAPI)
-			scanMode := getCycleScanMode(nextBloomCycle, bgHealInfo.BitrotStartCycle, bgHealInfo.BitrotStartTime)
+			scanMode := getCycleScanMode(cycleInfo.current, bgHealInfo.BitrotStartCycle, bgHealInfo.BitrotStartTime)
 			if bgHealInfo.CurrentScanMode != scanMode {
 				newHealInfo := bgHealInfo
 				newHealInfo.CurrentScanMode = scanMode
 				if scanMode == madmin.HealDeepScan {
 					newHealInfo.BitrotStartTime = time.Now().UTC()
-					newHealInfo.BitrotStartCycle = nextBloomCycle
+					newHealInfo.BitrotStartCycle = cycleInfo.current
 				}
 				saveBackgroundHealInfo(ctx, objAPI, newHealInfo)
 			}
@@ -198,23 +207,28 @@ func runDataScanner(pctx context.Context, objAPI ObjectLayer) {
 			// Wait before starting next cycle and wait on startup.
 			results := make(chan DataUsageInfo, 1)
 			go storeDataUsageInBackend(ctx, objAPI, results)
-			bf, err := globalNotificationSys.updateBloomFilter(ctx, nextBloomCycle)
+			bf, err := globalNotificationSys.updateBloomFilter(ctx, cycleInfo.current)
 			logger.LogIf(ctx, err)
-			err = objAPI.NSScanner(ctx, bf, results, uint32(nextBloomCycle), scanMode)
+			err = objAPI.NSScanner(ctx, bf, results, uint32(cycleInfo.current), scanMode)
 			logger.LogIf(ctx, err)
+			stopFn()
 			if err == nil {
 				// Store new cycle...
-				nextBloomCycle++
+				cycleInfo.next++
-				var tmp [8]byte
+				cycleInfo.current = 0
-				binary.LittleEndian.PutUint64(tmp[:], nextBloomCycle)
+				cycleInfo.cycleCompleted = append(cycleInfo.cycleCompleted, time.Now())
-				if err = saveConfig(ctx, objAPI, dataUsageBloomNamePath, tmp[:]); err != nil {
+				if len(cycleInfo.cycleCompleted) > dataUsageUpdateDirCycles {
+					cycleInfo.cycleCompleted = cycleInfo.cycleCompleted[len(cycleInfo.cycleCompleted)-dataUsageUpdateDirCycles:]
+				}
+				globalScannerMetrics.setCycle(&cycleInfo)
+				tmp := make([]byte, 8, 8+cycleInfo.Msgsize())
+				// Cycle for backward compat.
+				binary.LittleEndian.PutUint64(tmp, cycleInfo.next)
+				tmp, _ = cycleInfo.MarshalMsg(tmp)
+				err = saveConfig(ctx, objAPI, dataUsageBloomNamePath, tmp)
 				logger.LogIf(ctx, err)
 			}
 		}
-
-			// Reset the timer for next cycle.
-			scannerTimer.Reset(scannerCycle.Load())
-		}
 	}
 }
 
@@ -244,24 +258,11 @@ type folderScanner struct {
 	// Will not be closed when returned.
 	updates    chan<- dataUsageEntry
 	lastUpdate time.Time
+
+	// updateCurrentPath should be called whenever a new path is scanned.
+	updateCurrentPath func(string)
 }
 
-type scannerStats struct {
-	// All fields must be accessed atomically and aligned.
-
-	accTotalObjects  uint64
-	accTotalVersions uint64
-	accFolders       uint64
-	bucketsStarted   uint64
-	bucketsFinished  uint64
-	ilmChecks        uint64
-
-	// actions records actions performed.
-	actions [lifecycle.ActionCount]uint64
-}
-
-var globalScannerStats scannerStats
-
 // Cache structure and compaction:
 //
 // A cache structure will be kept with a tree of usages.
@@ -305,24 +306,14 @@ var globalScannerStats scannerStats
 // Before each operation sleepDuration is called which can be used to temporarily halt the scanner.
 // If the supplied context is canceled the function will return at the first chance.
 func scanDataFolder(ctx context.Context, poolIdx, setIdx int, basePath string, cache dataUsageCache, getSize getSizeFn, scanMode madmin.HealScanMode) (dataUsageCache, error) {
-	t := UTCNow()
-
 	logPrefix := color.Green("data-usage: ")
-	logSuffix := color.Blue("- %v + %v", basePath, cache.Info.Name)
-	atomic.AddUint64(&globalScannerStats.bucketsStarted, 1)
-	defer func() {
-		atomic.AddUint64(&globalScannerStats.bucketsFinished, 1)
-	}()
-	if intDataUpdateTracker.debug {
-		defer func() {
-			console.Debugf(logPrefix+" Scanner time: %v %s\n", time.Since(t), logSuffix)
-		}()
-	}
 
 	switch cache.Info.Name {
 	case "", dataUsageRoot:
 		return cache, errors.New("internal error: root scan attempted")
 	}
+	updatePath, closeDisk := globalScannerMetrics.currentPathUpdater(basePath, cache.Info.Name)
+	defer closeDisk()
 
 	s := folderScanner{
 		root:                  basePath,
@@ -335,6 +326,7 @@ func scanDataFolder(ctx context.Context, poolIdx, setIdx int, basePath string, c
 		healObjectSelect:      0,
 		scanMode:              scanMode,
 		updates:               cache.Info.updates,
+		updateCurrentPath:     updatePath,
 	}
 
 	// Add disks for set healing.
@@ -366,14 +358,8 @@ func scanDataFolder(ctx context.Context, poolIdx, setIdx int, basePath string, c
 			s.withFilter = nil
 		}
 	}
-	if s.dataUsageScannerDebug {
-		console.Debugf(logPrefix+"Start scanning. Bloom filter: %v %s\n", s.withFilter != nil, logSuffix)
-	}
 
 	done := ctx.Done()
-	if s.dataUsageScannerDebug {
-		console.Debugf(logPrefix+"Cycle: %v, Entries: %v %s\n", cache.Info.NextCycle, len(cache.Cache), logSuffix)
-	}
 
 	// Read top level in bucket.
 	select {
@@ -389,9 +375,6 @@ func scanDataFolder(ctx context.Context, poolIdx, setIdx int, basePath string, c
 		return cache, err
 	}
 
-	if s.dataUsageScannerDebug {
-		console.Debugf(logPrefix+"Finished scanner, %v entries (%+v) %s \n", len(s.newCache.Cache), *s.newCache.sizeRecursive(s.newCache.Info.Name), logSuffix)
-	}
 	s.newCache.Info.LastUpdate = UTCNow()
 	s.newCache.Info.NextCycle = cache.Info.NextCycle
 	return s.newCache, nil
@@ -420,10 +403,10 @@ func (f *folderScanner) sendUpdate() {
 func (f *folderScanner) scanFolder(ctx context.Context, folder cachedFolder, into *dataUsageEntry) error {
 	done := ctx.Done()
 	scannerLogPrefix := color.Green("folder-scanner:")
+
 	thisHash := hashPath(folder.name)
 	// Store initial compaction state.
 	wasCompacted := into.Compacted
-	atomic.AddUint64(&globalScannerStats.accFolders, 1)
 
 	for {
 		select {
@@ -648,7 +631,10 @@ func (f *folderScanner) scanFolder(ctx context.Context, folder cachedFolder, int
 					f.updateCache.replaceHashed(h, &thisHash, dataUsageEntry{})
 				}
 			}
+			f.updateCurrentPath(folder.name)
+			stopFn := globalScannerMetrics.log(scannerMetricScanFolder, f.root, folder.name)
 			scanFolder(folder)
+			stopFn()
 			// Add new folders if this is new and we don't have existing.
 			if !into.Compacted {
 				parent := f.updateCache.find(thisHash.Key())
@@ -676,7 +662,10 @@ func (f *folderScanner) scanFolder(ctx context.Context, folder cachedFolder, int
 					folder.objectHealProbDiv = f.healFolderInclude
 				}
 			}
+			f.updateCurrentPath(folder.name)
+			stopFn := globalScannerMetrics.log(scannerMetricScanFolder, f.root, folder.name, "EXISTING")
 			scanFolder(folder)
+			stopFn()
 		}
 
 		// Scan for healing
@@ -717,9 +706,8 @@ func (f *folderScanner) scanFolder(ctx context.Context, folder cachedFolder, int
 		healObjectsPrefix := color.Green("healObjects:")
 		for k := range abandonedChildren {
 			bucket, prefix := path2BucketObject(k)
-			if f.dataUsageScannerDebug {
+			stopFn := globalScannerMetrics.time(scannerMetricCheckMissing)
-				console.Debugf(scannerLogPrefix+" checking disappeared folder: %v/%v\n", bucket, prefix)
+			f.updateCurrentPath(k)
-			}
 
 			if bucket != resolver.bucket {
 				// Bucket might be missing as well with abandoned children.
@@ -807,6 +795,7 @@ func (f *folderScanner) scanFolder(ctx context.Context, folder cachedFolder, int
 				},
 			})
 
+			stopFn()
 			if f.dataUsageScannerDebug && err != nil && err != errFileNotFound {
 				console.Debugf(healObjectsPrefix+" checking returned value %v (%T)\n", err, err)
 			}
@@ -814,7 +803,9 @@ func (f *folderScanner) scanFolder(ctx context.Context, folder cachedFolder, int
 			// Add unless healing returned an error.
 			if foundObjs {
 				this := cachedFolder{name: k, parent: &thisHash, objectHealProbDiv: 1}
+				stopFn := globalScannerMetrics.log(scannerMetricScanFolder, f.root, this.name, "HEALED")
 				scanFolder(this)
+				stopFn()
 			}
 		}
 		break
@@ -964,7 +955,6 @@ func (i *scannerItem) applyLifecycle(ctx context.Context, o ObjectLayer, oi Obje
 		return false, size
 	}
 
-	atomic.AddUint64(&globalScannerStats.ilmChecks, 1)
 	versionID := oi.VersionID
 	rCfg, _ := globalBucketObjectLockSys.Get(i.bucket)
 	action := evalActionFromLifecycle(ctx, *i.lifeCycle, rCfg, oi, false)
@@ -975,7 +965,7 @@ func (i *scannerItem) applyLifecycle(ctx context.Context, o ObjectLayer, oi Obje
 			console.Debugf(applyActionsLogPrefix+" lifecycle: %q Initial scan: %v\n", i.objectPath(), action)
 		}
 	}
-	atomic.AddUint64(&globalScannerStats.actions[action], 1)
+	defer globalScannerMetrics.timeILM(action)
 
 	switch action {
 	case lifecycle.DeleteAction, lifecycle.DeleteVersionAction, lifecycle.DeleteRestoredAction, lifecycle.DeleteRestoredVersionAction:
@@ -1092,16 +1082,23 @@ func (i *scannerItem) applyVersionActions(ctx context.Context, o ObjectLayer, fi
 // The metadata will be compared to consensus on the object layer before any changes are applied.
 // If no metadata is supplied, -1 is returned if no action is taken.
 func (i *scannerItem) applyActions(ctx context.Context, o ObjectLayer, oi ObjectInfo, sizeS *sizeSummary) int64 {
+	done := globalScannerMetrics.time(scannerMetricILM)
 	applied, size := i.applyLifecycle(ctx, o, oi)
+	done()
+
 	// For instance, an applied lifecycle means we remove/transitioned an object
 	// from the current deployment, which means we don't have to call healing
 	// routine even if we are asked to do via heal flag.
 	if !applied {
 		if i.heal.enabled {
+			done := globalScannerMetrics.time(scannerMetricHealCheck)
 			size = i.applyHealing(ctx, o, oi)
+			done()
 		}
 		// replicate only if lifecycle rules are not applied.
+		done := globalScannerMetrics.time(scannerMetricCheckReplication)
 		i.healReplication(ctx, o, oi.Clone(), sizeS)
+		done()
 	}
 	return size
 }
@@ -1341,15 +1338,20 @@ type dynamicSleeper struct {
 
 	// cycle will be closed
 	cycle chan struct{}
+
+	// isScanner should be set when this is used by the scanner
+	// to record metrics.
+	isScanner bool
 }
 
 // newDynamicSleeper
-func newDynamicSleeper(factor float64, maxWait time.Duration) *dynamicSleeper {
+func newDynamicSleeper(factor float64, maxWait time.Duration, isScanner bool) *dynamicSleeper {
 	return &dynamicSleeper{
 		factor:    factor,
 		cycle:     make(chan struct{}),
 		maxSleep:  maxWait,
 		minSleep:  100 * time.Microsecond,
+		isScanner: isScanner,
 	}
 }
 
@@ -1379,15 +1381,24 @@ func (d *dynamicSleeper) Timer(ctx context.Context) func() {
 			select {
 			case <-ctx.Done():
 				if !timer.Stop() {
+					if d.isScanner {
+						globalScannerMetrics.incTime(scannerMetricYield, wantSleep)
+					}
 					<-timer.C
 				}
 				return
 			case <-timer.C:
+				if d.isScanner {
+					globalScannerMetrics.incTime(scannerMetricYield, wantSleep)
+				}
 				return
 			case <-cycle:
 				if !timer.Stop() {
 					// We expired.
 					<-timer.C
+					if d.isScanner {
+						globalScannerMetrics.incTime(scannerMetricYield, wantSleep)
+					}
 					return
 				}
 			}
@@ -1418,14 +1429,23 @@ func (d *dynamicSleeper) Sleep(ctx context.Context, base time.Duration) {
 		case <-ctx.Done():
 			if !timer.Stop() {
 				<-timer.C
+				if d.isScanner {
+					globalScannerMetrics.incTime(scannerMetricYield, wantSleep)
+				}
 			}
 			return
 		case <-timer.C:
+			if d.isScanner {
+				globalScannerMetrics.incTime(scannerMetricYield, wantSleep)
+			}
 			return
 		case <-cycle:
 			if !timer.Stop() {
 				// We expired.
 				<-timer.C
+				if d.isScanner {
+					globalScannerMetrics.incTime(scannerMetricYield, wantSleep)
+				}
 				return
 			}
 		}

cmd/data-update-tracker_test.go

@@ -63,7 +63,7 @@ func (t *testingLogger) Type() types.TargetType {
 	return types.TargetHTTP
 }
 
-func (t *testingLogger) Send(entry interface{}, errKind string) error {
+func (t *testingLogger) Send(entry interface{}) error {
 	t.mu.Lock()
 	defer t.mu.Unlock()
 	if t.t == nil {
@@ -75,7 +75,7 @@ func (t *testingLogger) Send(entry interface{}, errKind string) error {
 	}
 
 	t.t.Helper()
-	t.t.Log(e.Level, ":", errKind, e.Message)
+	t.t.Log(e.Level, ":", e.Message)
 	return nil
 }
 

cmd/data-usage-cache.go

@@ -1323,3 +1323,19 @@ func (z dataUsageHashMap) Msgsize() (s int) {
 	}
 	return
 }
+
+//msgp:encode ignore currentScannerCycle
+//msgp:decode ignore currentScannerCycle
+
+type currentScannerCycle struct {
+	current        uint64
+	next           uint64
+	started        time.Time
+	cycleCompleted []time.Time
+}
+
+// clone returns a clone.
+func (z currentScannerCycle) clone() currentScannerCycle {
+	z.cycleCompleted = append(make([]time.Time, 0, len(z.cycleCompleted)), z.cycleCompleted...)
+	return z
+}

cmd/data-usage-cache_gen.go

@@ -3,6 +3,8 @@ package cmd
 // Code generated by github.com/tinylib/msgp DO NOT EDIT.
 
 import (
+	"time"
+
 	"github.com/tinylib/msgp/msgp"
 )
 
@@ -284,6 +286,101 @@ func (z *allTierStats) Msgsize() (s int) {
 	return
 }
 
+// MarshalMsg implements msgp.Marshaler
+func (z *currentScannerCycle) MarshalMsg(b []byte) (o []byte, err error) {
+	o = msgp.Require(b, z.Msgsize())
+	// map header, size 4
+	// string "current"
+	o = append(o, 0x84, 0xa7, 0x63, 0x75, 0x72, 0x72, 0x65, 0x6e, 0x74)
+	o = msgp.AppendUint64(o, z.current)
+	// string "next"
+	o = append(o, 0xa4, 0x6e, 0x65, 0x78, 0x74)
+	o = msgp.AppendUint64(o, z.next)
+	// string "started"
+	o = append(o, 0xa7, 0x73, 0x74, 0x61, 0x72, 0x74, 0x65, 0x64)
+	o = msgp.AppendTime(o, z.started)
+	// string "cycleCompleted"
+	o = append(o, 0xae, 0x63, 0x79, 0x63, 0x6c, 0x65, 0x43, 0x6f, 0x6d, 0x70, 0x6c, 0x65, 0x74, 0x65, 0x64)
+	o = msgp.AppendArrayHeader(o, uint32(len(z.cycleCompleted)))
+	for za0001 := range z.cycleCompleted {
+		o = msgp.AppendTime(o, z.cycleCompleted[za0001])
+	}
+	return
+}
+
+// UnmarshalMsg implements msgp.Unmarshaler
+func (z *currentScannerCycle) UnmarshalMsg(bts []byte) (o []byte, err error) {
+	var field []byte
+	_ = field
+	var zb0001 uint32
+	zb0001, bts, err = msgp.ReadMapHeaderBytes(bts)
+	if err != nil {
+		err = msgp.WrapError(err)
+		return
+	}
+	for zb0001 > 0 {
+		zb0001--
+		field, bts, err = msgp.ReadMapKeyZC(bts)
+		if err != nil {
+			err = msgp.WrapError(err)
+			return
+		}
+		switch msgp.UnsafeString(field) {
+		case "current":
+			z.current, bts, err = msgp.ReadUint64Bytes(bts)
+			if err != nil {
+				err = msgp.WrapError(err, "current")
+				return
+			}
+		case "next":
+			z.next, bts, err = msgp.ReadUint64Bytes(bts)
+			if err != nil {
+				err = msgp.WrapError(err, "next")
+				return
+			}
+		case "started":
+			z.started, bts, err = msgp.ReadTimeBytes(bts)
+			if err != nil {
+				err = msgp.WrapError(err, "started")
+				return
+			}
+		case "cycleCompleted":
+			var zb0002 uint32
+			zb0002, bts, err = msgp.ReadArrayHeaderBytes(bts)
+			if err != nil {
+				err = msgp.WrapError(err, "cycleCompleted")
+				return
+			}
+			if cap(z.cycleCompleted) >= int(zb0002) {
+				z.cycleCompleted = (z.cycleCompleted)[:zb0002]
+			} else {
+				z.cycleCompleted = make([]time.Time, zb0002)
+			}
+			for za0001 := range z.cycleCompleted {
+				z.cycleCompleted[za0001], bts, err = msgp.ReadTimeBytes(bts)
+				if err != nil {
+					err = msgp.WrapError(err, "cycleCompleted", za0001)
+					return
+				}
+			}
+		default:
+			bts, err = msgp.Skip(bts)
+			if err != nil {
+				err = msgp.WrapError(err)
+				return
+			}
+		}
+	}
+	o = bts
+	return
+}
+
+// Msgsize returns an upper bound estimate of the number of bytes occupied by the serialized message
+func (z *currentScannerCycle) Msgsize() (s int) {
+	s = 1 + 8 + msgp.Uint64Size + 5 + msgp.Uint64Size + 8 + msgp.TimeSize + 15 + msgp.ArrayHeaderSize + (len(z.cycleCompleted) * (msgp.TimeSize))
+	return
+}
+
 // DecodeMsg implements msgp.Decodable
 func (z *dataUsageCache) DecodeMsg(dc *msgp.Reader) (err error) {
 	var field []byte

cmd/data-usage-cache_gen_test.go

@@ -122,6 +122,64 @@ func BenchmarkDecodeallTierStats(b *testing.B) {
 	}
 }
 
+func TestMarshalUnmarshalcurrentScannerCycle(t *testing.T) {
+	v := currentScannerCycle{}
+	bts, err := v.MarshalMsg(nil)
+	if err != nil {
+		t.Fatal(err)
+	}
+	left, err := v.UnmarshalMsg(bts)
+	if err != nil {
+		t.Fatal(err)
+	}
+	if len(left) > 0 {
+		t.Errorf("%d bytes left over after UnmarshalMsg(): %q", len(left), left)
+	}
+
+	left, err = msgp.Skip(bts)
+	if err != nil {
+		t.Fatal(err)
+	}
+	if len(left) > 0 {
+		t.Errorf("%d bytes left over after Skip(): %q", len(left), left)
+	}
+}
+
+func BenchmarkMarshalMsgcurrentScannerCycle(b *testing.B) {
+	v := currentScannerCycle{}
+	b.ReportAllocs()
+	b.ResetTimer()
+	for i := 0; i < b.N; i++ {
+		v.MarshalMsg(nil)
+	}
+}
+
+func BenchmarkAppendMsgcurrentScannerCycle(b *testing.B) {
+	v := currentScannerCycle{}
+	bts := make([]byte, 0, v.Msgsize())
+	bts, _ = v.MarshalMsg(bts[0:0])
+	b.SetBytes(int64(len(bts)))
+	b.ReportAllocs()
+	b.ResetTimer()
+	for i := 0; i < b.N; i++ {
+		bts, _ = v.MarshalMsg(bts[0:0])
+	}
+}
+
+func BenchmarkUnmarshalcurrentScannerCycle(b *testing.B) {
+	v := currentScannerCycle{}
+	bts, _ := v.MarshalMsg(nil)
+	b.ReportAllocs()
+	b.SetBytes(int64(len(bts)))
+	b.ResetTimer()
+	for i := 0; i < b.N; i++ {
+		_, err := v.UnmarshalMsg(bts)
+		if err != nil {
+			b.Fatal(err)
+		}
+	}
+}
+
 func TestMarshalUnmarshaldataUsageCache(t *testing.T) {
 	v := dataUsageCache{}
 	bts, err := v.MarshalMsg(nil)

cmd/erasure-sets.go

@@ -466,7 +466,7 @@ func newErasureSets(ctx context.Context, endpoints PoolEndpoints, storageDisks [
 				getDisks:              s.GetDisks(i),
 				getLockers:            s.GetLockers(i),
 				getEndpoints:          s.GetEndpoints(i),
-				deletedCleanupSleeper: newDynamicSleeper(10, 2*time.Second),
+				deletedCleanupSleeper: newDynamicSleeper(10, 2*time.Second, false),
 				nsMutex:               mutex,
 				bp:                    bp,
 				bpOld:                 bpOld,

cmd/erasure-single-drive.go

@@ -90,7 +90,7 @@ func newErasureSingle(ctx context.Context, storageDisk StorageAPI, format *forma
 		format:                format,
 		nsMutex:               newNSLock(false),
 		bp:                    bp,
-		deletedCleanupSleeper: newDynamicSleeper(10, 2*time.Second),
+		deletedCleanupSleeper: newDynamicSleeper(10, 2*time.Second, false),
 	}
 
 	// start cleanup stale uploads go-routine.

cmd/erasure.go

@@ -213,11 +213,13 @@ func getDisksInfo(disks []StorageAPI, endpoints []Endpoint) (disksInfo []madmin.
 				}
 			}
 			di.Metrics = &madmin.DiskMetrics{
-				APILatencies: make(map[string]interface{}),
+				LastMinute: make(map[string]madmin.TimedAction, len(info.Metrics.LastMinute)),
-				APICalls:     make(map[string]uint64),
+				APICalls:   make(map[string]uint64, len(info.Metrics.APICalls)),
+			}
+			for k, v := range info.Metrics.LastMinute {
+				if v.N > 0 {
+					di.Metrics.LastMinute[k] = v.asTimedAction()
 				}
-			for k, v := range info.Metrics.APILatencies {
-				di.Metrics.APILatencies[k] = v
 			}
 			for k, v := range info.Metrics.APICalls {
 				di.Metrics.APICalls[k] = v

cmd/fs-v1.go

@@ -31,7 +31,6 @@ import (
 	"sort"
 	"strings"
 	"sync"
-	"sync/atomic"
 	"time"
 
 	jsoniter "github.com/json-iterator/go"
@@ -349,6 +348,7 @@ func (fs *FSObjects) NSScanner(ctx context.Context, bf *bloomFilter, updates cha
 // A partially updated bucket may be returned.
 func (fs *FSObjects) scanBucket(ctx context.Context, bucket string, cache dataUsageCache) (dataUsageCache, error) {
 	defer close(cache.Info.updates)
+	defer globalScannerMetrics.log(scannerMetricScanBucketDisk, fs.fsPath, bucket)()
 
 	// Get bucket policy
 	// Check if the current bucket has a configured lifecycle policy
@@ -363,6 +363,16 @@ func (fs *FSObjects) scanBucket(ctx context.Context, bucket string, cache dataUs
 	// Load bucket info.
 	cache, err = scanDataFolder(ctx, -1, -1, fs.fsPath, cache, func(item scannerItem) (sizeSummary, error) {
 		bucket, object := item.bucket, item.objectPath()
+		stopFn := globalScannerMetrics.log(scannerMetricScanObject, fs.fsPath, PathJoin(item.bucket, item.objectPath()))
+		defer stopFn()
+
+		var fsMetaBytes []byte
+		done := globalScannerMetrics.timeSize(scannerMetricReadMetadata)
+		defer func() {
+			if done != nil {
+				done(len(fsMetaBytes))
+			}
+		}()
 		fsMetaBytes, err := xioutil.ReadFile(pathJoin(fs.fsPath, minioMetaBucket, bucketMetaPrefix, bucket, object, fs.metaJSONFile))
 		if err != nil && !osIsNotExist(err) {
 			if intDataUpdateTracker.debug {
@@ -391,11 +401,16 @@ func (fs *FSObjects) scanBucket(ctx context.Context, bucket string, cache dataUs
 			}
 			return sizeSummary{}, errSkipFile
 		}
+		done(len(fsMetaBytes))
+		done = nil
+
+		// FS has no "all versions". Increment the counter, though
+		globalScannerMetrics.incNoTime(scannerMetricApplyAll)
 
 		oi := fsMeta.ToObjectInfo(bucket, object, fi)
-		atomic.AddUint64(&globalScannerStats.accTotalVersions, 1)
+		doneVer := globalScannerMetrics.time(scannerMetricApplyVersion)
-		atomic.AddUint64(&globalScannerStats.accTotalObjects, 1)
 		sz := item.applyActions(ctx, fs, oi, &sizeSummary{})
+		doneVer()
 		if sz >= 0 {
 			return sizeSummary{totalSize: sz, versions: 1}, nil
 		}

cmd/globals.go

@@ -230,6 +230,7 @@ var (
 	globalTrace = pubsub.New(8)
 
 	// global Listen system to send S3 API events to registered listeners
+	// Objects are expected to be event.Event
 	globalHTTPListen = pubsub.New(0)
 
 	// global console system to send console logs to

cmd/http-tracer.go

@@ -125,7 +125,7 @@ type traceCtxt struct {
 // otherwise, generate a trace event with request information but no response.
 func httpTracer(h http.Handler) http.Handler {
 	return http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
-		if globalTrace.NumSubscribers() == 0 {
+		if globalTrace.NumSubscribers(madmin.TraceS3|madmin.TraceInternal) == 0 {
 			h.ServeHTTP(w, r)
 			return
 		}
@@ -149,6 +149,14 @@ func httpTracer(h http.Handler) http.Handler {
 		h.ServeHTTP(respRecorder, r)
 		reqEndTime := time.Now().UTC()
 
+		tt := madmin.TraceInternal
+		if strings.HasPrefix(tc.funcName, "s3.") {
+			tt = madmin.TraceS3
+		}
+		// No need to continue if no subscribers for actual type...
+		if globalTrace.NumSubscribers(tt) == 0 {
+			return
+		}
 		// Calculate input body size with headers
 		reqHeaders := r.Header.Clone()
 		reqHeaders.Set("Host", r.Host)
@@ -185,7 +193,15 @@ func httpTracer(h http.Handler) http.Handler {
 			funcName = "<unknown>"
 		}
 
-		rq := madmin.TraceRequestInfo{
+		t := madmin.TraceInfo{
+			TraceType: tt,
+			FuncName:  funcName,
+			NodeName:  nodeName,
+			Time:      reqStartTime,
+			Duration:  reqEndTime.Sub(respRecorder.StartTime),
+			Path:      reqPath,
+			HTTP: &madmin.TraceHTTPStats{
+				ReqInfo: madmin.TraceRequestInfo{
 					Time:     reqStartTime,
 					Proto:    r.Proto,
 					Method:   r.Method,
@@ -194,30 +210,20 @@ func httpTracer(h http.Handler) http.Handler {
 					Headers:  reqHeaders,
 					Path:     reqPath,
 					Body:     reqRecorder.Data(),
-		}
+				},
-
+				RespInfo: madmin.TraceResponseInfo{
-		rs := madmin.TraceResponseInfo{
 					Time:       reqEndTime,
 					Headers:    respRecorder.Header().Clone(),
 					StatusCode: respRecorder.StatusCode,
 					Body:       respRecorder.Body(),
-		}
+				},
-
+				CallStats: madmin.TraceCallStats{
-		cs := madmin.TraceCallStats{
+					Latency:         reqEndTime.Sub(respRecorder.StartTime),
-			Latency:         rs.Time.Sub(respRecorder.StartTime),
 					InputBytes:      inputBytes,
 					OutputBytes:     respRecorder.Size(),
 					TimeToFirstByte: respRecorder.TimeToFirstByte,
-		}
+				},
-
+			},
-		t := madmin.TraceInfo{
-			TraceType: madmin.TraceHTTP,
-			FuncName:  funcName,
-			NodeName:  nodeName,
-			Time:      reqStartTime,
-			ReqInfo:   rq,
-			RespInfo:  rs,
-			CallStats: cs,
 		}
 
 		globalTrace.Publish(t)

cmd/last-minute.go

@@ -21,6 +21,8 @@ package cmd
 
 import (
 	"time"
+
+	"github.com/minio/madmin-go"
 )
 
 const (
@@ -75,29 +77,49 @@ func sizeTagToString(tag int) string {
 // AccElem holds information for calculating an average value
 type AccElem struct {
 	Total int64
+	Size  int64
 	N     int64
 }
 
 // Add a duration to a single element.
 func (a *AccElem) add(dur time.Duration) {
+	if dur < 0 {
+		dur = 0
+	}
 	a.Total += int64(dur)
 	a.N++
 }
 
+// Add a duration to a single element.
+func (a *AccElem) addSize(dur time.Duration, sz int64) {
+	if dur < 0 {
+		dur = 0
+	}
+	a.Total += int64(dur)
+	a.Size += sz
+	a.N++
+}
+
 // Merge b into a.
 func (a *AccElem) merge(b AccElem) {
 	a.N += b.N
 	a.Total += b.Total
+	a.Size += b.Size
 }
 
-// Avg converts total to average.
+// Avg returns average time spent.
-func (a AccElem) avg() uint64 {
+func (a AccElem) avg() time.Duration {
 	if a.N >= 1 && a.Total > 0 {
-		return uint64(a.Total / a.N)
+		return time.Duration(a.Total / a.N)
 	}
 	return 0
 }
 
+// asTimedAction returns the element as a madmin.TimedAction.
+func (a AccElem) asTimedAction() madmin.TimedAction {
+	return madmin.TimedAction{AccTime: uint64(a.Total), Count: uint64(a.N), Bytes: uint64(a.Size)}
+}
+
 // lastMinuteLatency keeps track of last minute latency.
 type lastMinuteLatency struct {
 	Totals  [60]AccElem
@@ -118,6 +140,7 @@ func (l lastMinuteLatency) merge(o lastMinuteLatency) (merged lastMinuteLatency)
 		merged.Totals[i] = AccElem{
 			Total: l.Totals[i].Total + o.Totals[i].Total,
 			N:     l.Totals[i].N + o.Totals[i].N,
+			Size:  l.Totals[i].Size + o.Totals[i].Size,
 		}
 	}
 	return merged
@@ -132,8 +155,17 @@ func (l *lastMinuteLatency) add(t time.Duration) {
 	l.LastSec = sec
 }
 
+// Add  a new duration data
+func (l *lastMinuteLatency) addSize(t time.Duration, sz int64) {
+	sec := time.Now().Unix()
+	l.forwardTo(sec)
+	winIdx := sec % 60
+	l.Totals[winIdx].addSize(t, sz)
+	l.LastSec = sec
+}
+
 // Merge all recorded latencies of last minute into one
-func (l *lastMinuteLatency) getAvgData() AccElem {
+func (l *lastMinuteLatency) getTotal() AccElem {
 	var res AccElem
 	sec := time.Now().Unix()
 	l.forwardTo(sec)
@@ -160,11 +192,11 @@ func (l *lastMinuteLatency) forwardTo(t int64) {
 	}
 }
 
-// LastMinuteLatencies keeps track of last minute latencies.
+// LastMinuteHistogram keeps track of last minute sizes added.
-type LastMinuteLatencies [sizeLastElemMarker]lastMinuteLatency
+type LastMinuteHistogram [sizeLastElemMarker]lastMinuteLatency
 
-// Merge safely merges two LastMinuteLatencies structures into one
+// Merge safely merges two LastMinuteHistogram structures into one
-func (l LastMinuteLatencies) Merge(o LastMinuteLatencies) (merged LastMinuteLatencies) {
+func (l LastMinuteHistogram) Merge(o LastMinuteHistogram) (merged LastMinuteHistogram) {
 	for i := range l {
 		merged[i] = l[i].merge(o[i])
 	}
@@ -172,16 +204,16 @@ func (l LastMinuteLatencies) Merge(o LastMinuteLatencies) (merged LastMinuteLate
 }
 
 // Add latency t from object with the specified size.
-func (l *LastMinuteLatencies) Add(size int64, t time.Duration) {
+func (l *LastMinuteHistogram) Add(size int64, t time.Duration) {
 	l[sizeToTag(size)].add(t)
 }
 
 // GetAvgData will return the average for each bucket from the last time minute.
 // The number of objects is also included.
-func (l *LastMinuteLatencies) GetAvgData() [sizeLastElemMarker]AccElem {
+func (l *LastMinuteHistogram) GetAvgData() [sizeLastElemMarker]AccElem {
 	var res [sizeLastElemMarker]AccElem
 	for i, elem := range l[:] {
-		res[i] = elem.getAvgData()
+		res[i] = elem.getTotal()
 	}
 	return res
 }

cmd/last-minute_gen.go

@@ -30,6 +30,12 @@ func (z *AccElem) DecodeMsg(dc *msgp.Reader) (err error) {
 				err = msgp.WrapError(err, "Total")
 				return
 			}
+		case "Size":
+			z.Size, err = dc.ReadInt64()
+			if err != nil {
+				err = msgp.WrapError(err, "Size")
+				return
+			}
 		case "N":
 			z.N, err = dc.ReadInt64()
 			if err != nil {
@@ -49,9 +55,9 @@ func (z *AccElem) DecodeMsg(dc *msgp.Reader) (err error) {
 
 // EncodeMsg implements msgp.Encodable
 func (z AccElem) EncodeMsg(en *msgp.Writer) (err error) {
-	// map header, size 2
+	// map header, size 3
 	// write "Total"
-	err = en.Append(0x82, 0xa5, 0x54, 0x6f, 0x74, 0x61, 0x6c)
+	err = en.Append(0x83, 0xa5, 0x54, 0x6f, 0x74, 0x61, 0x6c)
 	if err != nil {
 		return
 	}
@@ -60,6 +66,16 @@ func (z AccElem) EncodeMsg(en *msgp.Writer) (err error) {
 		err = msgp.WrapError(err, "Total")
 		return
 	}
+	// write "Size"
+	err = en.Append(0xa4, 0x53, 0x69, 0x7a, 0x65)
+	if err != nil {
+		return
+	}
+	err = en.WriteInt64(z.Size)
+	if err != nil {
+		err = msgp.WrapError(err, "Size")
+		return
+	}
 	// write "N"
 	err = en.Append(0xa1, 0x4e)
 	if err != nil {
@@ -76,10 +92,13 @@ func (z AccElem) EncodeMsg(en *msgp.Writer) (err error) {
 // MarshalMsg implements msgp.Marshaler
 func (z AccElem) MarshalMsg(b []byte) (o []byte, err error) {
 	o = msgp.Require(b, z.Msgsize())
-	// map header, size 2
+	// map header, size 3
 	// string "Total"
-	o = append(o, 0x82, 0xa5, 0x54, 0x6f, 0x74, 0x61, 0x6c)
+	o = append(o, 0x83, 0xa5, 0x54, 0x6f, 0x74, 0x61, 0x6c)
 	o = msgp.AppendInt64(o, z.Total)
+	// string "Size"
+	o = append(o, 0xa4, 0x53, 0x69, 0x7a, 0x65)
+	o = msgp.AppendInt64(o, z.Size)
 	// string "N"
 	o = append(o, 0xa1, 0x4e)
 	o = msgp.AppendInt64(o, z.N)
@@ -110,6 +129,12 @@ func (z *AccElem) UnmarshalMsg(bts []byte) (o []byte, err error) {
 				err = msgp.WrapError(err, "Total")
 				return
 			}
+		case "Size":
+			z.Size, bts, err = msgp.ReadInt64Bytes(bts)
+			if err != nil {
+				err = msgp.WrapError(err, "Size")
+				return
+			}
 		case "N":
 			z.N, bts, err = msgp.ReadInt64Bytes(bts)
 			if err != nil {
@@ -130,12 +155,12 @@ func (z *AccElem) UnmarshalMsg(bts []byte) (o []byte, err error) {
 
 // Msgsize returns an upper bound estimate of the number of bytes occupied by the serialized message
 func (z AccElem) Msgsize() (s int) {
-	s = 1 + 6 + msgp.Int64Size + 2 + msgp.Int64Size
+	s = 1 + 6 + msgp.Int64Size + 5 + msgp.Int64Size + 2 + msgp.Int64Size
 	return
 }
 
 // DecodeMsg implements msgp.Decodable
-func (z *LastMinuteLatencies) DecodeMsg(dc *msgp.Reader) (err error) {
+func (z *LastMinuteHistogram) DecodeMsg(dc *msgp.Reader) (err error) {
 	var zb0001 uint32
 	zb0001, err = dc.ReadArrayHeader()
 	if err != nil {
@@ -195,6 +220,12 @@ func (z *LastMinuteLatencies) DecodeMsg(dc *msgp.Reader) (err error) {
 								err = msgp.WrapError(err, za0001, "Totals", za0002, "Total")
 								return
 							}
+						case "Size":
+							z[za0001].Totals[za0002].Size, err = dc.ReadInt64()
+							if err != nil {
+								err = msgp.WrapError(err, za0001, "Totals", za0002, "Size")
+								return
+							}
 						case "N":
 							z[za0001].Totals[za0002].N, err = dc.ReadInt64()
 							if err != nil {
@@ -229,7 +260,7 @@ func (z *LastMinuteLatencies) DecodeMsg(dc *msgp.Reader) (err error) {
 }
 
 // EncodeMsg implements msgp.Encodable
-func (z *LastMinuteLatencies) EncodeMsg(en *msgp.Writer) (err error) {
+func (z *LastMinuteHistogram) EncodeMsg(en *msgp.Writer) (err error) {
 	err = en.WriteArrayHeader(uint32(sizeLastElemMarker))
 	if err != nil {
 		err = msgp.WrapError(err)
@@ -248,9 +279,9 @@ func (z *LastMinuteLatencies) EncodeMsg(en *msgp.Writer) (err error) {
 			return
 		}
 		for za0002 := range z[za0001].Totals {
-			// map header, size 2
+			// map header, size 3
 			// write "Total"
-			err = en.Append(0x82, 0xa5, 0x54, 0x6f, 0x74, 0x61, 0x6c)
+			err = en.Append(0x83, 0xa5, 0x54, 0x6f, 0x74, 0x61, 0x6c)
 			if err != nil {
 				return
 			}
@@ -259,6 +290,16 @@ func (z *LastMinuteLatencies) EncodeMsg(en *msgp.Writer) (err error) {
 				err = msgp.WrapError(err, za0001, "Totals", za0002, "Total")
 				return
 			}
+			// write "Size"
+			err = en.Append(0xa4, 0x53, 0x69, 0x7a, 0x65)
+			if err != nil {
+				return
+			}
+			err = en.WriteInt64(z[za0001].Totals[za0002].Size)
+			if err != nil {
+				err = msgp.WrapError(err, za0001, "Totals", za0002, "Size")
+				return
+			}
 			// write "N"
 			err = en.Append(0xa1, 0x4e)
 			if err != nil {
@@ -285,7 +326,7 @@ func (z *LastMinuteLatencies) EncodeMsg(en *msgp.Writer) (err error) {
 }
 
 // MarshalMsg implements msgp.Marshaler
-func (z *LastMinuteLatencies) MarshalMsg(b []byte) (o []byte, err error) {
+func (z *LastMinuteHistogram) MarshalMsg(b []byte) (o []byte, err error) {
 	o = msgp.Require(b, z.Msgsize())
 	o = msgp.AppendArrayHeader(o, uint32(sizeLastElemMarker))
 	for za0001 := range z {
@@ -294,10 +335,13 @@ func (z *LastMinuteLatencies) MarshalMsg(b []byte) (o []byte, err error) {
 		o = append(o, 0x82, 0xa6, 0x54, 0x6f, 0x74, 0x61, 0x6c, 0x73)
 		o = msgp.AppendArrayHeader(o, uint32(60))
 		for za0002 := range z[za0001].Totals {
-			// map header, size 2
+			// map header, size 3
 			// string "Total"
-			o = append(o, 0x82, 0xa5, 0x54, 0x6f, 0x74, 0x61, 0x6c)
+			o = append(o, 0x83, 0xa5, 0x54, 0x6f, 0x74, 0x61, 0x6c)
 			o = msgp.AppendInt64(o, z[za0001].Totals[za0002].Total)
+			// string "Size"
+			o = append(o, 0xa4, 0x53, 0x69, 0x7a, 0x65)
+			o = msgp.AppendInt64(o, z[za0001].Totals[za0002].Size)
 			// string "N"
 			o = append(o, 0xa1, 0x4e)
 			o = msgp.AppendInt64(o, z[za0001].Totals[za0002].N)
@@ -310,7 +354,7 @@ func (z *LastMinuteLatencies) MarshalMsg(b []byte) (o []byte, err error) {
 }
 
 // UnmarshalMsg implements msgp.Unmarshaler
-func (z *LastMinuteLatencies) UnmarshalMsg(bts []byte) (o []byte, err error) {
+func (z *LastMinuteHistogram) UnmarshalMsg(bts []byte) (o []byte, err error) {
 	var zb0001 uint32
 	zb0001, bts, err = msgp.ReadArrayHeaderBytes(bts)
 	if err != nil {
@@ -370,6 +414,12 @@ func (z *LastMinuteLatencies) UnmarshalMsg(bts []byte) (o []byte, err error) {
 								err = msgp.WrapError(err, za0001, "Totals", za0002, "Total")
 								return
 							}
+						case "Size":
+							z[za0001].Totals[za0002].Size, bts, err = msgp.ReadInt64Bytes(bts)
+							if err != nil {
+								err = msgp.WrapError(err, za0001, "Totals", za0002, "Size")
+								return
+							}
 						case "N":
 							z[za0001].Totals[za0002].N, bts, err = msgp.ReadInt64Bytes(bts)
 							if err != nil {
@@ -405,8 +455,8 @@ func (z *LastMinuteLatencies) UnmarshalMsg(bts []byte) (o []byte, err error) {
 }
 
 // Msgsize returns an upper bound estimate of the number of bytes occupied by the serialized message
-func (z *LastMinuteLatencies) Msgsize() (s int) {
+func (z *LastMinuteHistogram) Msgsize() (s int) {
-	s = msgp.ArrayHeaderSize + (sizeLastElemMarker * (16 + (60 * (9 + msgp.Int64Size + msgp.Int64Size)) + msgp.Int64Size))
+	s = msgp.ArrayHeaderSize + (sizeLastElemMarker * (16 + (60 * (14 + msgp.Int64Size + msgp.Int64Size + msgp.Int64Size)) + msgp.Int64Size))
 	return
 }
 
@@ -460,6 +510,12 @@ func (z *lastMinuteLatency) DecodeMsg(dc *msgp.Reader) (err error) {
 							err = msgp.WrapError(err, "Totals", za0001, "Total")
 							return
 						}
+					case "Size":
+						z.Totals[za0001].Size, err = dc.ReadInt64()
+						if err != nil {
+							err = msgp.WrapError(err, "Totals", za0001, "Size")
+							return
+						}
 					case "N":
 						z.Totals[za0001].N, err = dc.ReadInt64()
 						if err != nil {
@@ -506,9 +562,9 @@ func (z *lastMinuteLatency) EncodeMsg(en *msgp.Writer) (err error) {
 		return
 	}
 	for za0001 := range z.Totals {
-		// map header, size 2
+		// map header, size 3
 		// write "Total"
-		err = en.Append(0x82, 0xa5, 0x54, 0x6f, 0x74, 0x61, 0x6c)
+		err = en.Append(0x83, 0xa5, 0x54, 0x6f, 0x74, 0x61, 0x6c)
 		if err != nil {
 			return
 		}
@@ -517,6 +573,16 @@ func (z *lastMinuteLatency) EncodeMsg(en *msgp.Writer) (err error) {
 			err = msgp.WrapError(err, "Totals", za0001, "Total")
 			return
 		}
+		// write "Size"
+		err = en.Append(0xa4, 0x53, 0x69, 0x7a, 0x65)
+		if err != nil {
+			return
+		}
+		err = en.WriteInt64(z.Totals[za0001].Size)
+		if err != nil {
+			err = msgp.WrapError(err, "Totals", za0001, "Size")
+			return
+		}
 		// write "N"
 		err = en.Append(0xa1, 0x4e)
 		if err != nil {
@@ -549,10 +615,13 @@ func (z *lastMinuteLatency) MarshalMsg(b []byte) (o []byte, err error) {
 	o = append(o, 0x82, 0xa6, 0x54, 0x6f, 0x74, 0x61, 0x6c, 0x73)
 	o = msgp.AppendArrayHeader(o, uint32(60))
 	for za0001 := range z.Totals {
-		// map header, size 2
+		// map header, size 3
 		// string "Total"
-		o = append(o, 0x82, 0xa5, 0x54, 0x6f, 0x74, 0x61, 0x6c)
+		o = append(o, 0x83, 0xa5, 0x54, 0x6f, 0x74, 0x61, 0x6c)
 		o = msgp.AppendInt64(o, z.Totals[za0001].Total)
+		// string "Size"
+		o = append(o, 0xa4, 0x53, 0x69, 0x7a, 0x65)
+		o = msgp.AppendInt64(o, z.Totals[za0001].Size)
 		// string "N"
 		o = append(o, 0xa1, 0x4e)
 		o = msgp.AppendInt64(o, z.Totals[za0001].N)
@@ -613,6 +682,12 @@ func (z *lastMinuteLatency) UnmarshalMsg(bts []byte) (o []byte, err error) {
 							err = msgp.WrapError(err, "Totals", za0001, "Total")
 							return
 						}
+					case "Size":
+						z.Totals[za0001].Size, bts, err = msgp.ReadInt64Bytes(bts)
+						if err != nil {
+							err = msgp.WrapError(err, "Totals", za0001, "Size")
+							return
+						}
 					case "N":
 						z.Totals[za0001].N, bts, err = msgp.ReadInt64Bytes(bts)
 						if err != nil {
@@ -648,6 +723,6 @@ func (z *lastMinuteLatency) UnmarshalMsg(bts []byte) (o []byte, err error) {
 
 // Msgsize returns an upper bound estimate of the number of bytes occupied by the serialized message
 func (z *lastMinuteLatency) Msgsize() (s int) {
-	s = 1 + 7 + msgp.ArrayHeaderSize + (60 * (9 + msgp.Int64Size + msgp.Int64Size)) + 8 + msgp.Int64Size
+	s = 1 + 7 + msgp.ArrayHeaderSize + (60 * (14 + msgp.Int64Size + msgp.Int64Size + msgp.Int64Size)) + 8 + msgp.Int64Size
 	return
 }

cmd/last-minute_gen_test.go

@@ -122,8 +122,8 @@ func BenchmarkDecodeAccElem(b *testing.B) {
 	}
 }
 
-func TestMarshalUnmarshalLastMinuteLatencies(t *testing.T) {
+func TestMarshalUnmarshalLastMinuteHistogram(t *testing.T) {
-	v := LastMinuteLatencies{}
+	v := LastMinuteHistogram{}
 	bts, err := v.MarshalMsg(nil)
 	if err != nil {
 		t.Fatal(err)
@@ -145,8 +145,8 @@ func TestMarshalUnmarshalLastMinuteLatencies(t *testing.T) {
 	}
 }
 
-func BenchmarkMarshalMsgLastMinuteLatencies(b *testing.B) {
+func BenchmarkMarshalMsgLastMinuteHistogram(b *testing.B) {
-	v := LastMinuteLatencies{}
+	v := LastMinuteHistogram{}
 	b.ReportAllocs()
 	b.ResetTimer()
 	for i := 0; i < b.N; i++ {
@@ -154,8 +154,8 @@ func BenchmarkMarshalMsgLastMinuteLatencies(b *testing.B) {
 	}
 }
 
-func BenchmarkAppendMsgLastMinuteLatencies(b *testing.B) {
+func BenchmarkAppendMsgLastMinuteHistogram(b *testing.B) {
-	v := LastMinuteLatencies{}
+	v := LastMinuteHistogram{}
 	bts := make([]byte, 0, v.Msgsize())
 	bts, _ = v.MarshalMsg(bts[0:0])
 	b.SetBytes(int64(len(bts)))
@@ -166,8 +166,8 @@ func BenchmarkAppendMsgLastMinuteLatencies(b *testing.B) {
 	}
 }
 
-func BenchmarkUnmarshalLastMinuteLatencies(b *testing.B) {
+func BenchmarkUnmarshalLastMinuteHistogram(b *testing.B) {
-	v := LastMinuteLatencies{}
+	v := LastMinuteHistogram{}
 	bts, _ := v.MarshalMsg(nil)
 	b.ReportAllocs()
 	b.SetBytes(int64(len(bts)))
@@ -180,17 +180,17 @@ func BenchmarkUnmarshalLastMinuteLatencies(b *testing.B) {
 	}
 }
 
-func TestEncodeDecodeLastMinuteLatencies(t *testing.T) {
+func TestEncodeDecodeLastMinuteHistogram(t *testing.T) {
-	v := LastMinuteLatencies{}
+	v := LastMinuteHistogram{}
 	var buf bytes.Buffer
 	msgp.Encode(&buf, &v)
 
 	m := v.Msgsize()
 	if buf.Len() > m {
-		t.Log("WARNING: TestEncodeDecodeLastMinuteLatencies Msgsize() is inaccurate")
+		t.Log("WARNING: TestEncodeDecodeLastMinuteHistogram Msgsize() is inaccurate")
 	}
 
-	vn := LastMinuteLatencies{}
+	vn := LastMinuteHistogram{}
 	err := msgp.Decode(&buf, &vn)
 	if err != nil {
 		t.Error(err)
@@ -204,8 +204,8 @@ func TestEncodeDecodeLastMinuteLatencies(t *testing.T) {
 	}
 }
 
-func BenchmarkEncodeLastMinuteLatencies(b *testing.B) {
+func BenchmarkEncodeLastMinuteHistogram(b *testing.B) {
-	v := LastMinuteLatencies{}
+	v := LastMinuteHistogram{}
 	var buf bytes.Buffer
 	msgp.Encode(&buf, &v)
 	b.SetBytes(int64(buf.Len()))
@@ -218,8 +218,8 @@ func BenchmarkEncodeLastMinuteLatencies(b *testing.B) {
 	en.Flush()
 }
 
-func BenchmarkDecodeLastMinuteLatencies(b *testing.B) {
+func BenchmarkDecodeLastMinuteHistogram(b *testing.B) {
-	v := LastMinuteLatencies{}
+	v := LastMinuteHistogram{}
 	var buf bytes.Buffer
 	msgp.Encode(&buf, &v)
 	b.SetBytes(int64(buf.Len()))

cmd/listen-notification-handlers.go

@@ -25,7 +25,8 @@ import (
 	"github.com/gorilla/mux"
 	"github.com/minio/minio/internal/event"
 	"github.com/minio/minio/internal/logger"
-	policy "github.com/minio/pkg/bucket/policy"
+	"github.com/minio/minio/internal/pubsub"
+	"github.com/minio/pkg/bucket/policy"
 )
 
 func (api objectAPIHandlers) ListenNotificationHandler(w http.ResponseWriter, r *http.Request) {
@@ -100,13 +101,14 @@ func (api objectAPIHandlers) ListenNotificationHandler(w http.ResponseWriter, r
 	pattern := event.NewPattern(prefix, suffix)
 
 	var eventNames []event.Name
+	var mask pubsub.Mask
 	for _, s := range values[peerRESTListenEvents] {
 		eventName, err := event.ParseName(s)
 		if err != nil {
 			writeErrorResponse(ctx, w, toAPIError(ctx, err), r.URL)
 			return
 		}
-
+		mask.MergeMaskable(eventName)
 		eventNames = append(eventNames, eventName)
 	}
 
@@ -123,11 +125,11 @@ func (api objectAPIHandlers) ListenNotificationHandler(w http.ResponseWriter, r
 
 	// Listen Publisher and peer-listen-client uses nonblocking send and hence does not wait for slow receivers.
 	// Use buffered channel to take care of burst sends or slow w.Write()
-	listenCh := make(chan interface{}, 4000)
+	listenCh := make(chan pubsub.Maskable, 4000)
 
 	peers, _ := newPeerRestClients(globalEndpoints)
 
-	listenFn := func(evI interface{}) bool {
+	err := globalHTTPListen.Subscribe(mask, listenCh, ctx.Done(), func(evI pubsub.Maskable) bool {
 		ev, ok := evI.(event.Event)
 		if !ok {
 			return false
@@ -138,14 +140,11 @@ func (api objectAPIHandlers) ListenNotificationHandler(w http.ResponseWriter, r
 			}
 		}
 		return rulesMap.MatchSimple(ev.EventName, ev.S3.Object.Key)
-	}
+	})
-
-	err := globalHTTPListen.Subscribe(listenCh, ctx.Done(), listenFn)
 	if err != nil {
 		writeErrorResponse(ctx, w, errorCodes.ToAPIErr(ErrSlowDown), r.URL)
 		return
 	}
-
 	if bucketName != "" {
 		values.Set(peerRESTListenBucket, bucketName)
 	}
@@ -173,7 +172,10 @@ func (api objectAPIHandlers) ListenNotificationHandler(w http.ResponseWriter, r
 					return
 				}
 			}
+			if len(listenCh) == 0 {
+				// Flush if nothing is queued
 				w.(http.Flusher).Flush()
+			}
 		case <-keepAliveTicker.C:
 			if _, err := w.Write([]byte(" ")); err != nil {
 				return

cmd/metrics-realtime.go

@@ -0,0 +1,109 @@
+// Copyright (c) 2015-2022 MinIO, Inc.
+//
+// This file is part of MinIO Object Storage stack
+//
+// This program is free software: you can redistribute it and/or modify
+// it under the terms of the GNU Affero General Public License as published by
+// the Free Software Foundation, either version 3 of the License, or
+// (at your option) any later version.
+//
+// This program is distributed in the hope that it will be useful
+// but WITHOUT ANY WARRANTY; without even the implied warranty of
+// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+// GNU Affero General Public License for more details.
+//
+// You should have received a copy of the GNU Affero General Public License
+// along with this program.  If not, see <http://www.gnu.org/licenses/>.
+
+package cmd
+
+import (
+	"context"
+	"time"
+
+	"github.com/minio/madmin-go"
+)
+
+func collectLocalMetrics(types madmin.MetricType, hosts map[string]struct{}) (m madmin.RealtimeMetrics) {
+	if types == madmin.MetricsNone {
+		return
+	}
+	if len(hosts) > 0 {
+		if _, ok := hosts[globalMinioAddr]; !ok {
+			return
+		}
+	}
+	if types.Contains(madmin.MetricsScanner) {
+		metrics := globalScannerMetrics.report()
+		m.Aggregated.Scanner = &metrics
+	}
+	if types.Contains(madmin.MetricsDisk) && !globalIsGateway {
+		m.Aggregated.Disk = collectDiskMetrics()
+	}
+	if types.Contains(madmin.MetricsOS) {
+		metrics := globalOSMetrics.report()
+		m.Aggregated.OS = &metrics
+	}
+	// Add types...
+
+	// ByHost is a shallow reference, so careful about sharing.
+	m.ByHost = map[string]madmin.Metrics{globalMinioAddr: m.Aggregated}
+	m.Hosts = append(m.Hosts, globalMinioAddr)
+
+	return m
+}
+
+func collectDiskMetrics() *madmin.DiskMetric {
+	objLayer := newObjectLayerFn()
+	disks := madmin.DiskMetric{
+		CollectedAt: time.Now(),
+	}
+
+	if objLayer == nil {
+		return nil
+	}
+	// only need Disks information in server mode.
+	storageInfo, errs := objLayer.LocalStorageInfo(GlobalContext)
+	for _, err := range errs {
+		if err != nil {
+			disks.Merge(&madmin.DiskMetric{NDisks: 1, Offline: 1})
+		}
+	}
+	for i, disk := range storageInfo.Disks {
+		if errs[i] != nil {
+			continue
+		}
+		var d madmin.DiskMetric
+		d.NDisks = 1
+		if disk.Healing {
+			d.Healing++
+		}
+		if disk.Metrics != nil {
+			d.LifeTimeOps = make(map[string]uint64, len(disk.Metrics.APICalls))
+			for k, v := range disk.Metrics.APICalls {
+				if v != 0 {
+					d.LifeTimeOps[k] = v
+				}
+			}
+			d.LastMinute.Operations = make(map[string]madmin.TimedAction, len(disk.Metrics.APICalls))
+			for k, v := range disk.Metrics.LastMinute {
+				if v.Count != 0 {
+					d.LastMinute.Operations[k] = v
+				}
+			}
+		}
+		disks.Merge(&d)
+	}
+	return &disks
+}
+
+func collectRemoteMetrics(ctx context.Context, types madmin.MetricType, hosts map[string]struct{}) (m madmin.RealtimeMetrics) {
+	if !globalIsDistErasure {
+		return
+	}
+	all := globalNotificationSys.GetMetrics(ctx, types, hosts)
+	for _, remote := range all {
+		m.Merge(&remote)
+	}
+	return m
+}

cmd/metrics-v2.go

@@ -1222,7 +1222,7 @@ func getScannerNodeMetrics() *MetricsGroup {
 					Help:      "Total number of unique objects scanned since server start",
 					Type:      counterMetric,
 				},
-				Value: float64(atomic.LoadUint64(&globalScannerStats.accTotalObjects)),
+				Value: float64(globalScannerMetrics.lifetime(scannerMetricScanObject)),
 			},
 			{
 				Description: MetricDescription{
@@ -1232,7 +1232,7 @@ func getScannerNodeMetrics() *MetricsGroup {
 					Help:      "Total number of object versions scanned since server start",
 					Type:      counterMetric,
 				},
-				Value: float64(atomic.LoadUint64(&globalScannerStats.accTotalVersions)),
+				Value: float64(globalScannerMetrics.lifetime(scannerMetricApplyVersion)),
 			},
 			{
 				Description: MetricDescription{
@@ -1242,7 +1242,7 @@ func getScannerNodeMetrics() *MetricsGroup {
 					Help:      "Total number of directories scanned since server start",
 					Type:      counterMetric,
 				},
-				Value: float64(atomic.LoadUint64(&globalScannerStats.accFolders)),
+				Value: float64(globalScannerMetrics.lifetime(scannerMetricScanFolder)),
 			},
 			{
 				Description: MetricDescription{
@@ -1252,7 +1252,7 @@ func getScannerNodeMetrics() *MetricsGroup {
 					Help:      "Total number of bucket scans started since server start",
 					Type:      counterMetric,
 				},
-				Value: float64(atomic.LoadUint64(&globalScannerStats.bucketsStarted)),
+				Value: float64(globalScannerMetrics.lifetime(scannerMetricScanBucketDisk) + uint64(globalScannerMetrics.activeDisks())),
 			},
 			{
 				Description: MetricDescription{
@@ -1262,7 +1262,7 @@ func getScannerNodeMetrics() *MetricsGroup {
 					Help:      "Total number of bucket scans finished since server start",
 					Type:      counterMetric,
 				},
-				Value: float64(atomic.LoadUint64(&globalScannerStats.bucketsFinished)),
+				Value: float64(globalScannerMetrics.lifetime(scannerMetricScanBucketDisk)),
 			},
 			{
 				Description: MetricDescription{
@@ -1272,12 +1272,12 @@ func getScannerNodeMetrics() *MetricsGroup {
 					Help:      "Total number of object versions checked for ilm actions since server start",
 					Type:      counterMetric,
 				},
-				Value: float64(atomic.LoadUint64(&globalScannerStats.ilmChecks)),
+				Value: float64(globalScannerMetrics.lifetime(scannerMetricILM)),
 			},
 		}
-		for i := range globalScannerStats.actions {
+		for i := range globalScannerMetrics.actions {
 			action := lifecycle.Action(i)
-			v := atomic.LoadUint64(&globalScannerStats.actions[action])
+			v := globalScannerMetrics.lifetimeActions(action)
 			if v == 0 {
 				continue
 			}
@@ -1861,11 +1861,10 @@ func getLocalDiskStorageMetrics() *MetricsGroup {
 			if disk.Metrics == nil {
 				continue
 			}
-			for apiName, latency := range disk.Metrics.APILatencies {
+			for apiName, latency := range disk.Metrics.LastMinute {
-				val := latency.(uint64)
 				metrics = append(metrics, Metric{
 					Description:    getNodeDiskAPILatencyMD(),
-					Value:          float64(val / 1000),
+					Value:          float64(latency.Avg().Microseconds()),
 					VariableLabels: map[string]string{"disk": disk.DrivePath, "api": "storage." + apiName},
 				})
 			}

cmd/notification.go

@@ -929,6 +929,38 @@ func (sys *NotificationSys) GetOSInfo(ctx context.Context) []madmin.OSInfo {
 	return reply
 }
 
+// GetMetrics - Get metrics from all peers.
+func (sys *NotificationSys) GetMetrics(ctx context.Context, t madmin.MetricType, hosts map[string]struct{}) []madmin.RealtimeMetrics {
+	reply := make([]madmin.RealtimeMetrics, len(sys.peerClients))
+
+	g := errgroup.WithNErrs(len(sys.peerClients))
+	for index, client := range sys.peerClients {
+		if client == nil {
+			continue
+		}
+		host := client.host.String()
+		if len(hosts) > 0 {
+			if _, ok := hosts[host]; !ok {
+				continue
+			}
+		}
+
+		index := index
+		g.Go(func() error {
+			var err error
+			reply[index], err = sys.peerClients[index].GetMetrics(ctx, t)
+			return err
+		}, index)
+	}
+
+	for index, err := range g.Wait() {
+		if err != nil {
+			reply[index].Errors = []string{err.Error()}
+		}
+	}
+	return reply
+}
+
 // GetSysConfig - Get information about system config
 // (only the config that are of concern to minio)
 func (sys *NotificationSys) GetSysConfig(ctx context.Context) []madmin.SysConfig {
@@ -1288,8 +1320,7 @@ func sendEvent(args eventArgs) {
 	if globalNotificationSys == nil {
 		return
 	}
-
+	if globalHTTPListen.NumSubscribers(args.EventName) > 0 {
-	if globalHTTPListen.NumSubscribers() > 0 {
 		globalHTTPListen.Publish(args.ToEvent(false))
 	}
 

cmd/os-instrumented.go

@@ -20,10 +20,12 @@ package cmd
 import (
 	"os"
 	"strings"
+	"sync/atomic"
 	"time"
 
 	"github.com/minio/madmin-go"
 	"github.com/minio/minio/internal/disk"
+	ioutilx "github.com/minio/minio/internal/ioutil"
 )
 
 //go:generate stringer -type=osMetric -trimprefix=osMetric $GOFILE
@@ -41,33 +43,67 @@ const (
 	osMetricRemove
 	osMetricStat
 	osMetricAccess
+	osMetricCreate
+	osMetricReadDirent
+	osMetricFdatasync
+	osMetricSync
 	// .... add more
 
 	osMetricLast
 )
 
+var globalOSMetrics osMetrics
+
+func init() {
+	// Inject metrics.
+	ioutilx.OsOpenFile = OpenFile
+	ioutilx.OpenFileDirectIO = OpenFileDirectIO
+	ioutilx.OsOpen = Open
+}
+
+type osMetrics struct {
+	// All fields must be accessed atomically and aligned.
+	operations [osMetricLast]uint64
+	latency    [osMetricLast]lockedLastMinuteLatency
+}
+
+// time an os action.
+func (o *osMetrics) time(s osMetric) func() {
+	startTime := time.Now()
+	return func() {
+		duration := time.Since(startTime)
+
+		atomic.AddUint64(&o.operations[s], 1)
+		o.latency[s].add(duration)
+	}
+}
+
+// incTime will increment time on metric s with a specific duration.
+func (o *osMetrics) incTime(s osMetric, d time.Duration) {
+	atomic.AddUint64(&o.operations[s], 1)
+	o.latency[s].add(d)
+}
+
 func osTrace(s osMetric, startTime time.Time, duration time.Duration, path string) madmin.TraceInfo {
 	return madmin.TraceInfo{
 		TraceType: madmin.TraceOS,
 		Time:      startTime,
 		NodeName:  globalLocalNodeName,
 		FuncName:  "os." + s.String(),
-		OSStats: madmin.TraceOSStats{
 		Duration:  duration,
 		Path:      path,
-		},
 	}
 }
 
 func updateOSMetrics(s osMetric, paths ...string) func() {
-	if globalTrace.NumSubscribers() == 0 {
+	if globalTrace.NumSubscribers(madmin.TraceOS) == 0 {
-		return func() {}
+		return globalOSMetrics.time(s)
 	}
 
 	startTime := time.Now()
 	return func() {
 		duration := time.Since(startTime)
-
+		globalOSMetrics.incTime(s, duration)
 		globalTrace.Publish(osTrace(s, startTime, duration, strings.Join(paths, " -> ")))
 	}
 }
@@ -133,3 +169,47 @@ func Stat(name string) (os.FileInfo, error) {
 	defer updateOSMetrics(osMetricStat, name)()
 	return os.Stat(name)
 }
+
+// Create captures time taken to call os.Create
+func Create(name string) (*os.File, error) {
+	defer updateOSMetrics(osMetricCreate, name)()
+	return os.Create(name)
+}
+
+// Fdatasync captures time taken to call Fdatasync
+func Fdatasync(f *os.File) error {
+	fn := ""
+	if f != nil {
+		fn = f.Name()
+	}
+	defer updateOSMetrics(osMetricFdatasync, fn)()
+	return disk.Fdatasync(f)
+}
+
+// report returns all os metrics.
+func (o *osMetrics) report() madmin.OSMetrics {
+	var m madmin.OSMetrics
+	m.CollectedAt = time.Now()
+	m.LifeTimeOps = make(map[string]uint64, osMetricLast)
+	for i := osMetric(0); i < osMetricLast; i++ {
+		if n := atomic.LoadUint64(&o.operations[i]); n > 0 {
+			m.LifeTimeOps[i.String()] = n
+		}
+	}
+	if len(m.LifeTimeOps) == 0 {
+		m.LifeTimeOps = nil
+	}
+
+	m.LastMinute.Operations = make(map[string]madmin.TimedAction, osMetricLast)
+	for i := osMetric(0); i < osMetricLast; i++ {
+		lm := o.latency[i].total()
+		if lm.N > 0 {
+			m.LastMinute.Operations[i.String()] = lm.asTimedAction()
+		}
+	}
+	if len(m.LastMinute.Operations) == 0 {
+		m.LastMinute.Operations = nil
+	}
+
+	return m
+}

cmd/os-readdir_other.go

@@ -153,5 +153,6 @@ func readDirWithOpts(dirPath string, opts readDirOpts) (entries []string, err er
 
 func globalSync() {
 	// no-op not sure about plan9/solaris support for syscall support
+	defer globalOSMetrics.time(osMetricSync)
 	syscall.Sync()
 }

cmd/os-readdir_unix.go

@@ -122,7 +122,9 @@ func readDirFn(dirPath string, fn func(name string, typ os.FileMode) error) erro
 	for {
 		if boff >= nbuf {
 			boff = 0
+			stop := globalOSMetrics.time(osMetricReadDirent)
 			nbuf, err = syscall.ReadDirent(int(f.Fd()), buf)
+			stop()
 			if err != nil {
 				if isSysErrNotDir(err) {
 					return nil
@@ -150,7 +152,7 @@ func readDirFn(dirPath string, fn func(name string, typ os.FileMode) error) erro
 		// support Dirent.Type and have DT_UNKNOWN (0) there
 		// instead.
 		if typ == unexpectedFileMode || typ&os.ModeSymlink == os.ModeSymlink {
-			fi, err := os.Stat(pathJoin(dirPath, string(name)))
+			fi, err := Stat(pathJoin(dirPath, string(name)))
 			if err != nil {
 				// It got deleted in the meantime, not found
 				// or returns too many symlinks ignore this
@@ -203,7 +205,9 @@ func readDirWithOpts(dirPath string, opts readDirOpts) (entries []string, err er
 	for count != 0 {
 		if boff >= nbuf {
 			boff = 0
+			stop := globalOSMetrics.time(osMetricReadDirent)
 			nbuf, err = syscall.ReadDirent(int(f.Fd()), buf)
+			stop()
 			if err != nil {
 				if isSysErrNotDir(err) {
 					return nil, errFileNotFound
@@ -266,5 +270,6 @@ func readDirWithOpts(dirPath string, opts readDirOpts) (entries []string, err er
 }
 
 func globalSync() {
+	defer globalOSMetrics.time(osMetricSync)
 	syscall.Sync()
 }

cmd/os-readdir_windows.go

@@ -35,7 +35,7 @@ func access(name string) error {
 // the directory itself, if the dirPath doesn't exist this function doesn't return
 // an error.
 func readDirFn(dirPath string, filter func(name string, typ os.FileMode) error) error {
-	f, err := os.Open(dirPath)
+	f, err := Open(dirPath)
 	if err != nil {
 		if osErrToFileErr(err) == errFileNotFound {
 			return nil
@@ -116,7 +116,7 @@ func readDirFn(dirPath string, filter func(name string, typ os.FileMode) error)
 
 // Return N entries at the directory dirPath.
 func readDirWithOpts(dirPath string, opts readDirOpts) (entries []string, err error) {
-	f, err := os.Open(dirPath)
+	f, err := Open(dirPath)
 	if err != nil {
 		return nil, osErrToFileErr(err)
 	}

cmd/osmetric_string.go

@@ -18,12 +18,16 @@ func _() {
 	_ = x[osMetricRemove-7]
 	_ = x[osMetricStat-8]
 	_ = x[osMetricAccess-9]
-	_ = x[osMetricLast-10]
+	_ = x[osMetricCreate-10]
+	_ = x[osMetricReadDirent-11]
+	_ = x[osMetricFdatasync-12]
+	_ = x[osMetricSync-13]
+	_ = x[osMetricLast-14]
 }
 
-const _osMetric_name = "RemoveAllMkdirAllRenameOpenFileOpenOpenFileDirectIOLstatRemoveStatAccessLast"
+const _osMetric_name = "RemoveAllMkdirAllRenameOpenFileOpenOpenFileDirectIOLstatRemoveStatAccessCreateReadDirentFdatasyncSyncLast"
 
-var _osMetric_index = [...]uint8{0, 9, 17, 23, 31, 35, 51, 56, 62, 66, 72, 76}
+var _osMetric_index = [...]uint8{0, 9, 17, 23, 31, 35, 51, 56, 62, 66, 72, 78, 88, 97, 101, 105}
 
 func (i osMetric) String() string {
 	if i >= osMetric(len(_osMetric_index)-1) {

cmd/peer-rest-client.go

@@ -34,6 +34,7 @@ import (
 	"github.com/minio/minio/internal/http"
 	xhttp "github.com/minio/minio/internal/http"
 	"github.com/minio/minio/internal/logger"
+	"github.com/minio/minio/internal/pubsub"
 	"github.com/minio/minio/internal/rest"
 	xnet "github.com/minio/pkg/net"
 	"github.com/tinylib/msgp/msgp"
@@ -193,6 +194,19 @@ func (client *peerRESTClient) GetMemInfo(ctx context.Context) (info madmin.MemIn
 	return info, err
 }
 
+// GetMetrics - fetch metrics from a remote node.
+func (client *peerRESTClient) GetMetrics(ctx context.Context, t madmin.MetricType) (info madmin.RealtimeMetrics, err error) {
+	values := make(url.Values)
+	values.Set(peerRESTTypes, strconv.FormatUint(uint64(t), 10))
+	respBody, err := client.callWithContext(ctx, peerRESTMethodMetrics, values, nil, -1)
+	if err != nil {
+		return
+	}
+	defer http.DrainBody(respBody)
+	err = gob.NewDecoder(respBody).Decode(&info)
+	return info, err
+}
+
 // GetProcInfo - fetch MinIO process information for a remote node.
 func (client *peerRESTClient) GetProcInfo(ctx context.Context) (info madmin.ProcInfo, err error) {
 	respBody, err := client.callWithContext(ctx, peerRESTMethodProcInfo, nil, nil, -1)
@@ -532,14 +546,9 @@ func (client *peerRESTClient) LoadTransitionTierConfig(ctx context.Context) erro
 	return nil
 }
 
-func (client *peerRESTClient) doTrace(traceCh chan interface{}, doneCh <-chan struct{}, traceOpts madmin.ServiceTraceOpts) {
+func (client *peerRESTClient) doTrace(traceCh chan<- pubsub.Maskable, doneCh <-chan struct{}, traceOpts madmin.ServiceTraceOpts) {
 	values := make(url.Values)
-	values.Set(peerRESTTraceErr, strconv.FormatBool(traceOpts.OnlyErrors))
+	traceOpts.AddParams(values)
-	values.Set(peerRESTTraceS3, strconv.FormatBool(traceOpts.S3))
-	values.Set(peerRESTTraceStorage, strconv.FormatBool(traceOpts.Storage))
-	values.Set(peerRESTTraceOS, strconv.FormatBool(traceOpts.OS))
-	values.Set(peerRESTTraceInternal, strconv.FormatBool(traceOpts.Internal))
-	values.Set(peerRESTTraceThreshold, traceOpts.Threshold.String())
 
 	// To cancel the REST request in case doneCh gets closed.
 	ctx, cancel := context.WithCancel(GlobalContext)
@@ -578,7 +587,7 @@ func (client *peerRESTClient) doTrace(traceCh chan interface{}, doneCh <-chan st
 	}
 }
 
-func (client *peerRESTClient) doListen(listenCh chan interface{}, doneCh <-chan struct{}, v url.Values) {
+func (client *peerRESTClient) doListen(listenCh chan<- pubsub.Maskable, doneCh <-chan struct{}, v url.Values) {
 	// To cancel the REST request in case doneCh gets closed.
 	ctx, cancel := context.WithCancel(GlobalContext)
 
@@ -617,7 +626,7 @@ func (client *peerRESTClient) doListen(listenCh chan interface{}, doneCh <-chan
 }
 
 // Listen - listen on peers.
-func (client *peerRESTClient) Listen(listenCh chan interface{}, doneCh <-chan struct{}, v url.Values) {
+func (client *peerRESTClient) Listen(listenCh chan<- pubsub.Maskable, doneCh <-chan struct{}, v url.Values) {
 	go func() {
 		for {
 			client.doListen(listenCh, doneCh, v)
@@ -633,7 +642,7 @@ func (client *peerRESTClient) Listen(listenCh chan interface{}, doneCh <-chan st
 }
 
 // Trace - send http trace request to peer nodes
-func (client *peerRESTClient) Trace(traceCh chan interface{}, doneCh <-chan struct{}, traceOpts madmin.ServiceTraceOpts) {
+func (client *peerRESTClient) Trace(traceCh chan<- pubsub.Maskable, doneCh <-chan struct{}, traceOpts madmin.ServiceTraceOpts) {
 	go func() {
 		for {
 			client.doTrace(traceCh, doneCh, traceOpts)
@@ -648,7 +657,7 @@ func (client *peerRESTClient) Trace(traceCh chan interface{}, doneCh <-chan stru
 	}()
 }
 
-func (client *peerRESTClient) doConsoleLog(logCh chan interface{}, doneCh <-chan struct{}) {
+func (client *peerRESTClient) doConsoleLog(logCh chan pubsub.Maskable, doneCh <-chan struct{}) {
 	// To cancel the REST request in case doneCh gets closed.
 	ctx, cancel := context.WithCancel(GlobalContext)
 
@@ -686,7 +695,7 @@ func (client *peerRESTClient) doConsoleLog(logCh chan interface{}, doneCh <-chan
 }
 
 // ConsoleLog - sends request to peer nodes to get console logs
-func (client *peerRESTClient) ConsoleLog(logCh chan interface{}, doneCh <-chan struct{}) {
+func (client *peerRESTClient) ConsoleLog(logCh chan pubsub.Maskable, doneCh <-chan struct{}) {
 	go func() {
 		for {
 			client.doConsoleLog(logCh, doneCh)

cmd/peer-rest-common.go

@@ -18,7 +18,7 @@
 package cmd
 
 const (
-	peerRESTVersion       = "v22" // Add bulk GetBucketStats
+	peerRESTVersion       = "v23" // Added /metrics
 	peerRESTVersionPrefix = SlashSeparator + peerRESTVersion
 	peerRESTPrefix        = minioReservedBucketPath + "/peer"
 	peerRESTPath          = peerRESTPrefix + peerRESTVersionPrefix
@@ -70,6 +70,7 @@ const (
 	peerRESTMethodGetLastDayTierStats         = "/getlastdaytierstats"
 	peerRESTMethodDevNull                     = "/devnull"
 	peerRESTMethodNetperf                     = "/netperf"
+	peerRESTMethodMetrics                     = "/metrics"
 )
 
 const (
@@ -84,16 +85,11 @@ const (
 	peerRESTSignal       = "signal"
 	peerRESTSubSys       = "sub-sys"
 	peerRESTProfiler     = "profiler"
-	peerRESTTraceErr       = "err"
-	peerRESTTraceInternal  = "internal"
-	peerRESTTraceStorage   = "storage"
-	peerRESTTraceS3        = "s3"
-	peerRESTTraceOS        = "os"
-	peerRESTTraceThreshold = "threshold"
 	peerRESTSize         = "size"
 	peerRESTConcurrent   = "concurrent"
 	peerRESTDuration     = "duration"
 	peerRESTStorageClass = "storage-class"
+	peerRESTTypes        = "types"
 
 	peerRESTListenBucket = "bucket"
 	peerRESTListenPrefix = "prefix"

cmd/peer-rest-server.go

@@ -35,6 +35,7 @@ import (
 	b "github.com/minio/minio/internal/bucket/bandwidth"
 	"github.com/minio/minio/internal/event"
 	"github.com/minio/minio/internal/logger"
+	"github.com/minio/minio/internal/pubsub"
 	"github.com/tinylib/msgp/msgp"
 )
 
@@ -411,6 +412,28 @@ func (s *peerRESTServer) GetMemInfoHandler(w http.ResponseWriter, r *http.Reques
 	logger.LogIf(ctx, gob.NewEncoder(w).Encode(info))
 }
 
+// GetMetricsHandler - returns server metrics.
+func (s *peerRESTServer) GetMetricsHandler(w http.ResponseWriter, r *http.Request) {
+	if !s.IsValid(w, r) {
+		s.writeErrorResponse(w, errors.New("Invalid request"))
+		return
+	}
+
+	var types madmin.MetricType
+	if t, _ := strconv.ParseUint(r.Form.Get(peerRESTTypes), 10, 64); t != 0 {
+		types = madmin.MetricType(t)
+	} else {
+		types = madmin.MetricsAll
+	}
+
+	ctx, cancel := context.WithCancel(r.Context())
+	defer cancel()
+
+	info := collectLocalMetrics(types, nil)
+
+	logger.LogIf(ctx, gob.NewEncoder(w).Encode(info))
+}
+
 // GetSysConfigHandler - returns system config information.
 // (only the config that are of concern to minio)
 func (s *peerRESTServer) GetSysConfigHandler(w http.ResponseWriter, r *http.Request) {
@@ -853,26 +876,26 @@ func (s *peerRESTServer) ListenHandler(w http.ResponseWriter, r *http.Request) {
 	pattern := event.NewPattern(prefix, suffix)
 
 	var eventNames []event.Name
+	var mask pubsub.Mask
 	for _, ev := range values[peerRESTListenEvents] {
 		eventName, err := event.ParseName(ev)
 		if err != nil {
 			s.writeErrorResponse(w, err)
 			return
 		}
-
+		mask.MergeMaskable(eventName)
 		eventNames = append(eventNames, eventName)
 	}
 
 	rulesMap := event.NewRulesMap(eventNames, pattern, event.TargetID{ID: mustGetUUID()})
 
-	doneCh := make(chan struct{})
+	doneCh := r.Context().Done()
-	defer close(doneCh)
 
 	// Listen Publisher uses nonblocking publish and hence does not wait for slow subscribers.
 	// Use buffered channel to take care of burst sends or slow w.Write()
-	ch := make(chan interface{}, 2000)
+	ch := make(chan pubsub.Maskable, 2000)
 
-	listenFn := func(evI interface{}) bool {
+	err := globalHTTPListen.Subscribe(mask, ch, doneCh, func(evI pubsub.Maskable) bool {
 		ev, ok := evI.(event.Event)
 		if !ok {
 			return false
@@ -883,14 +906,11 @@ func (s *peerRESTServer) ListenHandler(w http.ResponseWriter, r *http.Request) {
 			}
 		}
 		return rulesMap.MatchSimple(ev.EventName, ev.S3.Object.Key)
-	}
+	})
-
-	err := globalHTTPListen.Subscribe(ch, doneCh, listenFn)
 	if err != nil {
 		s.writeErrorResponse(w, err)
 		return
 	}
-
 	keepAliveTicker := time.NewTicker(500 * time.Millisecond)
 	defer keepAliveTicker.Stop()
 
@@ -901,7 +921,10 @@ func (s *peerRESTServer) ListenHandler(w http.ResponseWriter, r *http.Request) {
 			if err := enc.Encode(ev); err != nil {
 				return
 			}
+			if len(ch) == 0 {
+				// Flush if nothing is queued
 				w.(http.Flusher).Flush()
+			}
 		case <-r.Context().Done():
 			return
 		case <-keepAliveTicker.C:
@@ -913,23 +936,6 @@ func (s *peerRESTServer) ListenHandler(w http.ResponseWriter, r *http.Request) {
 	}
 }
 
-func extractTraceOptsFromPeerRequest(r *http.Request) (opts madmin.ServiceTraceOpts, err error) {
-	opts.S3 = r.Form.Get(peerRESTTraceS3) == "true"
-	opts.OS = r.Form.Get(peerRESTTraceOS) == "true"
-	opts.Storage = r.Form.Get(peerRESTTraceStorage) == "true"
-	opts.Internal = r.Form.Get(peerRESTTraceInternal) == "true"
-	opts.OnlyErrors = r.Form.Get(peerRESTTraceErr) == "true"
-
-	if t := r.Form.Get(peerRESTTraceThreshold); t != "" {
-		d, err := time.ParseDuration(t)
-		if err != nil {
-			return opts, err
-		}
-		opts.Threshold = d
-	}
-	return
-}
-
 // TraceHandler sends http trace messages back to peer rest client
 func (s *peerRESTServer) TraceHandler(w http.ResponseWriter, r *http.Request) {
 	if !s.IsValid(w, r) {
@@ -937,29 +943,27 @@ func (s *peerRESTServer) TraceHandler(w http.ResponseWriter, r *http.Request) {
 		return
 	}
 
-	traceOpts, err := extractTraceOptsFromPeerRequest(r)
+	var traceOpts madmin.ServiceTraceOpts
+	err := traceOpts.ParseParams(r)
 	if err != nil {
 		s.writeErrorResponse(w, errors.New("Invalid request"))
 		return
 	}
 
-	doneCh := make(chan struct{})
-	defer close(doneCh)
-
 	// Trace Publisher uses nonblocking publish and hence does not wait for slow subscribers.
 	// Use buffered channel to take care of burst sends or slow w.Write()
-	ch := make(chan interface{}, 2000)
+	ch := make(chan pubsub.Maskable, 2000)
-
+	mask := pubsub.MaskFromMaskable(traceOpts.TraceTypes())
-	traceFn := func(entry interface{}) bool {
+	err = globalTrace.Subscribe(mask, ch, r.Context().Done(), func(entry pubsub.Maskable) bool {
-		return mustTrace(entry, traceOpts)
+		if e, ok := entry.(madmin.TraceInfo); ok {
+			return shouldTrace(e, traceOpts)
 		}
-
+		return false
-	err = globalTrace.Subscribe(ch, doneCh, traceFn)
+	})
 	if err != nil {
 		s.writeErrorResponse(w, err)
 		return
 	}
-
 	keepAliveTicker := time.NewTicker(500 * time.Millisecond)
 	defer keepAliveTicker.Stop()
 
@@ -970,7 +974,6 @@ func (s *peerRESTServer) TraceHandler(w http.ResponseWriter, r *http.Request) {
 			if err := enc.Encode(entry); err != nil {
 				return
 			}
-			w.(http.Flusher).Flush()
 		case <-r.Context().Done():
 			return
 		case <-keepAliveTicker.C:
@@ -1042,13 +1045,12 @@ func (s *peerRESTServer) ConsoleLogHandler(w http.ResponseWriter, r *http.Reques
 	doneCh := make(chan struct{})
 	defer close(doneCh)
 
-	ch := make(chan interface{}, 2000)
+	ch := make(chan pubsub.Maskable, 2000)
-	err := globalConsoleSys.Subscribe(ch, doneCh, "", 0, string(logger.All), nil)
+	err := globalConsoleSys.Subscribe(ch, doneCh, "", 0, madmin.LogMaskAll, nil)
 	if err != nil {
 		s.writeErrorResponse(w, err)
 		return
 	}
-
 	keepAliveTicker := time.NewTicker(500 * time.Millisecond)
 	defer keepAliveTicker.Stop()
 
@@ -1275,6 +1277,7 @@ func registerPeerRESTHandlers(router *mux.Router) {
 	subrouter.Methods(http.MethodPost).Path(peerRESTVersionPrefix + peerRESTMethodServerInfo).HandlerFunc(httpTraceHdrs(server.ServerInfoHandler))
 	subrouter.Methods(http.MethodPost).Path(peerRESTVersionPrefix + peerRESTMethodProcInfo).HandlerFunc(httpTraceHdrs(server.GetProcInfoHandler))
 	subrouter.Methods(http.MethodPost).Path(peerRESTVersionPrefix + peerRESTMethodMemInfo).HandlerFunc(httpTraceHdrs(server.GetMemInfoHandler))
+	subrouter.Methods(http.MethodPost).Path(peerRESTVersionPrefix + peerRESTMethodMetrics).HandlerFunc(httpTraceHdrs(server.GetMetricsHandler)).Queries(restQueries(peerRESTTypes)...)
 	subrouter.Methods(http.MethodPost).Path(peerRESTVersionPrefix + peerRESTMethodSysErrors).HandlerFunc(httpTraceHdrs(server.GetSysErrorsHandler))
 	subrouter.Methods(http.MethodPost).Path(peerRESTVersionPrefix + peerRESTMethodSysServices).HandlerFunc(httpTraceHdrs(server.GetSysServicesHandler))
 	subrouter.Methods(http.MethodPost).Path(peerRESTVersionPrefix + peerRESTMethodSysConfig).HandlerFunc(httpTraceHdrs(server.GetSysConfigHandler))

cmd/scannermetric_string.go

@@ -0,0 +1,39 @@
+// Code generated by "stringer -type=scannerMetric -trimprefix=scannerMetric data-scanner-metric.go"; DO NOT EDIT.
+
+package cmd
+
+import "strconv"
+
+func _() {
+	// An "invalid array index" compiler error signifies that the constant values have changed.
+	// Re-run the stringer command to generate them again.
+	var x [1]struct{}
+	_ = x[scannerMetricReadMetadata-0]
+	_ = x[scannerMetricCheckMissing-1]
+	_ = x[scannerMetricSaveUsage-2]
+	_ = x[scannerMetricApplyAll-3]
+	_ = x[scannerMetricApplyVersion-4]
+	_ = x[scannerMetricTierObjSweep-5]
+	_ = x[scannerMetricHealCheck-6]
+	_ = x[scannerMetricILM-7]
+	_ = x[scannerMetricCheckReplication-8]
+	_ = x[scannerMetricYield-9]
+	_ = x[scannerMetricStartTrace-10]
+	_ = x[scannerMetricScanObject-11]
+	_ = x[scannerMetricLastRealtime-12]
+	_ = x[scannerMetricScanFolder-13]
+	_ = x[scannerMetricScanCycle-14]
+	_ = x[scannerMetricScanBucketDisk-15]
+	_ = x[scannerMetricLast-16]
+}
+
+const _scannerMetric_name = "ReadMetadataCheckMissingSaveUsageApplyAllApplyVersionTierObjSweepHealCheckILMCheckReplicationYieldStartTraceScanObjectLastRealtimeScanFolderScanCycleScanBucketDiskLast"
+
+var _scannerMetric_index = [...]uint8{0, 12, 24, 33, 41, 53, 65, 74, 77, 93, 98, 108, 118, 130, 140, 149, 163, 167}
+
+func (i scannerMetric) String() string {
+	if i >= scannerMetric(len(_scannerMetric_index)-1) {
+		return "scannerMetric(" + strconv.FormatInt(int64(i), 10) + ")"
+	}
+	return _scannerMetric_name[_scannerMetric_index[i]:_scannerMetric_index[i+1]]
+}

cmd/server-startup-msg.go

@@ -47,7 +47,7 @@ func printStartupMessage(apiEndpoints []string, err error) {
 	logger.Info(color.Bold("MinIO Object Storage Server"))
 	if err != nil {
 		if globalConsoleSys != nil {
-			globalConsoleSys.Send(fmt.Sprintf("Server startup failed with '%v', some features may be missing", err), string(logger.All))
+			globalConsoleSys.Send(fmt.Sprintf("Server startup failed with '%v', some features may be missing", err))
 		}
 	}
 

cmd/storage-datatypes.go

@@ -47,7 +47,7 @@ type DiskInfo struct {
 // the number of calls of each API and the moving average of
 // the duration of each API.
 type DiskMetrics struct {
-	APILatencies map[string]uint64 `json:"apiLatencies,omitempty"`
+	LastMinute map[string]AccElem `json:"apiLatencies,omitempty"`
 	APICalls   map[string]uint64  `json:"apiCalls,omitempty"`
 }
 

cmd/storage-datatypes_gen.go

@@ -291,35 +291,35 @@ func (z *DiskMetrics) DecodeMsg(dc *msgp.Reader) (err error) {
 			return
 		}
 		switch msgp.UnsafeString(field) {
-		case "APILatencies":
+		case "LastMinute":
 			var zb0002 uint32
 			zb0002, err = dc.ReadMapHeader()
 			if err != nil {
-				err = msgp.WrapError(err, "APILatencies")
+				err = msgp.WrapError(err, "LastMinute")
 				return
 			}
-			if z.APILatencies == nil {
+			if z.LastMinute == nil {
-				z.APILatencies = make(map[string]uint64, zb0002)
+				z.LastMinute = make(map[string]AccElem, zb0002)
-			} else if len(z.APILatencies) > 0 {
+			} else if len(z.LastMinute) > 0 {
-				for key := range z.APILatencies {
+				for key := range z.LastMinute {
-					delete(z.APILatencies, key)
+					delete(z.LastMinute, key)
 				}
 			}
 			for zb0002 > 0 {
 				zb0002--
 				var za0001 string
-				var za0002 uint64
+				var za0002 AccElem
 				za0001, err = dc.ReadString()
 				if err != nil {
-					err = msgp.WrapError(err, "APILatencies")
+					err = msgp.WrapError(err, "LastMinute")
 					return
 				}
-				za0002, err = dc.ReadUint64()
+				err = za0002.DecodeMsg(dc)
 				if err != nil {
-					err = msgp.WrapError(err, "APILatencies", za0001)
+					err = msgp.WrapError(err, "LastMinute", za0001)
 					return
 				}
-				z.APILatencies[za0001] = za0002
+				z.LastMinute[za0001] = za0002
 			}
 		case "APICalls":
 			var zb0003 uint32
@@ -365,25 +365,25 @@ func (z *DiskMetrics) DecodeMsg(dc *msgp.Reader) (err error) {
 // EncodeMsg implements msgp.Encodable
 func (z *DiskMetrics) EncodeMsg(en *msgp.Writer) (err error) {
 	// map header, size 2
-	// write "APILatencies"
+	// write "LastMinute"
-	err = en.Append(0x82, 0xac, 0x41, 0x50, 0x49, 0x4c, 0x61, 0x74, 0x65, 0x6e, 0x63, 0x69, 0x65, 0x73)
+	err = en.Append(0x82, 0xaa, 0x4c, 0x61, 0x73, 0x74, 0x4d, 0x69, 0x6e, 0x75, 0x74, 0x65)
 	if err != nil {
 		return
 	}
-	err = en.WriteMapHeader(uint32(len(z.APILatencies)))
+	err = en.WriteMapHeader(uint32(len(z.LastMinute)))
 	if err != nil {
-		err = msgp.WrapError(err, "APILatencies")
+		err = msgp.WrapError(err, "LastMinute")
 		return
 	}
-	for za0001, za0002 := range z.APILatencies {
+	for za0001, za0002 := range z.LastMinute {
 		err = en.WriteString(za0001)
 		if err != nil {
-			err = msgp.WrapError(err, "APILatencies")
+			err = msgp.WrapError(err, "LastMinute")
 			return
 		}
-		err = en.WriteUint64(za0002)
+		err = za0002.EncodeMsg(en)
 		if err != nil {
-			err = msgp.WrapError(err, "APILatencies", za0001)
+			err = msgp.WrapError(err, "LastMinute", za0001)
 			return
 		}
 	}
@@ -416,12 +416,16 @@ func (z *DiskMetrics) EncodeMsg(en *msgp.Writer) (err error) {
 func (z *DiskMetrics) MarshalMsg(b []byte) (o []byte, err error) {
 	o = msgp.Require(b, z.Msgsize())
 	// map header, size 2
-	// string "APILatencies"
+	// string "LastMinute"
-	o = append(o, 0x82, 0xac, 0x41, 0x50, 0x49, 0x4c, 0x61, 0x74, 0x65, 0x6e, 0x63, 0x69, 0x65, 0x73)
+	o = append(o, 0x82, 0xaa, 0x4c, 0x61, 0x73, 0x74, 0x4d, 0x69, 0x6e, 0x75, 0x74, 0x65)
-	o = msgp.AppendMapHeader(o, uint32(len(z.APILatencies)))
+	o = msgp.AppendMapHeader(o, uint32(len(z.LastMinute)))
-	for za0001, za0002 := range z.APILatencies {
+	for za0001, za0002 := range z.LastMinute {
 		o = msgp.AppendString(o, za0001)
-		o = msgp.AppendUint64(o, za0002)
+		o, err = za0002.MarshalMsg(o)
+		if err != nil {
+			err = msgp.WrapError(err, "LastMinute", za0001)
+			return
+		}
 	}
 	// string "APICalls"
 	o = append(o, 0xa8, 0x41, 0x50, 0x49, 0x43, 0x61, 0x6c, 0x6c, 0x73)
@@ -451,35 +455,35 @@ func (z *DiskMetrics) UnmarshalMsg(bts []byte) (o []byte, err error) {
 			return
 		}
 		switch msgp.UnsafeString(field) {
-		case "APILatencies":
+		case "LastMinute":
 			var zb0002 uint32
 			zb0002, bts, err = msgp.ReadMapHeaderBytes(bts)
 			if err != nil {
-				err = msgp.WrapError(err, "APILatencies")
+				err = msgp.WrapError(err, "LastMinute")
 				return
 			}
-			if z.APILatencies == nil {
+			if z.LastMinute == nil {
-				z.APILatencies = make(map[string]uint64, zb0002)
+				z.LastMinute = make(map[string]AccElem, zb0002)
-			} else if len(z.APILatencies) > 0 {
+			} else if len(z.LastMinute) > 0 {
-				for key := range z.APILatencies {
+				for key := range z.LastMinute {
-					delete(z.APILatencies, key)
+					delete(z.LastMinute, key)
 				}
 			}
 			for zb0002 > 0 {
 				var za0001 string
-				var za0002 uint64
+				var za0002 AccElem
 				zb0002--
 				za0001, bts, err = msgp.ReadStringBytes(bts)
 				if err != nil {
-					err = msgp.WrapError(err, "APILatencies")
+					err = msgp.WrapError(err, "LastMinute")
 					return
 				}
-				za0002, bts, err = msgp.ReadUint64Bytes(bts)
+				bts, err = za0002.UnmarshalMsg(bts)
 				if err != nil {
-					err = msgp.WrapError(err, "APILatencies", za0001)
+					err = msgp.WrapError(err, "LastMinute", za0001)
 					return
 				}
-				z.APILatencies[za0001] = za0002
+				z.LastMinute[za0001] = za0002
 			}
 		case "APICalls":
 			var zb0003 uint32
@@ -525,11 +529,11 @@ func (z *DiskMetrics) UnmarshalMsg(bts []byte) (o []byte, err error) {
 
 // Msgsize returns an upper bound estimate of the number of bytes occupied by the serialized message
 func (z *DiskMetrics) Msgsize() (s int) {
-	s = 1 + 13 + msgp.MapHeaderSize
+	s = 1 + 11 + msgp.MapHeaderSize
-	if z.APILatencies != nil {
+	if z.LastMinute != nil {
-		for za0001, za0002 := range z.APILatencies {
+		for za0001, za0002 := range z.LastMinute {
 			_ = za0002
-			s += msgp.StringPrefixSize + len(za0001) + msgp.Uint64Size
+			s += msgp.StringPrefixSize + len(za0001) + za0002.Msgsize()
 		}
 	}
 	s += 9 + msgp.MapHeaderSize

cmd/update.go

@@ -95,7 +95,7 @@ func getModTime(path string) (t time.Time, err error) {
 
 	// Version is minio non-standard, we will use minio binary's
 	// ModTime as release time.
-	fi, err := os.Stat(absPath)
+	fi, err := Stat(absPath)
 	if err != nil {
 		return t, fmt.Errorf("Unable to get ModTime of %s. %w", absPath, err)
 	}

cmd/utils.go

@@ -54,6 +54,7 @@ import (
 	"github.com/minio/minio/internal/fips"
 	"github.com/minio/minio/internal/handlers"
 	xhttp "github.com/minio/minio/internal/http"
+	ioutilx "github.com/minio/minio/internal/ioutil"
 	"github.com/minio/minio/internal/logger"
 	"github.com/minio/minio/internal/logger/message/audit"
 	"github.com/minio/minio/internal/rest"
@@ -305,7 +306,7 @@ func startProfiler(profilerType string) (minioProfiler, error) {
 			return nil, err
 		}
 		fn := filepath.Join(dirPath, "cpu.out")
-		f, err := os.Create(fn)
+		f, err := Create(fn)
 		if err != nil {
 			return nil, err
 		}
@@ -319,8 +320,8 @@ func startProfiler(profilerType string) (minioProfiler, error) {
 			if err != nil {
 				return nil, err
 			}
-			defer os.RemoveAll(dirPath)
+			defer RemoveAll(dirPath)
-			return ioutil.ReadFile(fn)
+			return ioutilx.ReadFile(fn)
 		}
 	case madmin.ProfilerCPUIO:
 		// at 10k or more goroutines fgprof is likely to become
@@ -335,7 +336,7 @@ func startProfiler(profilerType string) (minioProfiler, error) {
 			return nil, err
 		}
 		fn := filepath.Join(dirPath, "cpuio.out")
-		f, err := os.Create(fn)
+		f, err := Create(fn)
 		if err != nil {
 			return nil, err
 		}
@@ -349,8 +350,8 @@ func startProfiler(profilerType string) (minioProfiler, error) {
 			if err != nil {
 				return nil, err
 			}
-			defer os.RemoveAll(dirPath)
+			defer RemoveAll(dirPath)
-			return ioutil.ReadFile(fn)
+			return ioutilx.ReadFile(fn)
 		}
 	case madmin.ProfilerMEM:
 		runtime.GC()
@@ -400,7 +401,7 @@ func startProfiler(profilerType string) (minioProfiler, error) {
 			return nil, err
 		}
 		fn := filepath.Join(dirPath, "trace.out")
-		f, err := os.Create(fn)
+		f, err := Create(fn)
 		if err != nil {
 			return nil, err
 		}
@@ -415,8 +416,8 @@ func startProfiler(profilerType string) (minioProfiler, error) {
 			if err != nil {
 				return nil, err
 			}
-			defer os.RemoveAll(dirPath)
+			defer RemoveAll(dirPath)
-			return ioutil.ReadFile(fn)
+			return ioutilx.ReadFile(fn)
 		}
 	default:
 		return nil, errors.New("profiler type unknown")

cmd/xl-storage-disk-id-check.go

@@ -86,11 +86,11 @@ func (p *xlStorageDiskIDCheck) getMetrics() DiskMetrics {
 		p.metricsCache.TTL = 100 * time.Millisecond
 		p.metricsCache.Update = func() (interface{}, error) {
 			diskMetric := DiskMetrics{
-				APILatencies: make(map[string]uint64, len(p.apiLatencies)),
+				LastMinute: make(map[string]AccElem, len(p.apiLatencies)),
 				APICalls:   make(map[string]uint64, len(p.apiCalls)),
 			}
 			for i, v := range p.apiLatencies {
-				diskMetric.APILatencies[storageMetric(i).String()] = v.value()
+				diskMetric.LastMinute[storageMetric(i).String()] = v.total()
 			}
 			for i := range p.apiCalls {
 				diskMetric.APICalls[storageMetric(i).String()] = atomic.LoadUint64(&p.apiCalls[i])
@@ -113,10 +113,18 @@ func (e *lockedLastMinuteLatency) add(value time.Duration) {
 	e.lastMinuteLatency.add(value)
 }
 
-func (e *lockedLastMinuteLatency) value() uint64 {
+// addSize will add a duration and size.
+func (e *lockedLastMinuteLatency) addSize(value time.Duration, sz int64) {
 	e.Lock()
 	defer e.Unlock()
-	return e.lastMinuteLatency.getAvgData().avg()
+	e.lastMinuteLatency.addSize(value, sz)
+}
+
+// total returns the total call count and latency for the last minute.
+func (e *lockedLastMinuteLatency) total() AccElem {
+	e.Lock()
+	defer e.Unlock()
+	return e.lastMinuteLatency.getTotal()
 }
 
 func newXLStorageDiskIDCheck(storage *xlStorage) *xlStorageDiskIDCheck {
@@ -508,17 +516,26 @@ func storageTrace(s storageMetric, startTime time.Time, duration time.Duration,
 		Time:      startTime,
 		NodeName:  globalLocalNodeName,
 		FuncName:  "storage." + s.String(),
-		StorageStats: madmin.TraceStorageStats{
 		Duration:  duration,
 		Path:      path,
-		},
+	}
+}
+
+func scannerTrace(s scannerMetric, startTime time.Time, duration time.Duration, path string) madmin.TraceInfo {
+	return madmin.TraceInfo{
+		TraceType: madmin.TraceScanner,
+		Time:      startTime,
+		NodeName:  globalLocalNodeName,
+		FuncName:  "scanner." + s.String(),
+		Duration:  duration,
+		Path:      path,
 	}
 }
 
 // Update storage metrics
 func (p *xlStorageDiskIDCheck) updateStorageMetrics(s storageMetric, paths ...string) func(err *error) {
 	startTime := time.Now()
-	trace := globalTrace.NumSubscribers() > 0
+	trace := globalTrace.NumSubscribers(madmin.TraceStorage) > 0
 	return func(err *error) {
 		duration := time.Since(startTime)
 
@@ -655,18 +672,9 @@ func (p *xlStorageDiskIDCheck) TrackDiskHealth(ctx context.Context, s storageMet
 	atomic.StoreInt64(&p.health.lastStarted, time.Now().UnixNano())
 	ctx = context.WithValue(ctx, healthDiskCtxKey{}, &healthDiskCtxValue{lastSuccess: &p.health.lastSuccess})
 	si := p.updateStorageMetrics(s, paths...)
-	t := time.Now()
 	var once sync.Once
 	return ctx, func(errp *error) {
 		once.Do(func() {
-			if false {
-				var ers string
-				if errp != nil {
-					err := *errp
-					ers = fmt.Sprint(err)
-				}
-				fmt.Println(time.Now().Format(time.RFC3339), "op", s, "took", time.Since(t), "result:", ers, "disk:", p.storage.String(), "path:", strings.Join(paths, "/"))
-			}
 			p.health.tokens <- struct{}{}
 			if errp != nil {
 				err := *errp

cmd/xl-storage.go

@@ -32,7 +32,6 @@ import (
 	"runtime"
 	"strings"
 	"sync"
-	"sync/atomic"
 	"syscall"
 	"time"
 
@@ -477,8 +476,12 @@ func (s *xlStorage) NSScanner(ctx context.Context, cache dataUsageCache, updates
 			// if no xl.meta/xl.json found, skip the file.
 			return sizeSummary{}, errSkipFile
 		}
+		stopFn := globalScannerMetrics.log(scannerMetricScanObject, s.diskPath, PathJoin(item.bucket, item.objectPath()))
+		defer stopFn()
 
+		doneSz := globalScannerMetrics.timeSize(scannerMetricReadMetadata)
 		buf, err := s.readMetadata(ctx, item.Path)
+		doneSz(len(buf))
 		if err != nil {
 			if intDataUpdateTracker.debug {
 				console.Debugf(color.Green("scannerBucket:")+" object path missing: %v: %w\n", item.Path, err)
@@ -502,8 +505,11 @@ func (s *xlStorage) NSScanner(ctx context.Context, cache dataUsageCache, updates
 		if noTiers = globalTierConfigMgr.Empty(); !noTiers {
 			sizeS.tiers = make(map[string]tierStats)
 		}
-		atomic.AddUint64(&globalScannerStats.accTotalObjects, 1)
+
+		done := globalScannerMetrics.time(scannerMetricApplyAll)
 		fivs.Versions, err = item.applyVersionActions(ctx, objAPI, fivs.Versions)
+		done()
+
 		if err != nil {
 			if intDataUpdateTracker.debug {
 				console.Debugf(color.Green("scannerBucket:")+" applying version actions failed: %v: %w\n", item.Path, err)
@@ -514,9 +520,10 @@ func (s *xlStorage) NSScanner(ctx context.Context, cache dataUsageCache, updates
 		versioned := vcfg != nil && vcfg.Versioned(item.objectPath())
 
 		for _, version := range fivs.Versions {
-			atomic.AddUint64(&globalScannerStats.accTotalVersions, 1)
 			oi := version.ToObjectInfo(item.bucket, item.objectPath(), versioned)
+			done = globalScannerMetrics.time(scannerMetricApplyVersion)
 			sz := item.applyActions(ctx, objAPI, oi, &sizeS)
+			done()
 			if oi.VersionID != "" && sz == oi.Size {
 				sizeS.versions++
 			}
@@ -528,7 +535,6 @@ func (s *xlStorage) NSScanner(ctx context.Context, cache dataUsageCache, updates
 			//    tracking deleted transitioned objects
 			switch {
 			case noTiers, oi.DeleteMarker, oi.TransitionedObject.FreeVersion:
-
 				continue
 			}
 			tier := minioHotTier
@@ -541,7 +547,9 @@ func (s *xlStorage) NSScanner(ctx context.Context, cache dataUsageCache, updates
 		// apply tier sweep action on free versions
 		for _, freeVersion := range fivs.FreeVersions {
 			oi := freeVersion.ToObjectInfo(item.bucket, item.objectPath(), versioned)
+			done = globalScannerMetrics.time(scannerMetricTierObjSweep)
 			item.applyTierObjSweep(ctx, objAPI, oi)
+			done()
 		}
 		return sizeS, nil
 	}, scanMode)
@@ -1847,7 +1855,7 @@ func (s *xlStorage) CreateFile(ctx context.Context, volume, path string, fileSiz
 	}
 
 	defer func() {
-		disk.Fdatasync(w) // Only interested in flushing the size_t not mtime/atime
+		Fdatasync(w) // Only interested in flushing the size_t not mtime/atime
 		w.Close()
 	}()
 

go.mod

@@ -43,12 +43,12 @@ require (
 	github.com/lib/pq v1.10.4
 	github.com/miekg/dns v1.1.48
 	github.com/minio/cli v1.22.0
-	github.com/minio/console v0.19.0
+	github.com/minio/console v0.19.1-0.20220705203612-63e2793272dd
 	github.com/minio/csvparser v1.0.0
 	github.com/minio/dperf v0.4.2
 	github.com/minio/highwayhash v1.0.2
 	github.com/minio/kes v0.19.2
-	github.com/minio/madmin-go v1.3.20
+	github.com/minio/madmin-go v1.4.3
 	github.com/minio/minio-go/v7 v7.0.30
 	github.com/minio/pkg v1.1.26
 	github.com/minio/selfupdate v0.4.0
@@ -67,6 +67,7 @@ require (
 	github.com/pkg/errors v0.9.1
 	github.com/prometheus/client_golang v1.12.1
 	github.com/prometheus/client_model v0.2.0
+	github.com/prometheus/common v0.34.0
 	github.com/prometheus/procfs v0.7.3
 	github.com/rs/cors v1.7.0
 	github.com/rs/dnscache v0.0.0-20211102005908-e0241e321417
@@ -82,7 +83,7 @@ require (
 	go.etcd.io/etcd/client/v3 v3.5.4
 	go.uber.org/atomic v1.9.0
 	go.uber.org/zap v1.21.0
-	golang.org/x/crypto v0.0.0-20220411220226-7b82a4e95df4
+	golang.org/x/crypto v0.0.0-20220525230936-793ad666bf5e
 	golang.org/x/oauth2 v0.0.0-20220411215720-9780585627b5
 	golang.org/x/sys v0.0.0-20220502124256-b6088ccd6cba
 	golang.org/x/time v0.0.0-20220224211638-0e9765cccd65
@@ -100,7 +101,6 @@ require (
 	github.com/asaskevich/govalidator v0.0.0-20210307081110-f21760c49a8d // indirect
 	github.com/beorn7/perks v1.0.1 // indirect
 	github.com/bits-and-blooms/bitset v1.2.0 // indirect
-	github.com/briandowns/spinner v1.18.1 // indirect
 	github.com/charmbracelet/bubbles v0.10.3 // indirect
 	github.com/charmbracelet/bubbletea v0.20.0 // indirect
 	github.com/charmbracelet/lipgloss v0.5.0 // indirect
@@ -164,7 +164,7 @@ require (
 	github.com/matttproud/golang_protobuf_extensions v1.0.2-0.20181231171920-c182affec369 // indirect
 	github.com/minio/colorjson v1.0.2 // indirect
 	github.com/minio/filepath v1.0.0 // indirect
-	github.com/minio/mc v0.0.0-20220512134321-aa60a8db1e4d // indirect
+	github.com/minio/mc v0.0.0-20220705180830-01b87ecc02ff // indirect
 	github.com/minio/md5-simd v1.1.2 // indirect
 	github.com/mitchellh/mapstructure v1.4.3 // indirect
 	github.com/modern-go/concurrent v0.0.0-20180306012644-bacd9c7ef1dd // indirect
@@ -184,7 +184,6 @@ require (
 	github.com/posener/complete v1.2.3 // indirect
 	github.com/power-devops/perfstat v0.0.0-20220216144756-c35f1ee13d7c // indirect
 	github.com/pquerna/cachecontrol v0.1.0 // indirect
-	github.com/prometheus/common v0.34.0 // indirect
 	github.com/prometheus/prom2json v1.3.1 // indirect
 	github.com/rcrowley/go-metrics v0.0.0-20201227073835-cf1acfcdf475 // indirect
 	github.com/rivo/tview v0.0.0-20220216162559-96063d6082f3 // indirect
@@ -220,5 +219,4 @@ require (
 	gopkg.in/h2non/filetype.v1 v1.0.5 // indirect
 	gopkg.in/ini.v1 v1.66.4 // indirect
 	gopkg.in/square/go-jose.v2 v2.5.1 // indirect
-	maze.io/x/duration v0.0.0-20160924141736-faac084b6075 // indirect
 )

internal/config/heal/heal.go

@@ -71,7 +71,7 @@ func (opts Config) BitrotScanCycle() (d time.Duration) {
 // Wait waits for IOCount to go down or max sleep to elapse before returning.
 // usually used in healing paths to wait for specified amount of time to
 // throttle healing.
-func (opts Config) Wait(currentIO func() int, systemIO func() int) {
+func (opts Config) Wait(currentIO func() int, activeListeners func() int) {
 	configMutex.RLock()
 	maxIO, maxWait := opts.IOCount, opts.Sleep
 	configMutex.RUnlock()
@@ -87,7 +87,7 @@ func (opts Config) Wait(currentIO func() int, systemIO func() int) {
 	tmpMaxWait := maxWait
 
 	if currentIO != nil {
-		for currentIO() >= maxIO+systemIO() {
+		for currentIO() >= maxIO+activeListeners() {
 			if tmpMaxWait > 0 {
 				if tmpMaxWait < waitTick {
 					time.Sleep(tmpMaxWait)

internal/event/event.go

@@ -17,6 +17,10 @@
 
 package event
 
+import (
+	"github.com/minio/madmin-go"
+)
+
 const (
 	// NamespaceFormat - namespace log format used in some event targets.
 	NamespaceFormat = "namespace"
@@ -79,6 +83,12 @@ type Event struct {
 	ResponseElements  map[string]string `json:"responseElements"`
 	S3                Metadata          `json:"s3"`
 	Source            Source            `json:"source"`
+	Type              madmin.TraceType  `json:"-"`
+}
+
+// Mask returns the type as mask.
+func (e Event) Mask() uint64 {
+	return e.EventName.Mask()
 }
 
 // Log represents event information for some event targets.

internal/event/name.go

@@ -30,12 +30,12 @@ type Name int
 
 // Values of event Name
 const (
-	ObjectAccessedAll Name = 1 + iota
+	// Single event types (does not require expansion)
-	ObjectAccessedGet
+
+	ObjectAccessedGet Name = 1 + iota
 	ObjectAccessedGetRetention
 	ObjectAccessedGetLegalHold
 	ObjectAccessedHead
-	ObjectCreatedAll
 	ObjectCreatedCompleteMultipartUpload
 	ObjectCreatedCopy
 	ObjectCreatedPost
@@ -44,12 +44,10 @@ const (
 	ObjectCreatedPutLegalHold
 	ObjectCreatedPutTagging
 	ObjectCreatedDeleteTagging
-	ObjectRemovedAll
 	ObjectRemovedDelete
 	ObjectRemovedDeleteMarkerCreated
 	BucketCreated
 	BucketRemoved
-	ObjectReplicationAll
 	ObjectReplicationFailed
 	ObjectReplicationComplete
 	ObjectReplicationMissedThreshold
@@ -57,19 +55,28 @@ const (
 	ObjectReplicationNotTracked
 	ObjectRestorePostInitiated
 	ObjectRestorePostCompleted
-	ObjectRestorePostAll
-	ObjectTransitionAll
 	ObjectTransitionFailed
 	ObjectTransitionComplete
+
+	objectSingleTypesEnd
+	// Start Compound types that require expansion:
+
+	ObjectAccessedAll
+	ObjectCreatedAll
+	ObjectRemovedAll
+	ObjectReplicationAll
+	ObjectRestorePostAll
+	ObjectTransitionAll
 )
 
+// The number of single names should not exceed 64.
+// This will break masking. Use bit 63 as extension.
+var _ = uint64(1 << objectSingleTypesEnd)
+
 // Expand - returns expanded values of abbreviated event type.
 func (name Name) Expand() []Name {
 	switch name {
-	case BucketCreated:
+
-		return []Name{BucketCreated}
-	case BucketRemoved:
-		return []Name{BucketRemoved}
 	case ObjectAccessedAll:
 		return []Name{
 			ObjectAccessedGet, ObjectAccessedHead,
@@ -110,6 +117,19 @@ func (name Name) Expand() []Name {
 	}
 }
 
+// Mask returns the type as mask.
+// Compound "All" types are expanded.
+func (name Name) Mask() uint64 {
+	if name < objectSingleTypesEnd {
+		return 1 << (name - 1)
+	}
+	var mask uint64
+	for _, n := range name.Expand() {
+		mask |= 1 << (n - 1)
+	}
+	return mask
+}
+
 // String - returns string representation of event type.
 func (name Name) String() string {
 	switch name {

internal/ioutil/read_file.go

@@ -25,12 +25,21 @@ import (
 	"github.com/minio/minio/internal/disk"
 )
 
+var (
+	// OpenFileDirectIO allows overriding default function.
+	OpenFileDirectIO = disk.OpenFileDirectIO
+	// OsOpen allows overriding default function.
+	OsOpen = os.Open
+	// OsOpenFile allows overriding default function.
+	OsOpenFile = os.OpenFile
+)
+
 // ReadFileWithFileInfo reads the named file and returns the contents.
 // A successful call returns err == nil, not err == EOF.
 // Because ReadFile reads the whole file, it does not treat an EOF from Read
 // as an error to be reported, additionall returns os.FileInfo
 func ReadFileWithFileInfo(name string) ([]byte, fs.FileInfo, error) {
-	f, err := os.Open(name)
+	f, err := OsOpen(name)
 	if err != nil {
 		return nil, nil, err
 	}
@@ -53,11 +62,11 @@ func ReadFileWithFileInfo(name string) ([]byte, fs.FileInfo, error) {
 //
 // passes NOATIME flag for reads on Unix systems to avoid atime updates.
 func ReadFile(name string) ([]byte, error) {
-	f, err := disk.OpenFileDirectIO(name, readMode, 0o666)
+	f, err := OpenFileDirectIO(name, readMode, 0o666)
 	if err != nil {
 		// fallback if there is an error to read
 		// 'name' with O_DIRECT
-		f, err = os.OpenFile(name, readMode, 0o666)
+		f, err = OsOpenFile(name, readMode, 0o666)
 		if err != nil {
 			return nil, err
 		}

internal/logger/audit.go

@@ -27,6 +27,7 @@ import (
 	"time"
 
 	"github.com/klauspost/compress/gzhttp"
+	"github.com/minio/madmin-go"
 	xhttp "github.com/minio/minio/internal/http"
 	"github.com/minio/minio/internal/logger/message/audit"
 )
@@ -234,8 +235,8 @@ func AuditLog(ctx context.Context, w http.ResponseWriter, r *http.Request, reqCl
 
 	// Send audit logs only to http targets.
 	for _, t := range auditTgts {
-		if err := t.Send(entry, string(All)); err != nil {
+		if err := t.Send(entry); err != nil {
-			LogAlwaysIf(context.Background(), fmt.Errorf("event(%v) was not sent to Audit target (%v): %v", entry, t, err), All)
+			LogAlwaysIf(context.Background(), fmt.Errorf("event(%v) was not sent to Audit target (%v): %v", entry, t, err), madmin.LogKindAll)
 		}
 	}
 }

internal/logger/logger.go

@@ -31,6 +31,7 @@ import (
 	"time"
 
 	"github.com/minio/highwayhash"
+	"github.com/minio/madmin-go"
 	"github.com/minio/minio-go/v7/pkg/set"
 	xhttp "github.com/minio/minio/internal/http"
 	"github.com/minio/minio/internal/logger/message/log"
@@ -50,6 +51,10 @@ const (
 	InformationLvl Level = iota + 1
 	ErrorLvl
 	FatalLvl
+
+	Application = madmin.LogKindApplication
+	Minio       = madmin.LogKindMinio
+	All         = madmin.LogKindAll
 )
 
 var trimStrings []string
@@ -65,16 +70,15 @@ var matchingFuncNames = [...]string{
 }
 
 func (level Level) String() string {
-	var lvlStr string
 	switch level {
 	case InformationLvl:
-		lvlStr = "INFO"
+		return "INFO"
 	case ErrorLvl:
-		lvlStr = "ERROR"
+		return "ERROR"
 	case FatalLvl:
-		lvlStr = "FATAL"
+		return "FATAL"
 	}
-	return lvlStr
+	return ""
 }
 
 // quietFlag: Hide startup messages if enabled
@@ -237,18 +241,6 @@ func hashString(input string) string {
 	return hex.EncodeToString(hh.Sum(nil))
 }
 
-// Kind specifies the kind of error log
-type Kind string
-
-const (
-	// Minio errors
-	Minio Kind = "MINIO"
-	// Application errors
-	Application Kind = "APPLICATION"
-	// All errors
-	All Kind = "ALL"
-)
-
 // LogAlwaysIf prints a detailed error message during
 // the execution of the server.
 func LogAlwaysIf(ctx context.Context, err error, errKind ...interface{}) {
@@ -279,10 +271,10 @@ func LogIf(ctx context.Context, err error, errKind ...interface{}) {
 }
 
 func errToEntry(ctx context.Context, err error, errKind ...interface{}) log.Entry {
-	logKind := string(Minio)
+	logKind := madmin.LogKindAll
 	if len(errKind) > 0 {
-		if ek, ok := errKind[0].(Kind); ok {
+		if ek, ok := errKind[0].(madmin.LogKind); ok {
-			logKind = string(ek)
+			logKind = ek
 		}
 	}
 	req := GetReqInfo(ctx)
@@ -366,7 +358,7 @@ func consoleLogIf(ctx context.Context, err error, errKind ...interface{}) {
 
 	if consoleTgt != nil {
 		entry := errToEntry(ctx, err, errKind...)
-		consoleTgt.Send(entry, entry.LogKind)
+		consoleTgt.Send(entry)
 	}
 }
 
@@ -385,10 +377,10 @@ func logIf(ctx context.Context, err error, errKind ...interface{}) {
 	entry := errToEntry(ctx, err, errKind...)
 	// Iterate over all logger targets to send the log entry
 	for _, t := range systemTgts {
-		if err := t.Send(entry, entry.LogKind); err != nil {
+		if err := t.Send(entry); err != nil {
 			if consoleTgt != nil {
 				entry.Trace.Message = fmt.Sprintf("event(%#v) was not sent to Logger target (%#v): %#v", entry, t, err)
-				consoleTgt.Send(entry, entry.LogKind)
+				consoleTgt.Send(entry)
 			}
 		}
 	}

internal/logger/message/log/entry.go

@@ -20,6 +20,8 @@ package log
 import (
 	"strings"
 	"time"
+
+	"github.com/minio/madmin-go"
 )
 
 // ObjectVersion object version key/versionId
@@ -54,7 +56,7 @@ type API struct {
 type Entry struct {
 	DeploymentID string         `json:"deploymentid,omitempty"`
 	Level        string         `json:"level"`
-	LogKind      string    `json:"errKind"`
+	LogKind      madmin.LogKind `json:"errKind"`
 	Time         time.Time      `json:"time"`
 	API          *API           `json:"api,omitempty"`
 	RemoteHost   string         `json:"remotehost,omitempty"`
@@ -73,9 +75,15 @@ type Info struct {
 	Err        error  `json:"-"`
 }
 
-// SendLog returns true if log pertains to node specified in args.
+// Mask returns the mask based on the error level.
-func (l Info) SendLog(node, logKind string) bool {
+func (l Info) Mask() uint64 {
-	nodeFltr := (node == "" || strings.EqualFold(node, l.NodeName))
+	return l.LogKind.LogMask().Mask()
-	typeFltr := strings.EqualFold(logKind, "all") || strings.EqualFold(l.LogKind, logKind)
+}
-	return nodeFltr && typeFltr
+
+// SendLog returns true if log pertains to node specified in args.
+func (l Info) SendLog(node string, logKind madmin.LogMask) bool {
+	if logKind.Contains(l.LogKind.LogMask()) {
+		return node == "" || strings.EqualFold(node, l.NodeName)
+	}
+	return false
 }

internal/logger/target/http/http.go

@@ -203,7 +203,7 @@ func New(config Config) *Target {
 }
 
 // Send log message 'e' to http target.
-func (h *Target) Send(entry interface{}, errKind string) error {
+func (h *Target) Send(entry interface{}) error {
 	select {
 	case <-h.doneCh:
 		return nil

internal/logger/target/kafka/kafka.go

@@ -48,7 +48,7 @@ type Target struct {
 }
 
 // Send log message 'e' to kafka target.
-func (h *Target) Send(entry interface{}, errKind string) error {
+func (h *Target) Send(entry interface{}) error {
 	select {
 	case <-h.doneCh:
 		return nil

internal/logger/targets.go

@@ -33,7 +33,7 @@ type Target interface {
 	Endpoint() string
 	Init() error
 	Cancel()
-	Send(entry interface{}, errKind string) error
+	Send(entry interface{}) error
 	Type() types.TargetType
 }
 

internal/pubsub/mask.go

@@ -0,0 +1,66 @@
+package pubsub
+
+import (
+	"math"
+	"math/bits"
+)
+
+// Mask allows filtering by a bitset mask.
+type Mask uint64
+
+const (
+	// MaskAll is the mask for all entries.
+	MaskAll Mask = math.MaxUint64
+)
+
+// MaskFromMaskable extracts mask from an interface.
+func MaskFromMaskable(m Maskable) Mask {
+	return Mask(m.Mask())
+}
+
+// Contains returns whether *all* flags in other is present in t.
+func (t Mask) Contains(other Mask) bool {
+	return t&other == other
+}
+
+// Overlaps returns whether *any* flags in t overlaps with other.
+func (t Mask) Overlaps(other Mask) bool {
+	return t&other != 0
+}
+
+// SingleType returns whether t has a single type set.
+func (t Mask) SingleType() bool {
+	return bits.OnesCount64(uint64(t)) == 1
+}
+
+// FromUint64 will set a mask to the uint64 value.
+func (t *Mask) FromUint64(m uint64) {
+	*t = Mask(m)
+}
+
+// Merge will merge other into t.
+func (t *Mask) Merge(other Mask) {
+	*t |= other
+}
+
+// MergeMaskable will merge other into t.
+func (t *Mask) MergeMaskable(other Maskable) {
+	*t |= Mask(other.Mask())
+}
+
+// SetIf will add other if b is true.
+func (t *Mask) SetIf(b bool, other Mask) {
+	if b {
+		*t |= other
+	}
+}
+
+// Mask returns the mask as a uint64.
+func (t Mask) Mask() uint64 {
+	return uint64(t)
+}
+
+// Maskable implementations must return their mask as a 64 bit uint.
+type Maskable interface {
+	Mask() uint64
+}

internal/pubsub/pubsub.go

@@ -25,27 +25,31 @@ import (
 
 // Sub - subscriber entity.
 type Sub struct {
-	ch     chan interface{}
+	ch     chan Maskable
-	filter func(entry interface{}) bool
+	types  Mask
+	filter func(entry Maskable) bool
 }
 
 // PubSub holds publishers and subscribers
 type PubSub struct {
-	subs           []*Sub
+	// atomics, keep at top:
+	types          uint64
 	numSubscribers int32
 	maxSubscribers int32
+
+	// not atomics:
+	subs []*Sub
 	sync.RWMutex
 }
 
 // Publish message to the subscribers.
 // Note that publish is always nob-blocking send so that we don't block on slow receivers.
 // Hence receivers should use buffered channel so as not to miss the published events.
-func (ps *PubSub) Publish(item interface{}) {
+func (ps *PubSub) Publish(item Maskable) {
 	ps.RLock()
 	defer ps.RUnlock()
-
 	for _, sub := range ps.subs {
-		if sub.filter == nil || sub.filter(item) {
+		if sub.types.Contains(Mask(item.Mask())) && (sub.filter == nil || sub.filter(item)) {
 			select {
 			case sub.ch <- item:
 			default:
@@ -55,38 +59,54 @@ func (ps *PubSub) Publish(item interface{}) {
 }
 
 // Subscribe - Adds a subscriber to pubsub system
-func (ps *PubSub) Subscribe(subCh chan interface{}, doneCh <-chan struct{}, filter func(entry interface{}) bool) error {
+func (ps *PubSub) Subscribe(mask Mask, subCh chan Maskable, doneCh <-chan struct{}, filter func(entry Maskable) bool) error {
 	totalSubs := atomic.AddInt32(&ps.numSubscribers, 1)
 	if ps.maxSubscribers > 0 && totalSubs > ps.maxSubscribers {
 		atomic.AddInt32(&ps.numSubscribers, -1)
 		return fmt.Errorf("the limit of `%d` subscribers is reached", ps.maxSubscribers)
 	}
-
 	ps.Lock()
 	defer ps.Unlock()
 
-	sub := &Sub{subCh, filter}
+	sub := &Sub{ch: subCh, types: mask, filter: filter}
 	ps.subs = append(ps.subs, sub)
 
+	// We hold a lock, so we are safe to update
+	combined := Mask(atomic.LoadUint64(&ps.types))
+	combined.Merge(mask)
+	atomic.StoreUint64(&ps.types, uint64(combined))
+
 	go func() {
 		<-doneCh
 
 		ps.Lock()
 		defer ps.Unlock()
-
+		var remainTypes Mask
 		for i, s := range ps.subs {
 			if s == sub {
 				ps.subs = append(ps.subs[:i], ps.subs[i+1:]...)
+			} else {
+				remainTypes.Merge(s.types)
 			}
 		}
+		atomic.StoreUint64(&ps.types, uint64(remainTypes))
 		atomic.AddInt32(&ps.numSubscribers, -1)
 	}()
 
 	return nil
 }
 
-// NumSubscribers returns the number of current subscribers
+// NumSubscribers returns the number of current subscribers,
-func (ps *PubSub) NumSubscribers() int32 {
+// If t is non-nil, the type is checked against the active subscribed types,
+// and 0 will be returned if nobody is subscribed for the type,
+// otherwise the *total* number of subscribers is returned.
+func (ps *PubSub) NumSubscribers(m Maskable) int32 {
+	if m != nil {
+		types := Mask(atomic.LoadUint64(&ps.types))
+		if !types.Overlaps(Mask(m.Mask())) {
+			return 0
+		}
+	}
 	return atomic.LoadInt32(&ps.numSubscribers)
 }
 

internal/pubsub/pubsub_test.go

@@ -25,51 +25,87 @@ import (
 
 func TestSubscribe(t *testing.T) {
 	ps := New(2)
-	ch1 := make(chan interface{}, 1)
+	ch1 := make(chan Maskable, 1)
-	ch2 := make(chan interface{}, 1)
+	ch2 := make(chan Maskable, 1)
 	doneCh := make(chan struct{})
 	defer close(doneCh)
-	if err := ps.Subscribe(ch1, doneCh, nil); err != nil {
+	if err := ps.Subscribe(MaskAll, ch1, doneCh, nil); err != nil {
 		t.Fatalf("unexpected error: %v", err)
 	}
-	if err := ps.Subscribe(ch2, doneCh, nil); err != nil {
+	if err := ps.Subscribe(MaskAll, ch2, doneCh, nil); err != nil {
 		t.Fatalf("unexpected error: %v", err)
 	}
 	ps.Lock()
 	defer ps.Unlock()
+
+	if len(ps.subs) != 2 || ps.NumSubscribers(nil) != 2 {
+		t.Fatalf("expected 2 subscribers")
+	}
+}
+
+func TestNumSubscribersMask(t *testing.T) {
+	ps := New(2)
+	ch1 := make(chan Maskable, 1)
+	ch2 := make(chan Maskable, 1)
+	doneCh := make(chan struct{})
+	defer close(doneCh)
+	if err := ps.Subscribe(1, ch1, doneCh, nil); err != nil {
+		t.Fatalf("unexpected error: %v", err)
+	}
+	if err := ps.Subscribe(2, ch2, doneCh, nil); err != nil {
+		t.Fatalf("unexpected error: %v", err)
+	}
+	ps.Lock()
+	defer ps.Unlock()
+
 	if len(ps.subs) != 2 {
 		t.Fatalf("expected 2 subscribers")
 	}
+	if want, got := int32(2), ps.NumSubscribers(Mask(1)); got != want {
+		t.Fatalf("want %d subscribers, got %d", want, got)
+	}
+	if want, got := int32(2), ps.NumSubscribers(Mask(2)); got != want {
+		t.Fatalf("want %d subscribers, got %d", want, got)
+	}
+	if want, got := int32(2), ps.NumSubscribers(Mask(1|2)); got != want {
+		t.Fatalf("want %d subscribers, got %d", want, got)
+	}
+	if want, got := int32(2), ps.NumSubscribers(nil); got != want {
+		t.Fatalf("want %d subscribers, got %d", want, got)
+	}
+	if want, got := int32(0), ps.NumSubscribers(Mask(4)); got != want {
+		t.Fatalf("want %d subscribers, got %d", want, got)
+	}
 }
 
 func TestSubscribeExceedingLimit(t *testing.T) {
 	ps := New(2)
-	ch1 := make(chan interface{}, 1)
+	ch1 := make(chan Maskable, 1)
-	ch2 := make(chan interface{}, 1)
+	ch2 := make(chan Maskable, 1)
-	ch3 := make(chan interface{}, 1)
+	ch3 := make(chan Maskable, 1)
 	doneCh := make(chan struct{})
 	defer close(doneCh)
-	if err := ps.Subscribe(ch1, doneCh, nil); err != nil {
+	if err := ps.Subscribe(MaskAll, ch1, doneCh, nil); err != nil {
 		t.Fatalf("unexpected error: %v", err)
 	}
-	if err := ps.Subscribe(ch2, doneCh, nil); err != nil {
+	if err := ps.Subscribe(MaskAll, ch2, doneCh, nil); err != nil {
 		t.Fatalf("unexpected error: %v", err)
 	}
-	if err := ps.Subscribe(ch3, doneCh, nil); err == nil {
+	if err := ps.Subscribe(MaskAll, ch3, doneCh, nil); err == nil {
 		t.Fatalf("unexpected nil err")
 	}
 }
 
 func TestUnsubscribe(t *testing.T) {
 	ps := New(2)
-	ch1 := make(chan interface{}, 1)
+	ch1 := make(chan Maskable, 1)
-	ch2 := make(chan interface{}, 1)
+	ch2 := make(chan Maskable, 1)
 	doneCh1 := make(chan struct{})
 	doneCh2 := make(chan struct{})
-	if err := ps.Subscribe(ch1, doneCh1, nil); err != nil {
+	if err := ps.Subscribe(MaskAll, ch1, doneCh1, nil); err != nil {
 		t.Fatalf("unexpected error: %v", err)
 	}
-	if err := ps.Subscribe(ch2, doneCh2, nil); err != nil {
+	if err := ps.Subscribe(MaskAll, ch2, doneCh2, nil); err != nil {
 		t.Fatalf("unexpected error: %v", err)
 	}
 
@@ -84,40 +120,81 @@ func TestUnsubscribe(t *testing.T) {
 	close(doneCh2)
 }
 
+type maskString string
+
+func (m maskString) Mask() uint64 {
+	return 1
+}
+
 func TestPubSub(t *testing.T) {
 	ps := New(1)
-	ch1 := make(chan interface{}, 1)
+	ch1 := make(chan Maskable, 1)
 	doneCh1 := make(chan struct{})
 	defer close(doneCh1)
-	if err := ps.Subscribe(ch1, doneCh1, func(entry interface{}) bool { return true }); err != nil {
+	if err := ps.Subscribe(MaskAll, ch1, doneCh1, func(entry Maskable) bool { return true }); err != nil {
 		t.Fatalf("unexpected error: %v", err)
 	}
-	val := "hello"
+	val := maskString("hello")
 	ps.Publish(val)
 	msg := <-ch1
-	if msg != "hello" {
+	if msg != val {
 		t.Fatalf(fmt.Sprintf("expected %s , found %s", val, msg))
 	}
 }
 
 func TestMultiPubSub(t *testing.T) {
 	ps := New(2)
-	ch1 := make(chan interface{}, 1)
+	ch1 := make(chan Maskable, 1)
-	ch2 := make(chan interface{}, 1)
+	ch2 := make(chan Maskable, 1)
 	doneCh := make(chan struct{})
 	defer close(doneCh)
-	if err := ps.Subscribe(ch1, doneCh, func(entry interface{}) bool { return true }); err != nil {
+	if err := ps.Subscribe(MaskAll, ch1, doneCh, func(entry Maskable) bool { return true }); err != nil {
 		t.Fatalf("unexpected error: %v", err)
 	}
-	if err := ps.Subscribe(ch2, doneCh, func(entry interface{}) bool { return true }); err != nil {
+	if err := ps.Subscribe(MaskAll, ch2, doneCh, func(entry Maskable) bool { return true }); err != nil {
 		t.Fatalf("unexpected error: %v", err)
 	}
-	val := "hello"
+	val := maskString("hello")
 	ps.Publish(val)
 
 	msg1 := <-ch1
 	msg2 := <-ch2
-	if msg1 != "hello" && msg2 != "hello" {
+	if msg1 != val && msg2 != val {
 		t.Fatalf(fmt.Sprintf("expected both subscribers to have%s , found %s and  %s", val, msg1, msg2))
 	}
 }
+
+func TestMultiPubSubMask(t *testing.T) {
+	ps := New(3)
+	ch1 := make(chan Maskable, 1)
+	ch2 := make(chan Maskable, 1)
+	ch3 := make(chan Maskable, 1)
+	doneCh := make(chan struct{})
+	defer close(doneCh)
+	// Mask matches maskString, should get result
+	if err := ps.Subscribe(Mask(1), ch1, doneCh, func(entry Maskable) bool { return true }); err != nil {
+		t.Fatalf("unexpected error: %v", err)
+	}
+	// Mask matches maskString, should get result
+	if err := ps.Subscribe(Mask(1|2), ch2, doneCh, func(entry Maskable) bool { return true }); err != nil {
+		t.Fatalf("unexpected error: %v", err)
+	}
+	// Does NOT overlap maskString
+	if err := ps.Subscribe(Mask(2), ch3, doneCh, func(entry Maskable) bool { return true }); err != nil {
+		t.Fatalf("unexpected error: %v", err)
+	}
+	val := maskString("hello")
+	ps.Publish(val)
+
+	msg1 := <-ch1
+	msg2 := <-ch2
+	if msg1 != val && msg2 != val {
+		t.Fatalf(fmt.Sprintf("expected both subscribers to have%s , found %s and  %s", val, msg1, msg2))
+	}
+
+	select {
+	case msg := <-ch3:
+		t.Fatalf(fmt.Sprintf("unexpect msg, f got %s", msg))
+	default:
+	}
+}