Update cclog calls in latest IB Collector

Merge branch 'main' into cclog_update
Update cclog calls
2026-06-09 21:37:29 +02:00 · 2026-06-08 14:58:29 +02:00 · 2026-06-08 14:57:31 +02:00 · 2026-06-08 14:52:24 +02:00 · 2026-06-08 14:00:09 +02:00 · 2026-06-08 13:10:27 +02:00
32 changed files with 728 additions and 275 deletions
--- a/cc-metric-collector.go
+++ b/cc-metric-collector.go
@@ -132,11 +132,11 @@ func mainFunc() int {
 	if len(rcfg.ConfigFile.Interval) > 0 {
 		t, err := time.ParseDuration(rcfg.ConfigFile.Interval)
 		if err != nil {
-			cclog.Error("Configuration value 'interval' no valid duration")
+			cclog.Errorf("Configuration value interval=%s no valid duration", rcfg.ConfigFile.Interval)
 		}
 		rcfg.Interval = t
 		if rcfg.Interval == 0 {
-			cclog.Error("Configuration value 'interval' must be greater than zero")
+			cclog.Errorf("Configuration value interval=%s must be greater than zero", rcfg.ConfigFile.Interval)
 			return 1
 		}
 	}
@@ -145,11 +145,11 @@ func mainFunc() int {
 	if len(rcfg.ConfigFile.Duration) > 0 {
 		t, err := time.ParseDuration(rcfg.ConfigFile.Duration)
 		if err != nil {
-			cclog.Error("Configuration value 'duration' no valid duration")
+			cclog.Error("Configuration value duration=%s no valid duration", rcfg.ConfigFile.Duration)
 		}
 		rcfg.Duration = t
 		if rcfg.Duration == 0 {
-			cclog.Error("Configuration value 'duration' must be greater than zero")
+			cclog.Error("Configuration value duration=%s must be greater than zero", rcfg.ConfigFile.Duration)
 			return 1
 		}
 	}
--- a/collectors/beegfsmetaMetric.go
+++ b/collectors/beegfsmetaMetric.go
@@ -209,16 +209,16 @@ func (m *BeegfsMetaCollector) Read(interval time.Duration, output chan lp.CCMess
 				} else {
 					f1, err := strconv.ParseFloat(m.matches["other"], 32)
 					if err != nil {
-						cclog.ComponentError(
+						cclog.ComponentErrorf(
 							m.name,
-							fmt.Sprintf("Metric (other): Failed to convert str written '%s' to float: %v", m.matches["other"], err))
+							"Metric (other): Failed to convert str written '%s' to float: %v", m.matches["other"], err)
 						continue
 					}
 					f2, err := strconv.ParseFloat(split[i], 32)
 					if err != nil {
-						cclog.ComponentError(
+						cclog.ComponentErrorf(
 							m.name,
-							fmt.Sprintf("Metric (other): Failed to convert str written '%s' to float: %v", m.matches["other"], err))
+							"Metric (other): Failed to convert str written '%s' to float: %v", m.matches["other"], err)
 						continue
 					}
 					m.matches["beegfs_cstorage_other"] = fmt.Sprintf("%f", f1+f2)
--- a/collectors/beegfsstorageMetric.go
+++ b/collectors/beegfsstorageMetric.go
@@ -200,16 +200,16 @@ func (m *BeegfsStorageCollector) Read(interval time.Duration, output chan lp.CCM
 				} else {
 					f1, err := strconv.ParseFloat(m.matches["other"], 32)
 					if err != nil {
-						cclog.ComponentError(
+						cclog.ComponentErrorf(
 							m.name,
-							fmt.Sprintf("Metric (other): Failed to convert str written '%s' to float: %v", m.matches["other"], err))
+							"Metric (other): Failed to convert str written '%s' to float: %v", m.matches["other"], err)
 						continue
 					}
 					f2, err := strconv.ParseFloat(split[i], 32)
 					if err != nil {
-						cclog.ComponentError(
+						cclog.ComponentErrorf(
 							m.name,
-							fmt.Sprintf("Metric (other): Failed to convert str written '%s' to float: %v", m.matches["other"], err))
+							"Metric (other): Failed to convert str written '%s' to float: %v", m.matches["other"], err)
 						continue
 					}
 					m.matches["beegfs_cstorage_other"] = fmt.Sprintf("%f", f1+f2)
--- a/collectors/collectorManager.go
+++ b/collectors/collectorManager.go
@@ -50,6 +50,7 @@ var AvailableCollectors = map[string]MetricCollector{
 	"nfsiostat":       new(NfsIOStatCollector),
 	"slurm_cgroup":    new(SlurmCgroupCollector),
 	"smartmon":        new(SmartMonCollector),
+	"nvidia_gpm":      new(NvidiaGPMCollector),
 }

 // Metric collector manager data structure
@@ -99,17 +100,17 @@ func (cm *collectorManager) Init(ticker mct.MultiChanTicker, duration time.Durat
 	// Initialize configured collectors
 	for collectorName, collectorCfg := range cm.config {
 		if _, found := AvailableCollectors[collectorName]; !found {
-			cclog.ComponentError("CollectorManager", "SKIP unknown collector", collectorName)
+			cclog.ComponentErrorf("CollectorManager", "SKIP unknown collector %s", collectorName)
 			continue
 		}
 		collector := AvailableCollectors[collectorName]

 		err := collector.Init(collectorCfg)
 		if err != nil {
-			cclog.ComponentError("CollectorManager", fmt.Sprintf("Collector %s initialization failed: %v", collectorName, err))
+			cclog.ComponentErrorf("CollectorManager", "Collector %s initialization failed: %v", collectorName, err)
 			continue
 		}
-		cclog.ComponentDebug("CollectorManager", "ADD COLLECTOR", collector.Name())
+		cclog.ComponentDebugf("CollectorManager", "ADD COLLECTOR %s", collector.Name())
 		if collector.Parallel() {
 			cm.collectors = append(cm.collectors, collector)
 		} else {
@@ -155,7 +156,7 @@ func (cm *collectorManager) Start() {
 						return
 					default:
 						// Read metrics from collector c via goroutine
-						cclog.ComponentDebug("CollectorManager", c.Name(), t)
+						cclog.ComponentDebugf("CollectorManager: Read %s at %v", c.Name(), t)
 						cm.collector_wg.Add(1)
 						go func(myc MetricCollector) {
 							myc.Read(cm.duration, cm.output)
@@ -173,7 +174,7 @@ func (cm *collectorManager) Start() {
 						return
 					default:
 						// Read metrics from collector c
-						cclog.ComponentDebug("CollectorManager", c.Name(), t)
+						cclog.ComponentDebugf("CollectorManager: Read %s at %v", c.Name(), t)
 						c.Read(cm.duration, cm.output)
 					}
 				}
--- a/collectors/cpufreqCpuinfoMetric.go
+++ b/collectors/cpufreqCpuinfoMetric.go
@@ -139,16 +139,16 @@ func (m *CPUFreqCpuInfoCollector) Read(interval time.Duration, output chan lp.CC
 	const cpuInfoFile = "/proc/cpuinfo"
 	file, err := os.Open(cpuInfoFile)
 	if err != nil {
-		cclog.ComponentError(
+		cclog.ComponentErrorf(
 			m.name,
-			fmt.Sprintf("Read(): Failed to open file '%s': %v", cpuInfoFile, err))
+			"Read(): Failed to open file '%s': %v", cpuInfoFile, err)
 		return
 	}
 	defer func() {
 		if err := file.Close(); err != nil {
-			cclog.ComponentError(
+			cclog.ComponentErrorf(
 				m.name,
-				fmt.Sprintf("Read(): Failed to close file '%s': %v", cpuInfoFile, err))
+				"Read(): Failed to close file '%s': %v", cpuInfoFile, err)
 		}
 	}()

@@ -166,9 +166,9 @@ func (m *CPUFreqCpuInfoCollector) Read(interval time.Duration, output chan lp.CC
 				if !t.isHT {
 					value, err := strconv.ParseFloat(strings.TrimSpace(lineSplit[1]), 64)
 					if err != nil {
-						cclog.ComponentError(
+						cclog.ComponentErrorf(
 							m.name,
-							fmt.Sprintf("Read(): Failed to convert cpu MHz '%s' to float64: %v", lineSplit[1], err))
+							"Read(): Failed to convert cpu MHz '%s' to float64: %v", lineSplit[1], err)
 						return
 					}
 					if y, err := lp.NewMetric("cpufreq", t.tagSet, m.meta, value, now); err == nil {
--- a/collectors/cpufreqMetric.go
+++ b/collectors/cpufreqMetric.go
@@ -95,10 +95,7 @@ func (m *CPUFreqCollector) Init(config json.RawMessage) error {
 	}

 	// Initialized
-	cclog.ComponentDebug(
-		m.name,
-		"initialized",
-		len(m.topology), "non-hyper-threading CPUs")
+	cclog.ComponentDebugf(m.name, "initialized %d non-hyper-threading CPUs")
 	m.init = true
 	return nil
 }
@@ -116,16 +113,14 @@ func (m *CPUFreqCollector) Read(interval time.Duration, output chan lp.CCMessage
 		// Read current frequency
 		line, err := os.ReadFile(t.scalingCurFreqFile)
 		if err != nil {
-			cclog.ComponentError(
-				m.name,
-				fmt.Sprintf("Read(): Failed to read file '%s': %v", t.scalingCurFreqFile, err))
+			cclog.ComponentErrorf(
+				m.name, "Read(): Failed to read file '%s': %v", t.scalingCurFreqFile, err)
 			continue
 		}
 		cpuFreq, err := strconv.ParseInt(strings.TrimSpace(string(line)), 10, 64)
 		if err != nil {
 			cclog.ComponentError(
-				m.name,
-				fmt.Sprintf("Read(): Failed to convert CPU frequency '%s' to int64: %v", line, err))
+				m.name, "Read(): Failed to convert CPU frequency '%s' to int64: %v", line, err)
 			continue
 		}

--- a/collectors/cpustatMetric.go
+++ b/collectors/cpustatMetric.go
@@ -27,6 +27,7 @@ const CPUSTATFILE = `/proc/stat`

 type CpustatCollectorConfig struct {
 	ExcludeMetrics []string `json:"exclude_metrics,omitempty"`
+	excludeNumCPUs bool
 }

 type CpustatCollector struct {
@@ -79,6 +80,7 @@ func (m *CpustatCollector) Init(config json.RawMessage) error {
 			m.matches[match] = index
 		}
 	}
+	m.config.excludeNumCPUs = slices.Contains(m.config.ExcludeMetrics, "num_cpus")

 	// Check input file
 	file, err := os.Open(CPUSTATFILE)
@@ -95,11 +97,13 @@ func (m *CpustatCollector) Init(config json.RawMessage) error {
 		line := scanner.Text()
 		linefields := strings.Fields(line)
 		if strings.Compare(linefields[0], "cpu") == 0 {
+			// Kernel system statistics for all CPUs
 			m.olddata["cpu"] = make(map[string]int64)
 			for k, v := range m.matches {
 				m.olddata["cpu"][k], _ = strconv.ParseInt(linefields[v], 0, 64)
 			}
 		} else if strings.HasPrefix(linefields[0], "cpu") && strings.Compare(linefields[0], "cpu") != 0 {
+			// Kernel system statistics per CPU
 			cpustr := strings.TrimLeft(linefields[0], "cpu")
 			cpu, _ := strconv.Atoi(cpustr)
 			m.cputags[linefields[0]] = map[string]string{
@@ -167,15 +171,15 @@ func (m *CpustatCollector) Read(interval time.Duration, output chan lp.CCMessage

 	file, err := os.Open(CPUSTATFILE)
 	if err != nil {
-		cclog.ComponentError(
+		cclog.ComponentErrorf(
 			m.name,
-			fmt.Sprintf("Read(): Failed to open file '%s': %v", CPUSTATFILE, err))
+			"Read(): Failed to open file '%s': %v", CPUSTATFILE, err)
 	}
 	defer func() {
 		if err := file.Close(); err != nil {
-			cclog.ComponentError(
+			cclog.ComponentErrorf(
 				m.name,
-				fmt.Sprintf("Read(): Failed to close file '%s': %v", string(CPUSTATFILE), err))
+				"Read(): Failed to close file '%s': %v", string(CPUSTATFILE), err)
 		}
 	}()

@@ -191,9 +195,10 @@ func (m *CpustatCollector) Read(interval time.Duration, output chan lp.CCMessage
 		}
 	}

-	num_cpus_metric, err := lp.NewMetric("num_cpus", m.nodetags, m.meta, num_cpus, now)
-	if err == nil {
-		output <- num_cpus_metric
+	if !m.config.excludeNumCPUs {
+		if num_cpus_metric, err := lp.NewMetric("num_cpus", m.nodetags, m.meta, num_cpus, now); err == nil {
+			output <- num_cpus_metric
+		}
 	}

 	m.lastTimestamp = now
--- a/collectors/customCmdMetric.go
+++ b/collectors/customCmdMetric.go
@@ -64,9 +64,9 @@ func (m *CustomCmdCollector) Init(config json.RawMessage) error {
 		cmdFields := strings.Fields(c)
 		command := exec.Command(cmdFields[0], cmdFields[1:]...)
 		if _, err := command.Output(); err != nil {
-			cclog.ComponentWarn(
+			cclog.ComponentWarnf(
 				m.name,
-				fmt.Sprintf("%s Init(): Execution of command \"%s\" failed: %v", m.name, command.String(), err))
+				"%s Init(): Execution of command \"%s\" failed: %v", m.name, command.String(), err)
 			continue
 		}
 		m.cmdFieldsSlice = append(m.cmdFieldsSlice, cmdFields)
@@ -77,7 +77,7 @@ func (m *CustomCmdCollector) Init(config json.RawMessage) error {
 		if _, err := os.ReadFile(fileName); err != nil {
 			cclog.ComponentWarn(
 				m.name,
-				fmt.Sprintf("%s Init(): Reading of file \"%s\" failed: %v", m.name, fileName, err))
+				"%s Init(): Reading of file \"%s\" failed: %v", m.name, fileName, err)
 			continue
 		}
 		m.files = append(m.files, fileName)
@@ -100,20 +100,18 @@ func (m *CustomCmdCollector) Read(interval time.Duration, output chan lp.CCMessa
 		command := exec.Command(cmdFields[0], cmdFields[1:]...)
 		stdout, err := command.Output()
 		if err != nil {
-			cclog.ComponentError(
+			cclog.ComponentErrorf(
 				m.name,
-				fmt.Sprintf("Read(): Failed to read command output for command \"%s\": %v", command.String(), err),
-			)
+				"Read(): Failed to read command output for command \"%s\": %v", command.String(), err)
 			continue
 		}

 		// Read and decode influxDB line-protocol from command output
 		metrics, err := lp.FromBytes(stdout)
 		if err != nil {
-			cclog.ComponentError(
+			cclog.ComponentErrorf(
 				m.name,
-				fmt.Sprintf("Read(): Failed to decode influx Message: %v", err),
-			)
+				"Read(): Failed to decode influx Message: %v", err)
 			continue
 		}
 		for _, metric := range metrics {
@@ -128,20 +126,18 @@ func (m *CustomCmdCollector) Read(interval time.Duration, output chan lp.CCMessa
 	for _, filename := range m.files {
 		input, err := os.ReadFile(filename)
 		if err != nil {
-			cclog.ComponentError(
+			cclog.ComponentErrorf(
 				m.name,
-				fmt.Sprintf("Read(): Failed to read file \"%s\": %v\n", filename, err),
-			)
+				"Read(): Failed to read file \"%s\": %v\n", filename, err)
 			continue
 		}

 		// Read and decode influxDB line-protocol from file
 		metrics, err := lp.FromBytes(input)
 		if err != nil {
-			cclog.ComponentError(
+			cclog.ComponentErrorf(
 				m.name,
-				fmt.Sprintf("Read(): Failed to decode influx Message: %v", err),
-			)
+				"Read(): Failed to decode influx Message: %v", err)
 			continue
 		}
 		for _, metric := range metrics {
--- a/collectors/diskstatMetric.go
+++ b/collectors/diskstatMetric.go
@@ -77,16 +77,16 @@ func (m *DiskstatCollector) Read(interval time.Duration, output chan lp.CCMessag

 	file, err := os.Open(MOUNTFILE)
 	if err != nil {
-		cclog.ComponentError(
+		cclog.ComponentErrorf(
 			m.name,
-			fmt.Sprintf("Read(): Failed to open file '%s': %v", MOUNTFILE, err))
+			"Read(): Failed to open file '%s': %v", MOUNTFILE, err)
 		return
 	}
 	defer func() {
 		if err := file.Close(); err != nil {
 			cclog.ComponentError(
 				m.name,
-				fmt.Sprintf("Read(): Failed to close file '%s': %v", MOUNTFILE, err))
+				"Read(): Failed to close file '%s': %v", MOUNTFILE, err)
 		}
 	}()

--- a/collectors/gpfsMetric.go
+++ b/collectors/gpfsMetric.go
@@ -371,7 +371,7 @@ func (m *GpfsCollector) Init(config json.RawMessage) error {
 	if err != nil {
 		// if using sudo, exec.lookPath will return EACCES (file mode r-x------), this can be ignored
 		if m.config.Sudo && errors.Is(err, syscall.EACCES) {
-			cclog.ComponentWarn(m.name, fmt.Sprintf("got error looking for mmpmon binary '%s': %v . This is expected when using sudo, continuing.", m.config.Mmpmon, err))
+			cclog.ComponentWarnf(m.name, "got error looking for mmpmon binary '%s': %v . This is expected when using sudo, continuing.", m.config.Mmpmon, err)
 			// the file was given in the config, use it
 			p = m.config.Mmpmon
 		} else {
@@ -517,23 +517,23 @@ func (m *GpfsCollector) Read(interval time.Duration, output chan lp.CCMessage) {
 		// return code
 		rc, err := strconv.Atoi(key_value["_rc_"])
 		if err != nil {
-			cclog.ComponentError(m.name, fmt.Sprintf("Read(): Failed to convert return code '%s' to int: %v", key_value["_rc_"], err))
+			cclog.ComponentErrorf(m.name, "Read(): Failed to convert return code '%s' to int: %v", key_value["_rc_"], err)
 			continue
 		}
 		if rc != 0 {
-			cclog.ComponentError(m.name, fmt.Sprintf("Read(): Filesystem '%s' is not ok.", filesystem))
+			cclog.ComponentErrorf(m.name, "Read(): Filesystem '%s' is not ok.", filesystem)
 			continue
 		}

 		// timestamp
 		sec, err := strconv.ParseInt(key_value["_t_"], 10, 64)
 		if err != nil {
-			cclog.ComponentError(m.name, fmt.Sprintf("Read(): Failed to convert seconds '%s' to int64: %v", key_value["_t_"], err))
+			cclog.ComponentErrorf(m.name, "Read(): Failed to convert seconds '%s' to int64: %v", key_value["_t_"], err)
 			continue
 		}
 		msec, err := strconv.ParseInt(key_value["_tu_"], 10, 64)
 		if err != nil {
-			cclog.ComponentError(m.name, fmt.Sprintf("Read(): Failed to convert micro seconds '%s' to int64: %v", key_value["_tu_"], err))
+			cclog.ComponentErrorf(m.name, "Read(): Failed to convert micro seconds '%s' to int64: %v", key_value["_tu_"], err)
 			continue
 		}
 		timestamp := time.Unix(sec, msec*1000)
@@ -551,7 +551,7 @@ func (m *GpfsCollector) Read(interval time.Duration, output chan lp.CCMessage) {
 		for _, metric := range GpfsAbsMetrics {
 			value, err := strconv.ParseInt(key_value[metric.prefix], 10, 64)
 			if err != nil {
-				cclog.ComponentError(m.name, fmt.Sprintf("Read(): Failed to convert %s '%s' to int64: %v", metric.desc, key_value[metric.prefix], err))
+				cclog.ComponentErrorf(m.name, "Read(): Failed to convert %s '%s' to int64: %v", metric.desc, key_value[metric.prefix], err)
 				continue
 			}
 			newstate[metric.prefix] = value
@@ -636,7 +636,7 @@ func (m *GpfsCollector) Read(interval time.Duration, output chan lp.CCMessage) {
 				}
 			} else {
 				// the value could not be computed correctly
-				cclog.ComponentWarn(m.name, fmt.Sprintf("Read(): Could not compute value for filesystem %s of metric %s: vold_ok = %t, vnew_ok = %t", filesystem, metric.name, vold_ok, vnew_ok))
+				cclog.ComponentWarnf(m.name, "Read(): Could not compute value for filesystem %s of metric %s: vold_ok = %t, vnew_ok = %t", filesystem, metric.name, vold_ok, vnew_ok)
 			}
 		}

--- a/collectors/infinibandMetric.go
+++ b/collectors/infinibandMetric.go
@@ -23,20 +23,29 @@ import (
 	"golang.org/x/sys/unix"
 )

-const IB_BASEPATH = "/sys/class/infiniband/"
+// See: https://www.kernel.org/doc/Documentation/ABI/stable/sysfs-class-infiniband
+const (
+	ibBasePath     = "/sys/class/infiniband/"
+	ibDataUnit     = "bytes"
+	ibDataRateUnit = ibDataUnit + "/sec"
+	ibPkgUnit      = "packets"
+	ibPkgRateUnit  = ibPkgUnit + "/sec"
+)

 type InfinibandCollectorMetric struct {
-	name             string
-	path             string
-	unit             string
-	scale            int64
-	addToIBTotal     bool
-	addToIBTotalPkgs bool
-	lastState        int64
+	name               string
+	path               string
+	unit               string
+	unitRates          string
+	scaleByFourLanes   bool
+	addToIBTotal       bool
+	addToIBTotalPkgs   bool
+	lastState          uint64
+	lastStateAvailable bool
 }

 type InfinibandCollectorInfo struct {
-	LID              string                      // IB local Identifier (LID)
+	lid              string                      // IB local Identifier (LID)
 	device           string                      // IB device
 	port             string                      // IB device port
 	portCounterFiles []InfinibandCollectorMetric // mapping counter name -> InfinibandCollectorMetric
@@ -56,7 +65,7 @@ type InfinibandCollector struct {
 	lastTimestamp time.Time // Store time stamp of last tick to derive bandwidths
 }

-// Init initializes the Infiniband collector by walking through files below IB_BASEPATH
+// Init initializes the Infiniband collector by walking through files below ibBasePath
 func (m *InfinibandCollector) Init(config json.RawMessage) error {
 	// Check if already initialized
 	if m.init {
@@ -87,7 +96,7 @@ func (m *InfinibandCollector) Init(config json.RawMessage) error {
 	}

 	// Loop for all InfiniBand directories
-	globPattern := filepath.Join(IB_BASEPATH, "*", "ports", "*")
+	globPattern := filepath.Join(ibBasePath, "*", "ports", "*")
 	ibDirs, err := filepath.Glob(globPattern)
 	if err != nil {
 		return fmt.Errorf("%s Init(): unable to glob files with pattern %s: %w", m.name, globPattern, err)
@@ -122,36 +131,42 @@ func (m *InfinibandCollector) Init(config json.RawMessage) error {
 		countersDir := filepath.Join(path, "counters")
 		portCounterFiles := []InfinibandCollectorMetric{
 			{
-				name:         "ib_recv",
-				path:         filepath.Join(countersDir, "port_rcv_data"),
-				unit:         "bytes",
-				scale:        4,
-				addToIBTotal: true,
-				lastState:    -1,
+				// Total number of data octets, divided by 4 (lanes), received on all VLs.
+				// This is 64 bit counter
+				name:             "ib_recv",
+				path:             filepath.Join(countersDir, "port_rcv_data"),
+				unit:             ibDataUnit,
+				unitRates:        ibDataRateUnit,
+				scaleByFourLanes: true,
+				addToIBTotal:     true,
 			},
 			{
-				name:         "ib_xmit",
-				path:         filepath.Join(countersDir, "port_xmit_data"),
-				unit:         "bytes",
-				scale:        4,
-				addToIBTotal: true,
-				lastState:    -1,
+				// Total number of data octets, divided by 4 (lanes), transmitted on all VLs.
+				// This is 64 bit counter
+				name:             "ib_xmit",
+				path:             filepath.Join(countersDir, "port_xmit_data"),
+				unit:             ibDataUnit,
+				unitRates:        ibDataRateUnit,
+				scaleByFourLanes: true,
+				addToIBTotal:     true,
 			},
 			{
+				// Total number of packets received on all VLs from this port (this may include packets containing Errors.
+				// This is 64 bit counter.
 				name:             "ib_recv_pkts",
 				path:             filepath.Join(countersDir, "port_rcv_packets"),
-				unit:             "packets",
-				scale:            1,
+				unit:             ibPkgUnit,
+				unitRates:        ibPkgRateUnit,
 				addToIBTotalPkgs: true,
-				lastState:        -1,
 			},
 			{
+				// Total number of packets transmitted on all VLs from this port. This may include packets with errors.
+				// This is 64 bit counter.
 				name:             "ib_xmit_pkts",
 				path:             filepath.Join(countersDir, "port_xmit_packets"),
-				unit:             "packets",
-				scale:            1,
+				unit:             ibPkgUnit,
+				unitRates:        ibPkgRateUnit,
 				addToIBTotalPkgs: true,
-				lastState:        -1,
 			},
 		}
 		for _, counter := range portCounterFiles {
@@ -163,7 +178,7 @@ func (m *InfinibandCollector) Init(config json.RawMessage) error {

 		m.info = append(m.info,
 			InfinibandCollectorInfo{
-				LID:              LID,
+				lid:              LID,
 				device:           device,
 				port:             port,
 				portCounterFiles: portCounterFiles,
@@ -184,7 +199,7 @@ func (m *InfinibandCollector) Init(config json.RawMessage) error {
 	return nil
 }

-// Read reads Infiniband counter files below IB_BASEPATH
+// Read reads Infiniband counter files below ibBasePath
 func (m *InfinibandCollector) Read(interval time.Duration, output chan lp.CCMessage) {
 	// Check if already initialized
 	if !m.init {
@@ -201,36 +216,42 @@ func (m *InfinibandCollector) Read(interval time.Duration, output chan lp.CCMess
 	for i := range m.info {
 		info := &m.info[i]

-		var ib_total, ib_total_last_state,
-			ib_total_pkts, ib_total_pkts_last_state int64
-		var ib_total_last_state_available, ib_total_pkts_last_state_available bool
+		var ibTotal, ibTotalPkts uint64      // sum of xmit and recv counters
+		var ibTotalBw, ibTotalPktsBw float64 // sum of xmit and recv rates
+		var ibTotalBwAvailable, ibTotalPktsBwAvailable bool
 		for i := range info.portCounterFiles {
 			counterDef := &info.portCounterFiles[i]

 			// Read counter file
 			line, err := os.ReadFile(counterDef.path)
 			if err != nil {
-				cclog.ComponentError(
+				cclog.ComponentErrorf(
 					m.name,
-					fmt.Sprintf("Read(): Failed to read from file '%s': %v", counterDef.path, err))
+					"Read(): Failed to read from file '%s': %v", counterDef.path, err)
+				// Current counter can not be saved as last state
+				counterDef.lastStateAvailable = false
 				continue
 			}
 			data := strings.TrimSpace(string(line))

-			// convert counter to int64
-			v, err := strconv.ParseInt(data, 10, 64)
+			// convert counter to uint64
+			vRawCounter, err := strconv.ParseUint(data, 10, 64)
 			if err != nil {
-				cclog.ComponentError(
+				cclog.ComponentErrorf(
 					m.name,
-					fmt.Sprintf("Read(): Failed to convert Infininiband metrice %s='%s' to int64: %v", counterDef.name, data, err))
+					"Read(): Failed to convert Infininiband metrice %s='%s' to uint64: %v", counterDef.name, data, err)
+				// Current counter can not be saved as last state
+				counterDef.lastStateAvailable = false
 				continue
 			}
-			// Scale raw value
-			v *= counterDef.scale
+			vScaledCounter := vRawCounter
+			if counterDef.scaleByFourLanes {
+				vScaledCounter *= uint64(4)
+			}

 			// Send absolut values
 			if m.config.SendAbsoluteValues {
-				if y, err := lp.NewMetric(counterDef.name, info.tagSet, m.meta, v, now); err == nil {
+				if y, err := lp.NewMetric(counterDef.name, info.tagSet, m.meta, vScaledCounter, now); err == nil {
 					y.AddMeta("unit", counterDef.unit)
 					output <- y
 				}
@@ -238,63 +259,72 @@ func (m *InfinibandCollector) Read(interval time.Duration, output chan lp.CCMess

 			// Send derived values
 			if m.config.SendDerivedValues {
-				if counterDef.lastState >= 0 {
-					rate := float64((v - counterDef.lastState)) / timeDiff
+				if counterDef.lastStateAvailable {
+					var rate float64
+					// uint64 subtraction handles wraparound automatically
+					// in case vRawCounter < counterDef.lastState we would compute:
+					// math.MaxUint64 - lastState + vRawCounter + 1
+					// = (2^64 - 1) - lastState + vRawCounter + 1
+					// = 2^64 - lastState + vRawCounter
+					// ≡ vRawCounter - lastState (mod 2^64)
+					rate = float64(vRawCounter-counterDef.lastState) / timeDiff
+					if counterDef.scaleByFourLanes {
+						rate *= float64(4)
+					}
 					if y, err := lp.NewMetric(counterDef.name+"_bw", info.tagSet, m.meta, rate, now); err == nil {
-						y.AddMeta("unit", counterDef.unit+"/sec")
+						y.AddMeta("unit", counterDef.unitRates)
 						output <- y
 					}

-					// Sum up total values of last state
+					// Sum up rates for total rates
 					if m.config.SendTotalValues {
 						switch {
 						case counterDef.addToIBTotal:
-							ib_total_last_state += counterDef.lastState
-							ib_total_last_state_available = true
+							ibTotalBw += rate
+							ibTotalBwAvailable = true
 						case counterDef.addToIBTotalPkgs:
-							ib_total_pkts_last_state += counterDef.lastState
-							ib_total_pkts_last_state_available = true
+							ibTotalPktsBw += rate
+							ibTotalPktsBwAvailable = true
 						}
 					}
 				}
-				counterDef.lastState = v
+				counterDef.lastState = vRawCounter
+				counterDef.lastStateAvailable = true
 			}

 			// Sum up total values
 			if m.config.SendTotalValues {
 				switch {
 				case counterDef.addToIBTotal:
-					ib_total += v
+					ibTotal += vScaledCounter
 				case counterDef.addToIBTotalPkgs:
-					ib_total_pkts += v
+					ibTotalPkts += vScaledCounter
 				}
 			}
 		}

 		// Send total values
 		if m.config.SendTotalValues {
-			if y, err := lp.NewMetric("ib_total", info.tagSet, m.meta, ib_total, now); err == nil {
-				y.AddMeta("unit", "bytes")
+			if y, err := lp.NewMetric("ib_total", info.tagSet, m.meta, ibTotal, now); err == nil {
+				y.AddMeta("unit", ibDataUnit)
 				output <- y
 			}

-			if y, err := lp.NewMetric("ib_total_pkts", info.tagSet, m.meta, ib_total_pkts, now); err == nil {
-				y.AddMeta("unit", "packets")
+			if y, err := lp.NewMetric("ib_total_pkts", info.tagSet, m.meta, ibTotalPkts, now); err == nil {
+				y.AddMeta("unit", ibPkgUnit)
 				output <- y
 			}

-			if m.config.SendDerivedValues && ib_total_last_state_available {
-				rate := float64((ib_total - ib_total_last_state)) / timeDiff
-				if y, err := lp.NewMetric("ib_total_bw", info.tagSet, m.meta, rate, now); err == nil {
-					y.AddMeta("unit", "bytes/sec")
+			if m.config.SendDerivedValues && ibTotalBwAvailable {
+				if y, err := lp.NewMetric("ib_total_bw", info.tagSet, m.meta, ibTotalBw, now); err == nil {
+					y.AddMeta("unit", ibDataRateUnit)
 					output <- y
 				}
 			}

-			if m.config.SendDerivedValues && ib_total_pkts_last_state_available {
-				rate := float64((ib_total_pkts - ib_total_pkts_last_state)) / timeDiff
-				if y, err := lp.NewMetric("ib_total_pkts_bw", info.tagSet, m.meta, rate, now); err == nil {
-					y.AddMeta("unit", "packets/sec")
+			if m.config.SendDerivedValues && ibTotalPktsBwAvailable {
+				if y, err := lp.NewMetric("ib_total_pkts_bw", info.tagSet, m.meta, ibTotalPktsBw, now); err == nil {
+					y.AddMeta("unit", ibPkgRateUnit)
 					output <- y
 				}
 			}
--- a/collectors/iostatMetric.go
+++ b/collectors/iostatMetric.go
@@ -145,16 +145,16 @@ func (m *IOstatCollector) Read(interval time.Duration, output chan lp.CCMessage)

 	file, err := os.Open(IOSTATFILE)
 	if err != nil {
-		cclog.ComponentError(
+		cclog.ComponentErrorf(
 			m.name,
-			fmt.Sprintf("Read(): Failed to open file '%s': %v", IOSTATFILE, err))
+			"Read(): Failed to open file '%s': %v", IOSTATFILE, err)
 		return
 	}
 	defer func() {
 		if err := file.Close(); err != nil {
-			cclog.ComponentError(
+			cclog.ComponentErrorf(
 				m.name,
-				fmt.Sprintf("Read(): Failed to close file '%s': %v", IOSTATFILE, err))
+				"Read(): Failed to close file '%s': %v", IOSTATFILE, err)
 		}
 	}()

--- a/collectors/likwidMetric.go
+++ b/collectors/likwidMetric.go
@@ -12,6 +12,12 @@ package collectors
 #cgo LDFLAGS: -Wl,--unresolved-symbols=ignore-in-object-files
 #include <stdlib.h>
 #include <likwid.h>
+
+
+int cc_add_hwthread(int cpu_id) {
+    return HPMaddThread(cpu_id);
+}
+
 */
 import "C"

@@ -261,12 +267,12 @@ func (m *LikwidCollector) Init(config json.RawMessage) error {
 			}
 			for _, metric := range evset.Metrics {
 				// Try to evaluate the metric
-				cclog.ComponentDebug(m.name, "Checking", metric.Name)
+				cclog.ComponentDebugf(m.name, "Checking %s", metric.Name)
 				if !checkMetricType(metric.Type) {
-					cclog.ComponentError(m.name, "Metric", metric.Name, "uses invalid type", metric.Type)
+					cclog.ComponentErrorf(m.name, "Metric %s uses invalid type %s", metric.Name, metric.Type)
 					metric.Calc = ""
 				} else if !testLikwidMetricFormula(metric.Calc, params) {
-					cclog.ComponentError(m.name, "Metric", metric.Name, "cannot be calculated with given counters")
+					cclog.ComponentError(m.name, "Metric %s cannot be calculated with given counters", metric.Name)
 					metric.Calc = ""
 				} else {
 					globalParams = append(globalParams, metric.Name)
@@ -281,13 +287,13 @@ func (m *LikwidCollector) Init(config json.RawMessage) error {
 	for _, metric := range m.config.Metrics {
 		// Try to evaluate the global metric
 		if !checkMetricType(metric.Type) {
-			cclog.ComponentError(m.name, "Metric", metric.Name, "uses invalid type", metric.Type)
+			cclog.ComponentErrorf(m.name, "Metric %s uses invalid type %s", metric.Name, metric.Type)
 			metric.Calc = ""
 		} else if !testLikwidMetricFormula(metric.Calc, globalParams) {
-			cclog.ComponentError(m.name, "Metric", metric.Name, "cannot be calculated with given counters")
+			cclog.ComponentError(m.name, "Metric %s cannot be calculated with given counters", metric.Name)
 			metric.Calc = ""
 		} else if !checkMetricType(metric.Type) {
-			cclog.ComponentError(m.name, "Metric", metric.Name, "has invalid type")
+			cclog.ComponentError(m.name, "Metric %s has invalid type", metric.Name)
 			metric.Calc = ""
 		} else {
 			totalMetrics++
@@ -328,7 +334,7 @@ func (m *LikwidCollector) Init(config json.RawMessage) error {
 		for _, c := range m.cpulist {
 			m.measureThread.Call(
 				func() {
-					retCode := C.HPMaddThread(C.uint32_t(c))
+					retCode := C.cc_add_hwthread(C.int(c))
 					if retCode != 0 {
 						err := fmt.Errorf("C.HPMaddThread(%v) failed with return code %v", c, retCode)
 						cclog.ComponentError(m.name, err.Error())
@@ -375,16 +381,16 @@ func (m *LikwidCollector) takeMeasurement(evidx int, evset LikwidEventsetConfig,
 	// Watch changes for the lock file ()
 	watcher, err := fsnotify.NewWatcher()
 	if err != nil {
-		cclog.ComponentError(
+		cclog.ComponentErrorf(
 			m.name,
-			fmt.Sprintf("takeMeasurement(): Failed to create a new fsnotify.Watcher: %v", err))
+			"takeMeasurement(): Failed to create a new fsnotify.Watcher: %v", err)
 		return true, err
 	}
 	defer func() {
 		if err := watcher.Close(); err != nil {
-			cclog.ComponentError(
+			cclog.ComponentErrorf(
 				m.name,
-				fmt.Sprintf("takeMeasurement(): Failed to close fsnotify.Watcher: %v", err))
+				"takeMeasurement(): Failed to close fsnotify.Watcher: %v", err)
 		}
 	}()
 	if len(m.config.LockfilePath) > 0 {
@@ -597,7 +603,7 @@ func (m *LikwidCollector) calcEventsetMetrics(evset LikwidEventsetConfig, interv
 			if tid >= 0 && len(metric.Calc) > 0 {
 				value, err := agg.EvalFloat64Condition(metric.Calc, evset.results[tid])
 				if err != nil {
-					cclog.ComponentError(m.name, "Calculation for metric", metric.Name, "failed:", err.Error())
+					cclog.ComponentErrorf(m.name, "Calculation for metric %s failed: %s", metric.Name, err.Error())
 					value = 0.0
 				}
 				if m.config.InvalidToZero && (math.IsNaN(value) || math.IsInf(value, 0)) {
@@ -762,7 +768,7 @@ func (m *LikwidCollector) calcGlobalMetrics(groups []LikwidEventsetConfig, inter
 				// Evaluate the metric
 				value, err := agg.EvalFloat64Condition(metric.Calc, params)
 				if err != nil {
-					cclog.ComponentError(m.name, "Calculation for metric", metric.Name, "failed:", err.Error())
+					cclog.ComponentErrorf(m.name, "Calculation for metric %s failed: %s", metric.Name, err.Error())
 					value = 0.0
 				}
 				if m.config.InvalidToZero && (math.IsNaN(value) || math.IsInf(value, 0)) {
--- a/collectors/loadavgMetric.go
+++ b/collectors/loadavgMetric.go
@@ -89,9 +89,9 @@ func (m *LoadavgCollector) Read(interval time.Duration, output chan lp.CCMessage
 	}
 	buffer, err := os.ReadFile(LOADAVGFILE)
 	if err != nil {
-		cclog.ComponentError(
+		cclog.ComponentErrorf(
 			m.name,
-			fmt.Sprintf("Read(): Failed to read file '%s': %v", LOADAVGFILE, err))
+			"Read(): Failed to read file '%s': %v", LOADAVGFILE, err)
 		return
 	}
 	now := time.Now()
@@ -101,9 +101,9 @@ func (m *LoadavgCollector) Read(interval time.Duration, output chan lp.CCMessage
 	for i, name := range m.load_matches {
 		x, err := strconv.ParseFloat(ls[i], 64)
 		if err != nil {
-			cclog.ComponentError(
+			cclog.ComponentErrorf(
 				m.name,
-				fmt.Sprintf("Read(): Failed to convert '%s' to float64: %v", ls[i], err))
+				"Read(): Failed to convert '%s' to float64: %v", ls[i], err)
 			continue
 		}
 		if m.load_skips[i] {
@@ -120,9 +120,9 @@ func (m *LoadavgCollector) Read(interval time.Duration, output chan lp.CCMessage
 	for i, name := range m.proc_matches {
 		x, err := strconv.ParseInt(lv[i], 10, 64)
 		if err != nil {
-			cclog.ComponentError(
+			cclog.ComponentErrorf(
 				m.name,
-				fmt.Sprintf("Read(): Failed to convert '%s' to float64: %v", lv[i], err))
+				"Read(): Failed to convert '%s' to float64: %v", lv[i], err)
 			continue
 		}
 		if m.proc_skips[i] {
--- a/collectors/memstatMetric.go
+++ b/collectors/memstatMetric.go
@@ -72,7 +72,8 @@ func getStats(filename string) map[string]MemstatStats {
 	for scanner.Scan() {
 		line := scanner.Text()
 		linefields := strings.Fields(line)
-		if len(linefields) == 3 {
+		switch len(linefields) {
+		case 3:
 			v, err := strconv.ParseFloat(linefields[1], 64)
 			if err == nil {
 				stats[strings.Trim(linefields[0], ":")] = MemstatStats{
@@ -80,10 +81,10 @@ func getStats(filename string) map[string]MemstatStats {
 					unit:  linefields[2],
 				}
 			}
-		} else if len(linefields) == 5 {
+		case 5:
 			v, err := strconv.ParseFloat(linefields[3], 64)
 			if err == nil {
-				cclog.ComponentDebug("getStats", strings.Trim(linefields[2], ":"), v, linefields[4])
+				cclog.ComponentDebug("MemstatCollector", "getStats %s value %v unit %s", strings.Trim(linefields[2], ":"), v, linefields[4])
 				stats[strings.Trim(linefields[2], ":")] = MemstatStats{
 					value: v,
 					unit:  linefields[4],
@@ -106,7 +107,10 @@ func (m *MemstatCollector) Init(config json.RawMessage) error {
 			return fmt.Errorf("%s Init(): Error decoding JSON config: %w", m.name, err)
 		}
 	}
-	m.meta = map[string]string{"source": m.name, "group": "Memory"}
+	m.meta = map[string]string{
+		"source": m.name,
+		"group":  "Memory",
+	}
 	m.stats = make(map[string]int64)
 	m.matches = make(map[string]string)
 	m.tags = map[string]string{"type": "node"}
@@ -145,7 +149,7 @@ func (m *MemstatCollector) Init(config json.RawMessage) error {
 		"KernelStack":     "mem_kernelstack",
 	}
 	for k, v := range matches {
-		if !slices.Contains(m.config.ExcludeMetrics, k) {
+		if !slices.Contains(m.config.ExcludeMetrics, v) {
 			m.matches[k] = v
 		}
 	}
@@ -153,7 +157,7 @@ func (m *MemstatCollector) Init(config json.RawMessage) error {
 	if !slices.Contains(m.config.ExcludeMetrics, "mem_used") {
 		m.sendMemUsed = true
 	}
-	if len(m.matches) == 0 {
+	if len(m.matches) == 0 && !m.sendMemUsed {
 		return fmt.Errorf("%s Init(): no metrics to collect", m.name)
 	}
 	if err := m.setup(); err != nil {
--- a/collectors/netstatMetric.go
+++ b/collectors/netstatMetric.go
@@ -222,16 +222,16 @@ func (m *NetstatCollector) Read(interval time.Duration, output chan lp.CCMessage

 	file, err := os.Open(NETSTATFILE)
 	if err != nil {
-		cclog.ComponentError(
+		cclog.ComponentErrorf(
 			m.name,
-			fmt.Sprintf("Read(): Failed to open file '%s': %v", NETSTATFILE, err))
+			"Read(): Failed to open file '%s': %v", NETSTATFILE, err)
 		return
 	}
 	defer func() {
 		if err := file.Close(); err != nil {
-			cclog.ComponentError(
+			cclog.ComponentErrorf(
 				m.name,
-				fmt.Sprintf("Read(): Failed to close file '%s': %v", NETSTATFILE, err))
+				"Read(): Failed to close file '%s': %v", NETSTATFILE, err)
 		}
 	}()

--- a/collectors/nfsMetric.go
+++ b/collectors/nfsMetric.go
@@ -125,10 +125,9 @@ func (m *nfsCollector) Read(interval time.Duration, output chan lp.CCMessage) {
 	timestamp := time.Now()

 	if err := m.updateStats(); err != nil {
-		cclog.ComponentError(
+		cclog.ComponentErrorf(
 			m.name,
-			fmt.Sprintf("Read(): updateStats() failed: %v", err),
-		)
+			"Read(): updateStats() failed: %v", err)
 		return
 	}
 	var prefix string
--- a/collectors/numastatsMetric.go
+++ b/collectors/numastatsMetric.go
@@ -117,7 +117,7 @@ func (m *NUMAStatsCollector) Init(config json.RawMessage) error {
 	}

 	// Initialized
-	cclog.ComponentDebug(m.name, "initialized", len(m.topology), "NUMA domains")
+	cclog.ComponentDebugf(m.name, "initialized %d NUMA domains", len(m.topology))
 	m.init = true
 	return nil
 }
--- a/collectors/nvidiaGPMMetric.go
+++ b/collectors/nvidiaGPMMetric.go
@@ -0,0 +1,396 @@
+package collectors
+
+import (
+	"encoding/json"
+	"errors"
+	"fmt"
+	"slices"
+	"strconv"
+	"strings"
+	"time"
+
+	cclog "github.com/ClusterCockpit/cc-lib/v2/ccLogger"
+	lp "github.com/ClusterCockpit/cc-lib/v2/ccMessage"
+	"github.com/NVIDIA/go-nvml/pkg/nvml"
+)
+
+type NvidiaGPMMetricDef struct {
+	name    string
+	outname string
+	id      nvml.GpmMetricId
+	unit    string
+}
+
+var NvidiaGPMMetrics []NvidiaGPMMetricDef = []NvidiaGPMMetricDef{
+	{
+		name:    "GRAPHICS_UTIL",
+		outname: "nv_gpm_graphics_util",
+		id:      nvml.GPM_METRIC_GRAPHICS_UTIL,
+		unit:    "%",
+	},
+	{
+		name:    "SM_UTIL",
+		outname: "nv_gpm_sm_util",
+		id:      nvml.GPM_METRIC_SM_UTIL,
+		unit:    "%",
+	},
+	{
+		name:    "SM_OCCUPANCY",
+		outname: "nv_gpm_sm_occupancy",
+		id:      nvml.GPM_METRIC_SM_OCCUPANCY,
+		unit:    "%",
+	},
+	{
+		name:    "INTEGER_UTIL",
+		outname: "nv_gpm_integer_util",
+		id:      nvml.GPM_METRIC_INTEGER_UTIL,
+		unit:    "%",
+	},
+	{
+		name:    "ANY_TENSOR_UTIL",
+		outname: "nv_gpm_any_tensor_util",
+		id:      nvml.GPM_METRIC_ANY_TENSOR_UTIL,
+		unit:    "%",
+	},
+	{
+		name:    "DFMA_TENSOR_UTIL",
+		outname: "nv_gpm_dfma_tensor_util",
+		id:      nvml.GPM_METRIC_DFMA_TENSOR_UTIL,
+		unit:    "%",
+	},
+	{
+		name:    "HMMA_TENSOR_UTIL",
+		outname: "nv_gpm_hmma_tensor_util",
+		id:      nvml.GPM_METRIC_HMMA_TENSOR_UTIL,
+		unit:    "%",
+	},
+	{
+		name:    "IMMA_TENSOR_UTIL",
+		outname: "nv_gpm_imma_tensor_util",
+		id:      nvml.GPM_METRIC_IMMA_TENSOR_UTIL,
+		unit:    "%",
+	},
+	{
+		name:    "DRAM_BW_UTIL",
+		outname: "nv_gpm_dram_bw_util",
+		id:      nvml.GPM_METRIC_DRAM_BW_UTIL,
+		unit:    "%",
+	},
+	{
+		name:    "FP64_UTIL",
+		outname: "nv_gpm_fp64_util",
+		id:      nvml.GPM_METRIC_FP64_UTIL,
+		unit:    "%",
+	},
+	{
+		name:    "FP32_UTIL",
+		outname: "nv_gpm_fp32_util",
+		id:      nvml.GPM_METRIC_FP32_UTIL,
+		unit:    "%",
+	},
+	{
+		name:    "FP16_UTIL",
+		outname: "nv_gpm_fp16_util",
+		id:      nvml.GPM_METRIC_FP16_UTIL,
+		unit:    "%",
+	},
+}
+
+type NvidiaGPMCollectorConfig struct {
+	Metrics               []string `json:"metrics,omitempty"`
+	ExcludeDevices        []string `json:"exclude_devices,omitempty"`
+	AddPciInfoTag         bool     `json:"add_pci_info_tag,omitempty"`
+	UsePciInfoAsTypeId    bool     `json:"use_pci_info_as_type_id,omitempty"`
+	AddUuidMeta           bool     `json:"add_uuid_meta,omitempty"`
+	AddBoardNumberMeta    bool     `json:"add_board_number_meta,omitempty"`
+	AddSerialMeta         bool     `json:"add_serial_meta,omitempty"`
+	ProcessMigDevices     bool     `json:"process_mig_devices,omitempty"`
+	UseUuidForMigDevices  bool     `json:"use_uuid_for_mig_device,omitempty"`
+	UseSliceForMigDevices bool     `json:"use_slice_for_mig_device,omitempty"`
+}
+
+type NvidiaGPMCollectorDevice struct {
+	device        nvml.Device
+	tags          map[string]string
+	meta          map[string]string
+	startTime     time.Time
+	endTime       time.Time
+	measurement   nvml.GpmMetricsGetType
+	metricsLookup map[int]NvidiaGPMMetricDef
+}
+
+type NvidiaGPMCollector struct {
+	metricCollector
+
+	config   NvidiaGPMCollectorConfig
+	gpus     []NvidiaGPMCollectorDevice
+	num_gpus int
+}
+
+func (m *NvidiaGPMCollector) Init(config json.RawMessage) error {
+	var err error = nil
+	m.name = "NvidiaGPMCollector"
+	m.parallel = true
+	if err := m.setup(); err != nil {
+		return fmt.Errorf("%s Init(): setup() call failed: %w", m.name, err)
+	}
+	if len(config) > 0 {
+		d := json.NewDecoder(strings.NewReader(string(config)))
+		d.DisallowUnknownFields()
+		if err = d.Decode(&m.config); err != nil {
+			return fmt.Errorf("%s Init(): Error decoding JSON config: %w", m.name, err)
+		}
+	}
+	m.meta = map[string]string{
+		"source": m.name,
+		"group":  "NvidiaGPM",
+	}
+
+	// Initialize NVIDIA Management Library (NVML)
+	ret := nvml.Init()
+
+	// Error: NVML library not found
+	// (nvml.ErrorString can not be used in this case)
+	if ret == nvml.ERROR_LIBRARY_NOT_FOUND {
+		return fmt.Errorf("%s Init(): NVML library not found", m.name)
+	}
+	if ret != nvml.SUCCESS {
+		err = errors.New(nvml.ErrorString(ret))
+		return fmt.Errorf("%s Init(): Unable to initialize NVML: %w", m.name, err)
+	}
+
+	// Number of NVIDIA GPUs
+	num_gpus, ret := nvml.DeviceGetCount()
+	if ret != nvml.SUCCESS {
+		err = errors.New(nvml.ErrorString(ret))
+		return fmt.Errorf("%s Init(): Unable to get device count: %w", m.name, err)
+	}
+
+	// For all GPUs
+	m.gpus = make([]NvidiaGPMCollectorDevice, 0, num_gpus)
+	for i := range num_gpus {
+
+		// Skip excluded devices by ID
+		str_i := strconv.Itoa(i)
+		if slices.Contains(m.config.ExcludeDevices, str_i) {
+			cclog.ComponentDebugf(m.name, "Skipping excluded device %s", str_i)
+			continue
+		}
+
+		// Get device handle
+		device, ret := nvml.DeviceGetHandleByIndex(i)
+		if ret != nvml.SUCCESS {
+			err = errors.New(nvml.ErrorString(ret))
+			cclog.ComponentErrorf(m.name, "Unable to get device at index %d: %s", i, err.Error())
+			continue
+		}
+
+		supportInfo, ret := nvml.GpmQueryDeviceSupport(device)
+		if ret != nvml.SUCCESS {
+			err = errors.New(nvml.ErrorString(ret))
+			cclog.ComponentErrorf(m.name, "Unable to query GPM support for device at index %d: %s", i, err.Error())
+			continue
+		} else {
+			if supportInfo.IsSupportedDevice == uint32(nvml.FEATURE_DISABLED) {
+				cclog.ComponentErrorf(m.name, "Device at index %d does not support GPM metrics", i)
+				continue
+			}
+		}
+
+		stream, ret := nvml.GpmQueryIfStreamingEnabled(device)
+		if ret != nvml.SUCCESS {
+			err = errors.New(nvml.ErrorString(ret))
+			cclog.ComponentErrorf(m.name, "Unable to query GPM streaming for device at index %d: %s", i, err.Error())
+			continue
+		} else {
+			if stream == uint32(nvml.FEATURE_DISABLED) {
+				ret = nvml.GpmSetStreamingEnabled(device, uint32(nvml.FEATURE_ENABLED))
+				if ret != nvml.SUCCESS {
+					err = errors.New(nvml.ErrorString(ret))
+					cclog.ComponentErrorf(m.name, "Unable to set streaming mode for device at index %d: %s", i, err.Error())
+				}
+			}
+		}
+
+		// Get device's PCI info
+		pciInfo, ret := nvml.DeviceGetPciInfo(device)
+		if ret != nvml.SUCCESS {
+			err = errors.New(nvml.ErrorString(ret))
+			cclog.ComponentErrorf(m.name, "Unable to get PCI info for device at index %d: %s", i, err.Error())
+			continue
+		}
+		// Create PCI ID in the common format used by the NVML.
+		pci_id := fmt.Sprintf(
+			nvml.DEVICE_PCI_BUS_ID_FMT,
+			pciInfo.Domain,
+			pciInfo.Bus,
+			pciInfo.Device)
+
+		// Skip excluded devices specified by PCI ID
+		if slices.Contains(m.config.ExcludeDevices, pci_id) {
+			cclog.ComponentDebugf(m.name, "Skipping excluded device %s", pci_id)
+			continue
+		}
+		ss, nvmlErr := nvml.GpmSampleAlloc()
+		if nvmlErr != nvml.SUCCESS {
+			err = errors.New(nvml.ErrorString(ret))
+			cclog.ComponentErrorf(m.name, "Failed to allocate GPM sample for device %d: %s", i, err.Error())
+			continue
+		}
+		es, nvmlErr := nvml.GpmSampleAlloc()
+		if nvmlErr != nvml.SUCCESS {
+			err = errors.New(nvml.ErrorString(ret))
+			cclog.ComponentErrorf(m.name, "Failed to allocate GPM sample for device %d: %s", i, err.Error())
+			continue
+		}
+
+		// Select which value to use as 'type-id'.
+		// The PCI ID is commonly required in SLURM environments because the
+		// numberic IDs used by SLURM and the ones used by NVML might differ
+		// depending on the job type. The PCI ID is more reliable but is commonly
+		// not recorded for a job, so it must be added manually in prologue or epilogue
+		// e.g. to the comment field
+		tid := str_i
+		if m.config.UsePciInfoAsTypeId {
+			tid = pci_id
+		}
+
+		// Now we got all infos together, populate the device list
+		g := NvidiaGPMCollectorDevice{}
+
+		// Add device handle
+		g.device = device
+
+		// Add tags
+		g.tags = map[string]string{
+			"type":    "accelerator",
+			"type-id": tid,
+		}
+
+		// Add PCI info as tag if not already used as 'type-id'
+		if m.config.AddPciInfoTag && !m.config.UsePciInfoAsTypeId {
+			g.tags["pci_identifier"] = pci_id
+		}
+
+		g.meta = map[string]string{
+			"source": m.name,
+			"group":  "Nvidia",
+		}
+
+		if m.config.AddBoardNumberMeta {
+			board, ret := nvml.DeviceGetBoardPartNumber(device)
+			if ret != nvml.SUCCESS {
+				err = errors.New(nvml.ErrorString(ret))
+				cclog.ComponentError(m.name, "Unable to get boart part number for device at index", i, ":", err.Error())
+			} else {
+				g.meta["board_number"] = board
+			}
+		}
+		if m.config.AddSerialMeta {
+			serial, ret := nvml.DeviceGetSerial(device)
+			if ret != nvml.SUCCESS {
+				err = errors.New(nvml.ErrorString(ret))
+				cclog.ComponentError(m.name, "Unable to get serial number for device at index", i, ":", err.Error())
+			} else {
+				g.meta["serial"] = serial
+			}
+		}
+		if m.config.AddUuidMeta {
+			uuid, ret := nvml.DeviceGetUUID(device)
+			if ret != nvml.SUCCESS {
+				err = errors.New(nvml.ErrorString(ret))
+				cclog.ComponentError(m.name, "Unable to get UUID for device at index", i, ":", err.Error())
+			} else {
+				g.meta["uuid"] = uuid
+			}
+		}
+
+		g.measurement.Sample1 = ss
+		g.measurement.Sample2 = es
+		g.measurement.Version = nvml.GPM_METRICS_GET_VERSION
+		g.metricsLookup = make(map[int]NvidiaGPMMetricDef)
+		metIdx := 0
+		for _, inmetric := range m.config.Metrics {
+			for _, defmetric := range NvidiaGPMMetrics {
+				if inmetric == defmetric.outname || inmetric == defmetric.name {
+					g.measurement.Metrics[metIdx] = nvml.GpmMetric{
+						MetricId: uint32(defmetric.id),
+					}
+					g.metricsLookup[metIdx] = defmetric
+					metIdx += 1
+				}
+			}
+		}
+		g.measurement.NumMetrics = uint32(metIdx)
+		m.gpus = append(m.gpus, g)
+	}
+	cclog.ComponentDebugf(m.name, "Found %d Nvidia GPUs with GPM support", len(m.gpus))
+	m.num_gpus = len(m.gpus)
+	m.init = true
+	return err
+}
+
+func (m *NvidiaGPMCollector) Read(interval time.Duration, output chan lp.CCMessage) {
+	var err error
+	if !m.init {
+		return
+	}
+	for i, gpu := range m.gpus {
+		gpu.startTime = time.Now()
+		nvmlErr := gpu.measurement.Sample1.Get(gpu.device)
+		if nvmlErr != nvml.SUCCESS {
+			err = errors.New(nvml.ErrorString(nvmlErr))
+			cclog.ComponentError(m.name, "Unable to get start GPM sample for device at index", i, ":", err.Error())
+			continue
+		}
+	}
+	time.Sleep(interval)
+
+	for i, gpu := range m.gpus {
+		gpu.endTime = time.Now()
+		nvmlErr := gpu.measurement.Sample2.Get(gpu.device)
+		if nvmlErr != nvml.SUCCESS {
+			err = errors.New(nvml.ErrorString(nvmlErr))
+			cclog.ComponentError(m.name, "Unable to get stop GPM sample for device at index", i, ":", err.Error())
+			continue
+		}
+	}
+
+	for i, gpu := range m.gpus {
+		nvmlErr := nvml.GpmMetricsGet(&gpu.measurement)
+		if nvmlErr != nvml.SUCCESS {
+			err = errors.New(nvml.ErrorString(nvmlErr))
+			cclog.ComponentError(m.name, "Unable to get evaluate GPM sample for device at index", i, ":", err.Error())
+			continue
+		}
+		for idx, metricDef := range gpu.metricsLookup {
+			y, err := lp.NewMetric(metricDef.outname, gpu.tags, gpu.meta, gpu.measurement.Metrics[idx].Value, time.Now())
+			if err == nil {
+				y.AddMeta("unit", metricDef.unit)
+				output <- y
+			}
+		}
+	}
+
+}
+
+func (m *NvidiaGPMCollector) Close() {
+	if m.init {
+		for i, gpu := range m.gpus {
+			ret := gpu.measurement.Sample1.Free()
+			if ret != nvml.SUCCESS {
+				err := errors.New(nvml.ErrorString(ret))
+				cclog.ComponentErrorf(m.name, "Unable to free start sample for device at index %d: %s", i, err.Error())
+			}
+			ret = gpu.measurement.Sample2.Free()
+			if ret != nvml.SUCCESS {
+				err := errors.New(nvml.ErrorString(ret))
+				cclog.ComponentErrorf(m.name, "Unable to free stop sample for device at index %d: %s", i, err.Error())
+			}
+		}
+		if ret := nvml.Shutdown(); ret != nvml.SUCCESS {
+			cclog.ComponentError(m.name, "nvml.Shutdown() not successful")
+		}
+		m.init = false
+	}
+}
--- a/collectors/nvidiaGPMMetric.md
+++ b/collectors/nvidiaGPMMetric.md
@@ -0,0 +1,54 @@
+<!--
+---
+title: "Nvidia NVML GPM metric collector"
+description: Collect metrics for Nvidia GPUs using the NVML GPM interface
+categories: [cc-metric-collector]
+tags: ['Admin']
+weight: 2
+hugo_path: docs/reference/cc-metric-collector/collectors/nvidiaGPM.md
+---
+-->
+
+## `nvidiaGPM` collector
+
+```json
+  "nvidia_gpm": {
+    "metrics": [
+      "nv_fb_mem_used",
+      "nv_fan"
+    ],
+    "exclude_devices": [
+      "0","1", "0000000:ff:01.0"
+    ],
+    
+    "process_mig_devices": false,
+    "use_pci_info_as_type_id": true,
+    "add_pci_info_tag": false,
+    "add_uuid_meta": false,
+    "add_board_number_meta": false,
+    "add_serial_meta": false,
+    "use_uuid_for_mig_device": false,
+    "use_slice_for_mig_device": false
+  }
+```
+
+The `nvidia_gpm` collector can be configured to leave out specific devices with the `exclude_devices` option. It takes IDs as supplied to the NVML with `nvmlDeviceGetHandleByIndex()` or the PCI address in NVML format (`%08X:%02X:%02X.0`). Commonly only the physical GPUs are monitored. If MIG devices should be analyzed as well, set `process_mig_devices` (adds `stype=mig,stype-id=<mig_index>`). With the options `use_uuid_for_mig_device` and `use_slice_for_mig_device`, the `<mig_index>` can be replaced with the UUID (e.g. `MIG-6a9f7cc8-6d5b-5ce0-92de-750edc4d8849`) or the MIG slice name (e.g. `1g.5gb`).
+
+The metrics sent by the `nvidia_gpm` collector use `accelerator` as `type` tag. For the `type-id`, it uses the device handle index by default. With the `use_pci_info_as_type_id` option, the PCI ID is used instead. If both values should be added as tags, activate the `add_pci_info_tag` option. It uses the device handle index as `type-id` and adds the PCI ID as separate `pci_identifier` tag.
+
+Optionally, it is possible to add the UUID, the board part number and the serial to the meta informations. They are not sent to the sinks (if not configured otherwise).
+
+
+Available Metrics:
+* `nv_gpm_graphics_util`
+* `nv_gpm_sm_util`
+* `nv_gpm_sm_occupancy`
+* `nv_gpm_integer_util`
+* `nv_gpm_any_tensor_util`
+* `nv_gpm_dfma_tensor_util`
+* `nv_gpm_hmma_tensor_util`
+* `nv_gpm_imma_tensor_util`
+* `nv_gpm_dram_bw_util`
+* `nv_gpm_fp64_util`
+* `nv_gpm_fp32_util`
+* `nv_gpm_fp16_util`
--- a/collectors/nvidiaMetric.go
+++ b/collectors/nvidiaMetric.go
@@ -113,7 +113,7 @@ func (m *NvidiaCollector) Init(config json.RawMessage) error {
 		// Skip excluded devices by ID
 		str_i := strconv.Itoa(i)
 		if slices.Contains(m.config.ExcludeDevices, str_i) {
-			cclog.ComponentDebug(m.name, "Skipping excluded device", str_i)
+			cclog.ComponentDebugf(m.name, "Skipping excluded device %s", str_i)
 			continue
 		}

@@ -121,7 +121,7 @@ func (m *NvidiaCollector) Init(config json.RawMessage) error {
 		device, ret := nvml.DeviceGetHandleByIndex(i)
 		if ret != nvml.SUCCESS {
 			err = errors.New(nvml.ErrorString(ret))
-			cclog.ComponentError(m.name, "Unable to get device at index", i, ":", err.Error())
+			cclog.ComponentErrorf(m.name, "Unable to get device at index %d: %s", i, err.Error())
 			continue
 		}

@@ -129,7 +129,7 @@ func (m *NvidiaCollector) Init(config json.RawMessage) error {
 		pciInfo, ret := nvml.DeviceGetPciInfo(device)
 		if ret != nvml.SUCCESS {
 			err = errors.New(nvml.ErrorString(ret))
-			cclog.ComponentError(m.name, "Unable to get PCI info for device at index", i, ":", err.Error())
+			cclog.ComponentErrorf(m.name, "Unable to get PCI info for device at index %d: %s", i, err.Error())
 			continue
 		}
 		// Create PCI ID in the common format used by the NVML.
@@ -141,7 +141,7 @@ func (m *NvidiaCollector) Init(config json.RawMessage) error {

 		// Skip excluded devices specified by PCI ID
 		if slices.Contains(m.config.ExcludeDevices, pci_id) {
-			cclog.ComponentDebug(m.name, "Skipping excluded device", pci_id)
+			cclog.ComponentDebugf(m.name, "Skipping excluded device %s", pci_id)
 			continue
 		}

@@ -183,7 +183,7 @@ func (m *NvidiaCollector) Init(config json.RawMessage) error {
 		if m.config.AddBoardNumberMeta {
 			board, ret := nvml.DeviceGetBoardPartNumber(device)
 			if ret != nvml.SUCCESS {
-				cclog.ComponentError(m.name, "Unable to get boart part number for device at index", i, ":", err.Error())
+				cclog.ComponentErrorf(m.name, "Unable to get boart part number for device at index %d: %s", i, err.Error())
 			} else {
 				g.meta["board_number"] = board
 			}
@@ -191,7 +191,7 @@ func (m *NvidiaCollector) Init(config json.RawMessage) error {
 		if m.config.AddSerialMeta {
 			serial, ret := nvml.DeviceGetSerial(device)
 			if ret != nvml.SUCCESS {
-				cclog.ComponentError(m.name, "Unable to get serial number for device at index", i, ":", err.Error())
+				cclog.ComponentErrorf(m.name, "Unable to get serial number for device at index %d: %s", i, err.Error())
 			} else {
 				g.meta["serial"] = serial
 			}
@@ -199,7 +199,7 @@ func (m *NvidiaCollector) Init(config json.RawMessage) error {
 		if m.config.AddUuidMeta {
 			uuid, ret := nvml.DeviceGetUUID(device)
 			if ret != nvml.SUCCESS {
-				cclog.ComponentError(m.name, "Unable to get UUID for device at index", i, ":", err.Error())
+				cclog.ComponentErrorf(m.name, "Unable to get UUID for device at index %d: %s", i, err.Error())
 			} else {
 				g.meta["uuid"] = uuid
 			}
@@ -1115,31 +1115,6 @@ func readNVLinkStats(device *NvidiaCollectorDevice, output chan lp.CCMessage) er
 	return nil
 }

-func readEfficiency(device *NvidiaCollectorDevice, output chan lp.CCMessage) error {
-	if !device.excludeMetrics["nv_util_eff"] {
-		maxPower, ret := nvml.DeviceGetEnforcedPowerLimit(device.device)
-		if ret == nvml.SUCCESS {
-			curPower, ret := nvml.DeviceGetPowerUsage(device.device)
-			if ret == nvml.SUCCESS {
-				util, ret := nvml.DeviceGetUtilizationRates(device.device)
-				if ret == nvml.SUCCESS {
-					factor := float64(curPower) / float64(maxPower)
-					eff := uint32(float64(util.Gpu) * factor)
-					if eff > 100 {
-						eff = 100
-					}
-					y, err := lp.NewMetric("nv_util_eff", device.tags, device.meta, eff, time.Now())
-					if err == nil {
-						y.AddTag("unit", "percent")
-						output <- y
-					}
-				}
-			}
-		}
-	}
-	return nil
-}
-
 func (m *NvidiaCollector) Read(interval time.Duration, output chan lp.CCMessage) {
 	var err error
 	if !m.init {
@@ -1153,102 +1128,97 @@ func (m *NvidiaCollector) Read(interval time.Duration, output chan lp.CCMessage)
 		}
 		err = readMemoryInfo(device, output)
 		if err != nil {
-			cclog.ComponentDebug(m.name, "readMemoryInfo for device", name, "failed")
+			cclog.ComponentDebugf(m.name, "readMemoryInfo for device %s failed", name)
 		}

 		err = readUtilization(device, output)
 		if err != nil {
-			cclog.ComponentDebug(m.name, "readUtilization for device", name, "failed")
+			cclog.ComponentDebugf(m.name, "readUtilization for device %s failed", name)
 		}

 		err = readTemp(device, output)
 		if err != nil {
-			cclog.ComponentDebug(m.name, "readTemp for device", name, "failed")
+			cclog.ComponentDebugf(m.name, "readTemp for device %s failed", name)
 		}

 		err = readFan(device, output)
 		if err != nil {
-			cclog.ComponentDebug(m.name, "readFan for device", name, "failed")
+			cclog.ComponentDebugf(m.name, "readFan for device %s failed", name)
 		}

 		err = readEccMode(device, output)
 		if err != nil {
-			cclog.ComponentDebug(m.name, "readEccMode for device", name, "failed")
+			cclog.ComponentDebugf(m.name, "readEccMode for device %s failed", name)
 		}

 		err = readPerfState(device, output)
 		if err != nil {
-			cclog.ComponentDebug(m.name, "readPerfState for device", name, "failed")
+			cclog.ComponentDebugf(m.name, "readPerfState for device %s failed", name)
 		}

 		err = readPowerUsage(device, output)
 		if err != nil {
-			cclog.ComponentDebug(m.name, "readPowerUsage for device", name, "failed")
+			cclog.ComponentDebugf(m.name, "readPowerUsage for device %s failed", name)
 		}

 		err = readEnergyConsumption(device, output)
 		if err != nil {
-			cclog.ComponentDebug(m.name, "readEnergyConsumption for device", name, "failed")
+			cclog.ComponentDebugf(m.name, "readEnergyConsumption for device %s failed", name)
 		}

 		err = readClocks(device, output)
 		if err != nil {
-			cclog.ComponentDebug(m.name, "readClocks for device", name, "failed")
+			cclog.ComponentDebugf(m.name, "readClocks for device %s failed", name)
 		}

 		err = readMaxClocks(device, output)
 		if err != nil {
-			cclog.ComponentDebug(m.name, "readMaxClocks for device", name, "failed")
+			cclog.ComponentDebugf(m.name, "readMaxClocks for device %s failed", name)
 		}

 		err = readEccErrors(device, output)
 		if err != nil {
-			cclog.ComponentDebug(m.name, "readEccErrors for device", name, "failed")
+			cclog.ComponentDebugf(m.name, "readEccErrors for device %s failed", name)
 		}

 		err = readPowerLimit(device, output)
 		if err != nil {
-			cclog.ComponentDebug(m.name, "readPowerLimit for device", name, "failed")
+			cclog.ComponentDebugf(m.name, "readPowerLimit for device %s failed", name)
 		}

 		err = readEncUtilization(device, output)
 		if err != nil {
-			cclog.ComponentDebug(m.name, "readEncUtilization for device", name, "failed")
+			cclog.ComponentDebugf(m.name, "readEncUtilization for device %s failed", name)
 		}

 		err = readDecUtilization(device, output)
 		if err != nil {
-			cclog.ComponentDebug(m.name, "readDecUtilization for device", name, "failed")
+			cclog.ComponentDebugf(m.name, "readDecUtilization for device %s failed", name)
 		}

 		err = readRemappedRows(device, output)
 		if err != nil {
-			cclog.ComponentDebug(m.name, "readRemappedRows for device", name, "failed")
+			cclog.ComponentDebugf(m.name, "readRemappedRows for device %s failed", name)
 		}

 		err = readBarMemoryInfo(device, output)
 		if err != nil {
-			cclog.ComponentDebug(m.name, "readBarMemoryInfo for device", name, "failed")
+			cclog.ComponentDebugf(m.name, "readBarMemoryInfo for device %s failed", name)
 		}

 		err = readProcessCounts(device, output)
 		if err != nil {
-			cclog.ComponentDebug(m.name, "readProcessCounts for device", name, "failed")
+			cclog.ComponentDebugf(m.name, "readProcessCounts for device %s failed", name)
 		}

 		err = readViolationStats(device, output)
 		if err != nil {
-			cclog.ComponentDebug(m.name, "readViolationStats for device", name, "failed")
+			cclog.ComponentDebugf(m.name, "readViolationStats for device %s failed", name)
 		}

 		err = readNVLinkStats(device, output)
 		if err != nil {
-			cclog.ComponentDebug(m.name, "readNVLinkStats for device", name, "failed")
-		}
-
-		err = readEfficiency(device, output)
-		if err != nil {
-			cclog.ComponentDebug(m.name, "readEfficiency for device", name, "failed")
+			cclog.ComponentDebugf(m.name, "readNVLinkStats for device %s failed", name)
 		}
 	}

@@ -1274,7 +1244,7 @@ func (m *NvidiaCollector) Read(interval time.Duration, output chan lp.CCMessage)
 			if maxMig == 0 {
 				continue
 			}
-			cclog.ComponentDebug(m.name, "Reading MIG devices for GPU", i)
+			cclog.ComponentDebugf(m.name, "Reading MIG devices for GPU %d", i)

 			for j := range maxMig {
 				mdev, ret := nvml.DeviceGetMigDeviceHandleByIndex(m.gpus[i].device, j)
@@ -1298,7 +1268,7 @@ func (m *NvidiaCollector) Read(interval time.Duration, output chan lp.CCMessage)
 				if m.config.UseUuidForMigDevices {
 					uuid, ret := nvml.DeviceGetUUID(mdev)
 					if ret != nvml.SUCCESS {
-						cclog.ComponentError(m.name, "Unable to get UUID for mig device at index", j, ":", err.Error())
+						cclog.ComponentErrorf(m.name, "Unable to get UUID for mig device at index %d: %s", j, err.Error())
 					} else {
 						migDevice.tags["stype-id"] = uuid
 					}
--- a/collectors/nvidiaMetric.md
+++ b/collectors/nvidiaMetric.md
@@ -85,6 +85,5 @@ Metrics:
 * `nv_energy`
 * `nv_energy_abs`
 * `nv_average_power`
-* `nv_util_eff` (`nv_util` * (`nv_power_usage` / `nv_power_max_limit`))

 Some metrics add the additional sub type tag (`stype`) like the `nv_nvlink_*` metrics set `stype=nvlink,stype-id=<link_number>`. 
--- a/collectors/raplMetric.go
+++ b/collectors/raplMetric.go
@@ -208,11 +208,10 @@ func (m *RAPLCollector) Init(config json.RawMessage) error {
 	}

 	// Initialized
-	cclog.ComponentDebug(
+	cclog.ComponentDebugf(
 		m.name,
-		"initialized",
-		len(m.RAPLZoneInfo),
-		"zones with running average power limit (RAPL) monitoring attributes")
+		"initialized %d zones with running average power limit (RAPL) monitoring attributes",
+		len(m.RAPLZoneInfo))
 	m.init = true

 	return err
--- a/collectors/rocmsmiMetric.go
+++ b/collectors/rocmsmiMetric.go
@@ -124,7 +124,7 @@ func (m *RocmSmiCollector) Init(config json.RawMessage) error {
 		if m.config.AddSerialMeta {
 			serial, ret := rocm_smi.DeviceGetSerialNumber(device)
 			if ret != rocm_smi.STATUS_SUCCESS {
-				cclog.ComponentError(m.name, "Unable to get serial number for device at index", i, ":", rocm_smi.StatusStringNoError(ret))
+				cclog.ComponentErrorf(m.name, "Unable to get serial number for device at index %d: %s", i, rocm_smi.StatusStringNoError(ret))
 			} else {
 				dev.meta["serial"] = serial
 			}
@@ -152,7 +152,7 @@ func (m *RocmSmiCollector) Read(interval time.Duration, output chan lp.CCMessage
 	for _, dev := range m.devices {
 		metrics, ret := rocm_smi.DeviceGetMetrics(dev.device)
 		if ret != rocm_smi.STATUS_SUCCESS {
-			cclog.ComponentError(m.name, "Unable to get metrics for device at index", dev.index, ":", rocm_smi.StatusStringNoError(ret))
+			cclog.ComponentErrorf(m.name, "Unable to get metrics for device at index %d: %s", dev.index, rocm_smi.StatusStringNoError(ret))
 			continue
 		}

--- a/collectors/schedstatMetric.go
+++ b/collectors/schedstatMetric.go
@@ -147,15 +147,15 @@ func (m *SchedstatCollector) Read(interval time.Duration, output chan lp.CCMessa

 	file, err := os.Open(SCHEDSTATFILE)
 	if err != nil {
-		cclog.ComponentError(
+		cclog.ComponentErrorf(
 			m.name,
-			fmt.Sprintf("Read(): Failed to open file '%s': %v", SCHEDSTATFILE, err))
+			"Read(): Failed to open file '%s': %v", SCHEDSTATFILE, err)
 	}
 	defer func() {
 		if err := file.Close(); err != nil {
-			cclog.ComponentError(
+			cclog.ComponentErrorf(
 				m.name,
-				fmt.Sprintf("Read(): Failed to close file '%s': %v", SCHEDSTATFILE, err))
+				"Read(): Failed to close file '%s': %v", SCHEDSTATFILE, err)
 		}
 	}()

--- a/collectors/slurmCgroupMetric.go
+++ b/collectors/slurmCgroupMetric.go
@@ -240,7 +240,7 @@ func (m *SlurmCgroupCollector) Read(interval time.Duration, output chan lp.CCMes
 	globPattern := filepath.Join(m.cgroupBase, "job_*")
 	jobDirs, err := filepath.Glob(globPattern)
 	if err != nil {
-		cclog.ComponentError(m.name, "Error globbing job directories:", err.Error())
+		cclog.ComponentErrorf(m.name, "Error globbing job directories: %s", err.Error())
 		return
 	}

@@ -249,7 +249,7 @@ func (m *SlurmCgroupCollector) Read(interval time.Duration, output chan lp.CCMes

 		jobdata, err := m.ReadJobData(jKey)
 		if err != nil {
-			cclog.ComponentError(m.name, "Error reading job data for", jKey, ":", err.Error())
+			cclog.ComponentError(m.name, "Error reading job data for %s: %s", jKey, err.Error())
 			continue
 		}

--- a/collectors/smartmonMetric.go
+++ b/collectors/smartmonMetric.go
@@ -228,12 +228,12 @@ func (m *SmartMonCollector) Read(interval time.Duration, output chan lp.CCMessag

 		stdout, err := command.Output()
 		if err != nil {
-			cclog.ComponentError(m.name, "cannot read data for device", d.Name)
+			cclog.ComponentErrorf(m.name, "cannot read data for device %s", d.Name)
 			continue
 		}
 		err = json.Unmarshal(stdout, &data)
 		if err != nil {
-			cclog.ComponentError(m.name, "cannot unmarshal data for device", d.Name)
+			cclog.ComponentErrorf(m.name, "cannot unmarshal data for device %s", d.Name)
 			continue
 		}
 		if !m.excludeMetric.temp {
--- a/collectors/tempMetric.go
+++ b/collectors/tempMetric.go
@@ -188,16 +188,16 @@ func (m *TempCollector) Read(interval time.Duration, output chan lp.CCMessage) {
 		// Read sensor file
 		buffer, err := os.ReadFile(sensor.file)
 		if err != nil {
-			cclog.ComponentError(
+			cclog.ComponentErrorf(
 				m.name,
-				fmt.Sprintf("Read(): Failed to read file '%s': %v", sensor.file, err))
+				"Read(): Failed to read file '%s': %v", sensor.file, err)
 			continue
 		}
 		x, err := strconv.ParseInt(strings.TrimSpace(string(buffer)), 10, 64)
 		if err != nil {
-			cclog.ComponentError(
+			cclog.ComponentErrorf(
 				m.name,
-				fmt.Sprintf("Read(): Failed to convert temperature '%s' to int64: %v", buffer, err))
+				"Read(): Failed to convert temperature '%s' to int64: %v", buffer, err)
 			continue
 		}
 		x /= 1000
--- a/collectors/topprocsMetric.go
+++ b/collectors/topprocsMetric.go
@@ -77,9 +77,9 @@ func (m *TopProcsCollector) Read(interval time.Duration, output chan lp.CCMessag
 	command := exec.Command("ps", "-Ao", "comm", "--sort=-pcpu")
 	stdout, err := command.Output()
 	if err != nil {
-		cclog.ComponentError(
+		cclog.ComponentErrorf(
 			m.name,
-			fmt.Sprintf("Read(): Failed to read output from command \"%s\": %v", command.String(), err))
+			"Read(): Failed to read output from command \"%s\": %v", command.String(), err)
 		return
 	}

--- a/go.mod
+++ b/go.mod
@@ -7,10 +7,10 @@ require (
 	github.com/ClusterCockpit/go-rocm-smi v0.4.0
 	github.com/NVIDIA/go-nvml v0.13.0-1
 	github.com/PaesslerAG/gval v1.2.4
-	github.com/fsnotify/fsnotify v1.10.0
-	github.com/tklauser/go-sysconf v0.3.16
-	golang.design/x/thread v0.0.0-20210122121316-335e9adffdf1
-	golang.org/x/sys v0.43.0
+	github.com/fsnotify/fsnotify v1.10.1
+	github.com/tklauser/go-sysconf v0.4.0
+	golang.design/x/thread v0.3.2
+	golang.org/x/sys v0.45.0
 )

 require (
@@ -38,7 +38,7 @@ require (
 	github.com/santhosh-tekuri/jsonschema/v5 v5.3.1 // indirect
 	github.com/shopspring/decimal v1.4.0 // indirect
 	github.com/stmcginnis/gofish v0.21.6 // indirect
-	github.com/tklauser/numcpus v0.11.0 // indirect
+	github.com/tklauser/numcpus v0.12.0 // indirect
 	go.yaml.in/yaml/v2 v2.4.4 // indirect
 	golang.org/x/crypto v0.50.0 // indirect
 	golang.org/x/net v0.53.0 // indirect
--- a/go.sum
+++ b/go.sum
@@ -53,8 +53,8 @@ github.com/expr-lang/expr v1.17.8 h1:W1loDTT+0PQf5YteHSTpju2qfUfNoBt4yw9+wOEU9VM
 github.com/expr-lang/expr v1.17.8/go.mod h1:8/vRC7+7HBzESEqt5kKpYXxrxkr31SaO8r40VO/1IT4=
 github.com/frankban/quicktest v1.13.0 h1:yNZif1OkDfNoDfb9zZa9aXIpejNR4F23Wely0c+Qdqk=
 github.com/frankban/quicktest v1.13.0/go.mod h1:qLE0fzW0VuyUAJgPU19zByoIr0HtCHN/r/VLSOOIySU=
-github.com/fsnotify/fsnotify v1.10.0 h1:Xx/5Ydg9CeBDX/wi4VJqStNtohYjitZhhlHt4h3St1M=
-github.com/fsnotify/fsnotify v1.10.0/go.mod h1:TLheqan6HD6GBK6PrDWyDPBaEV8LspOxvPSjC+bVfgo=
+github.com/fsnotify/fsnotify v1.10.1 h1:b0/UzAf9yR5rhf3RPm9gf3ehBPpf0oZKIjtpKrx59Ho=
+github.com/fsnotify/fsnotify v1.10.1/go.mod h1:TLheqan6HD6GBK6PrDWyDPBaEV8LspOxvPSjC+bVfgo=
 github.com/go-ole/go-ole v1.3.0 h1:Dt6ye7+vXGIKZ7Xtk4s6/xVdGDQynvom7xCFEdWr6uE=
 github.com/go-ole/go-ole v1.3.0/go.mod h1:5LS6F96DhAwUc7C+1HLexzMXY1xGRSryjyPPKW6zv78=
 github.com/gogo/protobuf v1.3.2 h1:Ov1cvc58UF3b5XjBnZv7+opcTcQFZebYjWzi34vdm4Q=
@@ -163,18 +163,18 @@ github.com/stretchr/testify v1.11.1 h1:7s2iGBzp5EwR7/aIZr8ao5+dra3wiQyKjjFuvgVKu
 github.com/stretchr/testify v1.11.1/go.mod h1:wZwfW3scLgRK+23gO65QZefKpKQRnfz6sD981Nm4B6U=
 github.com/testcontainers/testcontainers-go v0.26.0 h1:uqcYdoOHBy1ca7gKODfBd9uTHVK3a7UL848z09MVZ0c=
 github.com/testcontainers/testcontainers-go v0.26.0/go.mod h1:ICriE9bLX5CLxL9OFQ2N+2N+f+803LNJ1utJb1+Inx0=
-github.com/tklauser/go-sysconf v0.3.16 h1:frioLaCQSsF5Cy1jgRBrzr6t502KIIwQ0MArYICU0nA=
-github.com/tklauser/go-sysconf v0.3.16/go.mod h1:/qNL9xxDhc7tx3HSRsLWNnuzbVfh3e7gh/BmM179nYI=
-github.com/tklauser/numcpus v0.11.0 h1:nSTwhKH5e1dMNsCdVBukSZrURJRoHbSEQjdEbY+9RXw=
-github.com/tklauser/numcpus v0.11.0/go.mod h1:z+LwcLq54uWZTX0u/bGobaV34u6V7KNlTZejzM6/3MQ=
+github.com/tklauser/go-sysconf v0.4.0 h1:7H0uAN+7RkwWRaxhYXDLqa5V3LPrJeV8wmD9dRUgPQU=
+github.com/tklauser/go-sysconf v0.4.0/go.mod h1:8mTNWyog7H+MpKijp4VmKJAd2bbYQ2zuUwkYRbUArPI=
+github.com/tklauser/numcpus v0.12.0 h1:NR85qdvHA9pFse3x3weVZ0r0ST8R6l5RHbZrlRaqob4=
+github.com/tklauser/numcpus v0.12.0/go.mod h1:ABHeXzJnr/qqwguhClkZKT1/8VABcYrsyUiUGobwWJg=
 github.com/yusufpapurcu/wmi v1.2.3 h1:E1ctvB7uKFMOJw3fdOW32DwGE9I7t++CRUEMKvFoFiw=
 github.com/yusufpapurcu/wmi v1.2.3/go.mod h1:SBZ9tNy3G9/m5Oi98Zks0QjeHVDvuK0qfxQmPyzfmi0=
 go.uber.org/goleak v1.3.0 h1:2K3zAYmnTNqV73imy9J1T3WC+gmCePx2hEGkimedGto=
 go.uber.org/goleak v1.3.0/go.mod h1:CoHD4mav9JJNrW/WLlf7HGZPjdw8EucARQHekz1X6bE=
 go.yaml.in/yaml/v2 v2.4.4 h1:tuyd0P+2Ont/d6e2rl3be67goVK4R6deVxCUX5vyPaQ=
 go.yaml.in/yaml/v2 v2.4.4/go.mod h1:gMZqIpDtDqOfM0uNfy0SkpRhvUryYH0Z6wdMYcacYXQ=
-golang.design/x/thread v0.0.0-20210122121316-335e9adffdf1 h1:P7S/GeHBAFEZIYp0ePPs2kHXoazz8q2KsyxHyQVGCJg=
-golang.design/x/thread v0.0.0-20210122121316-335e9adffdf1/go.mod h1:9CWpnTUmlQkfdpdutA1nNf4iE5lAVt3QZOu0Z6hahBE=
+golang.design/x/thread v0.3.2 h1:FmD1glspGrQCe6FuQLmSrT6wz2CSzq7vKVDluyiMnqo=
+golang.design/x/thread v0.3.2/go.mod h1:6+Hi2rMOgMHZdKDWaqNHyWtoFUx1HxZ06LfHPh5Z/hQ=
 golang.org/x/crypto v0.50.0 h1:zO47/JPrL6vsNkINmLoo/PH1gcxpls50DNogFvB5ZGI=
 golang.org/x/crypto v0.50.0/go.mod h1:3muZ7vA7PBCE6xgPX7nkzzjiUq87kRItoJQM1Yo8S+Q=
 golang.org/x/exp v0.0.0-20231005195138-3e424a577f31 h1:9k5exFQKQglLo+RoP+4zMjOFE14P6+vyR0baDAi0Rcs=
@@ -183,9 +183,8 @@ golang.org/x/mod v0.13.0 h1:I/DsJXRlw/8l/0c24sM9yb0T4z9liZTduXvdAWYiysY=
 golang.org/x/mod v0.13.0/go.mod h1:hTbmBsO62+eylJbnUtE2MGJUyE7QWk4xUqPFrRgJ+7c=
 golang.org/x/net v0.53.0 h1:d+qAbo5L0orcWAr0a9JweQpjXF19LMXJE8Ey7hwOdUA=
 golang.org/x/net v0.53.0/go.mod h1:JvMuJH7rrdiCfbeHoo3fCQU24Lf5JJwT9W3sJFulfgs=
-golang.org/x/sys v0.0.0-20210122093101-04d7465088b8/go.mod h1:h1NjWce9XRLGQEsW7wpKNCjG9DtNlClVuFLEZdDNbEs=
-golang.org/x/sys v0.43.0 h1:Rlag2XtaFTxp19wS8MXlJwTvoh8ArU6ezoyFsMyCTNI=
-golang.org/x/sys v0.43.0/go.mod h1:4GL1E5IUh+htKOUEOaiffhrAeqysfVGipDYzABqnCmw=
+golang.org/x/sys v0.45.0 h1:dO4czNzziLiiXplLQgBCEpCvXQ3dnkn0SdaZSYdQ+FY=
+golang.org/x/sys v0.45.0/go.mod h1:4GL1E5IUh+htKOUEOaiffhrAeqysfVGipDYzABqnCmw=
 golang.org/x/time v0.15.0 h1:bbrp8t3bGUeFOx08pvsMYRTCVSMk89u4tKbNOZbp88U=
 golang.org/x/time v0.15.0/go.mod h1:Y4YMaQmXwGQZoFaVFk4YpCt4FLQMYKZe9oeV/f4MSno=
 golang.org/x/tools v0.14.0 h1:jvNa2pY0M4r62jkRQ6RwEZZyPcymeL9XZMLBbV7U2nc=
--- a/internal/metricRouter/metricRouter.go
+++ b/internal/metricRouter/metricRouter.go
@@ -35,18 +35,18 @@ type metricRouterTagConfig struct {

 // Metric router configuration
 type metricRouterConfig struct {
-	HostnameTagName   string                               `json:"hostname_tag"`        // Key name used when adding the hostname to a metric (default 'hostname')
-	AddTags           []metricRouterTagConfig              `json:"add_tags"`            // List of tags that are added when the condition is met
-	DelTags           []metricRouterTagConfig              `json:"delete_tags"`         // List of tags that are removed when the condition is met
-	IntervalAgg       []agg.MetricAggregatorIntervalConfig `json:"interval_aggregates"` // List of aggregation function processed at the end of an interval
-	DropMetrics       []string                             `json:"drop_metrics"`        // List of metric names to drop. For fine-grained dropping use drop_metrics_if
-	DropMetricsIf     []string                             `json:"drop_metrics_if"`     // List of evaluatable terms to drop metrics
-	RenameMetrics     map[string]string                    `json:"rename_metrics"`      // Map to rename metric name from key to value
-	IntervalStamp     bool                                 `json:"interval_timestamp"`  // Update timestamp periodically by ticker each interval?
-	NumCacheIntervals int                                  `json:"num_cache_intervals"` // Number of intervals of cached metrics for evaluation
-	MaxForward        int                                  `json:"max_forward"`         // Number of maximal forwarded metrics at one select
-	NormalizeUnits    bool                                 `json:"normalize_units"`     // Check unit meta flag and normalize it using cc-units
-	ChangeUnitPrefix  map[string]string                    `json:"change_unit_prefix"`  // Add prefix that should be applied to the metrics
+	HostnameTagName   string                               `json:"hostname_tag,omitempty"`        // Key name used when adding the hostname to a metric (default 'hostname')
+	AddTags           []metricRouterTagConfig              `json:"add_tags,omitempty"`            // List of tags that are added when the condition is met
+	DelTags           []metricRouterTagConfig              `json:"delete_tags,omitempty"`         // List of tags that are removed when the condition is met
+	IntervalAgg       []agg.MetricAggregatorIntervalConfig `json:"interval_aggregates,omitempty"` // List of aggregation function processed at the end of an interval
+	DropMetrics       []string                             `json:"drop_metrics,omitempty"`        // List of metric names to drop. For fine-grained dropping use drop_metrics_if
+	DropMetricsIf     []string                             `json:"drop_metrics_if,omitempty"`     // List of evaluatable terms to drop metrics
+	RenameMetrics     map[string]string                    `json:"rename_metrics,omitempty"`      // Map to rename metric name from key to value
+	IntervalStamp     bool                                 `json:"interval_timestamp,omitempty"`  // Update timestamp periodically by ticker each interval?
+	NumCacheIntervals int                                  `json:"num_cache_intervals,omitempty"` // Number of intervals of cached metrics for evaluation
+	MaxForward        int                                  `json:"max_forward,omitempty"`         // Number of maximal forwarded metrics at one select
+	NormalizeUnits    bool                                 `json:"normalize_units,omitempty"`     // Check unit meta flag and normalize it using cc-units
+	ChangeUnitPrefix  map[string]string                    `json:"change_unit_prefix,omitempty"`  // Add prefix that should be applied to the metrics
 	MessageProcessor  json.RawMessage                      `json:"process_messages,omitempty"`
 }

@@ -297,7 +297,7 @@ func (r *metricRouter) Start() {

 			case timestamp := <-timeChan:
 				r.timestamp = timestamp
-				cclog.ComponentDebug("MetricRouter", "Update timestamp", r.timestamp.UnixNano())
+				cclog.ComponentDebugf("MetricRouter", "Update timestamp %d", r.timestamp.UnixNano())

 			case p := <-r.coll_input:
 				coll_forward(p)
Author	SHA1	Message	Date
Thomas Roehl	c9dfca622f	Update cclog calls in latest IB Collector	2026-06-08 14:58:29 +02:00
Thomas Roehl	3b0638e815	Merge branch 'main' into cclog_update	2026-06-08 14:57:31 +02:00
Thomas Roehl	037b4f1526	Update cclog calls	2026-06-08 14:52:24 +02:00
Holger Obermaier	bed5491068	Fix Overflows in Infiniband collector (#219 ) * Add information about the used infiniband counters * Change datatype from int64 to uint64 * uint64 subtraction handles wraparound automatically * Compute total rates by summing up the xmit and recv rates. This avoids overflows in the raw counters * Check for cases where the current counter can not be saved as last state * Use golang variable naming convention (camelCase)	2026-06-08 14:00:09 +02:00
dependabot[bot]	a2eba41150	Bump golang.design/x/thread Bumps [golang.design/x/thread](https://github.com/golang-design/thread) from 0.0.0-20210122121316-335e9adffdf1 to 0.3.2. - [Release notes](https://github.com/golang-design/thread/releases) - [Commits](https://github.com/golang-design/thread/commits/v0.3.2) --- updated-dependencies: - dependency-name: golang.design/x/thread dependency-version: 0.3.2 dependency-type: direct:production update-type: version-update:semver-minor ... Signed-off-by: dependabot[bot] <support@github.com>	2026-06-08 13:10:27 +02:00
Thomas Roehl	5d55ee7a77	Fix lint errors	2026-06-02 16:35:11 +02:00
Thomas Roehl	5938368a76	Add collector to get Nvidia GPM metrics	2026-06-02 13:52:44 +02:00
dependabot[bot]	077204d39f	Bump github.com/tklauser/go-sysconf from 0.3.16 to 0.4.0 Bumps [github.com/tklauser/go-sysconf](https://github.com/tklauser/go-sysconf) from 0.3.16 to 0.4.0. - [Release notes](https://github.com/tklauser/go-sysconf/releases) - [Commits](https://github.com/tklauser/go-sysconf/compare/v0.3.16...v0.4.0) --- updated-dependencies: - dependency-name: github.com/tklauser/go-sysconf dependency-version: 0.4.0 dependency-type: direct:production update-type: version-update:semver-minor ... Signed-off-by: dependabot[bot] <support@github.com>	2026-05-25 12:17:29 +02:00
dependabot[bot]	dcc9746df4	Bump golang.org/x/sys from 0.43.0 to 0.45.0 Bumps [golang.org/x/sys](https://github.com/golang/sys) from 0.43.0 to 0.45.0. - [Commits](https://github.com/golang/sys/compare/v0.43.0...v0.45.0) --- updated-dependencies: - dependency-name: golang.org/x/sys dependency-version: 0.45.0 dependency-type: direct:production update-type: version-update:semver-minor ... Signed-off-by: dependabot[bot] <support@github.com>	2026-05-25 12:10:50 +02:00
dependabot[bot]	2c51a3ed72	Bump github.com/fsnotify/fsnotify from 1.10.0 to 1.10.1 Bumps [github.com/fsnotify/fsnotify](https://github.com/fsnotify/fsnotify) from 1.10.0 to 1.10.1. - [Release notes](https://github.com/fsnotify/fsnotify/releases) - [Changelog](https://github.com/fsnotify/fsnotify/blob/main/CHANGELOG.md) - [Commits](https://github.com/fsnotify/fsnotify/compare/v1.10.0...v1.10.1) --- updated-dependencies: - dependency-name: github.com/fsnotify/fsnotify dependency-version: 1.10.1 dependency-type: direct:production update-type: version-update:semver-patch ... Signed-off-by: dependabot[bot] <support@github.com>	2026-05-25 12:10:16 +02:00
Holger Obermaier	656ea73d12	Fix: num_cpus could not be excluded	2026-05-07 14:47:23 +02:00
Holger Obermaier	330f923596	Fixed exclude_metrics and check for used metrics	2026-05-07 12:25:07 +02:00