add probe help

pkg/exporter/probe/legacy.go

@@ -20,6 +20,11 @@ func BuildStandardMetricsLabelValues(entity *nettop.Entity) []string {
 	return append(metaPodLabels, BuildAdditionalLabelsValues(entity.GetLabels())...)
 }
 
+type LegacyMetric struct {
+	Name string
+	Help string
+}
+
 func InitAdditionalLabels(additionalLabels []string) error {
 	if len(additionalLabels) == 0 {
 		return nil
@@ -84,15 +89,15 @@ func newMetricsName(module, name string) string {
 
 type LegacyCollector func() (map[string]map[uint32]uint64, error)
 
-func NewLegacyBatchMetrics(module string, metrics []string, collector LegacyCollector) prometheus.Collector {
+func NewLegacyBatchMetrics(module string, metrics []LegacyMetric, collector LegacyCollector) prometheus.Collector {
 	return newLegacyBatchMetrics(module, metrics, collector)
 }
 
-func newLegacyBatchMetrics(module string, metrics []string, collector LegacyCollector) prometheus.Collector {
+func newLegacyBatchMetrics(module string, metrics []LegacyMetric, collector LegacyCollector) prometheus.Collector {
 	descs := make(map[string]*prometheus.Desc)
 	for _, m := range metrics {
-		newName := newMetricsName(module, m)
-		descs[newName] = prometheus.NewDesc(newName, "", StandardMetricsLabels, nil)
+		newName := newMetricsName(module, m.Name)
+		descs[newName] = prometheus.NewDesc(newName, m.Help, StandardMetricsLabels, nil)
 	}
 	return &legacyBatchMetrics{
 		module:    module,

pkg/exporter/probe/nlconntrack/conntrackmetrics.go

@@ -26,7 +26,19 @@ var (
 	MaxEntries = "maxentries"
 
 	// stats of conntrack status summary
-	conntrackMetrics = []string{Found, Invalid, Ignore, Insert, InsertFailed, Drop, EarlyDrop, Error, SearchRestart, Entries, MaxEntries}
+	conntrackMetrics = []probe.LegacyMetric{
+		{Name: Found, Help: "The total number of tracked connections found in the conntrack table."},
+		{Name: Invalid, Help: "The total number of invalid connections encountered."},
+		{Name: Ignore, Help: "The total number of connections that were ignored by the conntrack module."},
+		{Name: Insert, Help: "The total number of connections inserted into the conntrack table."},
+		{Name: InsertFailed, Help: "The total number of failed attempts to insert a connection into the conntrack table."},
+		{Name: Drop, Help: "The total number of connections dropped from the conntrack table."},
+		{Name: EarlyDrop, Help: "The total number of connections dropped early before they were fully established."},
+		{Name: Error, Help: "The total number of errors encountered while managing connections in the conntrack table."},
+		{Name: SearchRestart, Help: "The total number of times the search for a connection entry was restarted."},
+		{Name: Entries, Help: "The current number of connections tracked in the conntrack table."},
+		{Name: MaxEntries, Help: "The maximum number of entries allowed in the conntrack table."},
+	}
 )
 
 func metricsProbeCreator() (probe.MetricsProbe, error) {
@@ -80,7 +92,7 @@ func (c *conntrackMetricsProbe) CollectOnce() (map[string]map[uint32]uint64, err
 	}
 
 	for _, metric := range conntrackMetrics {
-		resMap[metric] = map[uint32]uint64{uint32(nettop.InitNetns): stats[metric]}
+		resMap[metric.Name] = map[uint32]uint64{uint32(nettop.InitNetns): stats[metric.Name]}
 	}
 
 	return resMap, nil

pkg/exporter/probe/nlqdisc/nlqdiscstats.go

@@ -52,7 +52,14 @@ var (
 	Backlog    = "backlog"
 	Overlimits = "overlimits"
 
-	qdiscMetrics = []string{Bytes, Packets, Drops, Qlen, Backlog, Overlimits}
+	qdiscMetrics = []probe.LegacyMetric{
+		{Name: Bytes, Help: "The total number of bytes transmitted through the queuing discipline."},
+		{Name: Packets, Help: "The total number of packets transmitted through the queuing discipline."},
+		{Name: Drops, Help: "The total number of packets dropped by the queuing discipline."},
+		{Name: Qlen, Help: "The current length of the queue (the number of packets queued)."},
+		{Name: Backlog, Help: "The total amount of data currently in the queue (in bytes)."},
+		{Name: Overlimits, Help: "The total number of packets that exceeded the configured limits."},
+	}
 )
 
 func init() {
@@ -80,7 +87,7 @@ func (p *Probe) Stop(_ context.Context) error {
 func (p *Probe) CollectOnce() (map[string]map[uint32]uint64, error) {
 	resMap := make(map[string]map[uint32]uint64)
 	for _, metric := range qdiscMetrics {
-		resMap[metric] = make(map[uint32]uint64)
+		resMap[metric.Name] = make(map[uint32]uint64)
 	}
 
 	ets := nettop.GetAllUniqueNetnsEntity()

pkg/exporter/probe/procfd/procfd.go

@@ -35,8 +35,8 @@ func fdProbeCreator() (probe.MetricsProbe, error) {
 		Subsystem:      probeName,
 		VariableLabels: probe.StandardMetricsLabels,
 		SingleMetricsOpts: []probe.SingleMetricsOpts{
-			{Name: OpenFD, ValueType: prometheus.GaugeValue},
-			{Name: OpenSocket, ValueType: prometheus.GaugeValue},
+			{Name: OpenFD, Help: "The total number of open file descriptors for the process", ValueType: prometheus.GaugeValue},
+			{Name: OpenSocket, Help: "The total number of open sockets for the process", ValueType: prometheus.GaugeValue},
 		},
 	}
 	metrics := probe.NewBatchMetrics(opts, p.collectOnce)

pkg/exporter/probe/procio/procio.go

@@ -36,10 +36,10 @@ func ioProbeCreator() (probe.MetricsProbe, error) {
 		Subsystem:      probeName,
 		VariableLabels: probe.StandardMetricsLabels,
 		SingleMetricsOpts: []probe.SingleMetricsOpts{
-			{Name: IOReadSyscall, ValueType: prometheus.CounterValue},
-			{Name: IOWriteSyscall, ValueType: prometheus.CounterValue},
-			{Name: IOReadBytes, ValueType: prometheus.CounterValue},
-			{Name: IOWriteBytes, ValueType: prometheus.CounterValue},
+			{Name: IOReadSyscall, Help: "The total number of read system calls made by the process", ValueType: prometheus.CounterValue},
+			{Name: IOWriteSyscall, Help: "The total number of write system calls made by the process", ValueType: prometheus.CounterValue},
+			{Name: IOReadBytes, Help: "The total number of bytes read by the process", ValueType: prometheus.CounterValue},
+			{Name: IOWriteBytes, Help: "The total number of bytes written by the process", ValueType: prometheus.CounterValue},
 		},
 	}
 	metrics := probe.NewBatchMetrics(opts, p.collectOnce)

pkg/exporter/probe/procipvs/ipvsservicestats.go

@@ -25,7 +25,13 @@ var (
 	IncomingBytes   = "incomingbytes"
 	OutgoingBytes   = "outgoingbytes"
 
-	IPVSMetrics = []string{Connections, IncomingPackets, OutgoingBytes, IncomingBytes, OutgoingPackets}
+	IPVSMetrics = []probe.LegacyMetric{
+		{Name: Connections, Help: "The total number of connections handled by the IPVS (IP Virtual Server)"},
+		{Name: IncomingPackets, Help: "The total number of incoming packets processed by the IPVS"},
+		{Name: OutgoingBytes, Help: "The total number of bytes sent out by the IPVS"},
+		{Name: IncomingBytes, Help: "The total number of bytes received by the IPVS"},
+		{Name: OutgoingPackets, Help: "The total number of outgoing packets processed by the IPVS"},
+	}
 )
 
 func init() {

pkg/exporter/probe/procnetdev/procnetdev.go

@@ -23,7 +23,16 @@ const (
 )
 
 var (
-	NetdevMetrics = []string{RxBytes, RxErrors, TxBytes, TxErrors, RxPackets, RxDropped, TxPackets, TxDropped}
+	NetdevMetrics = []probe.LegacyMetric{
+		{Name: RxBytes, Help: "The total number of bytes received on the network interface."},
+		{Name: RxErrors, Help: "The total number of errors encountered while receiving on the network interface."},
+		{Name: TxBytes, Help: "The total number of bytes transmitted on the network interface."},
+		{Name: TxErrors, Help: "The total number of errors encountered while transmitting on the network interface."},
+		{Name: RxPackets, Help: "The total number of packets received on the network interface."},
+		{Name: RxDropped, Help: "The total number of received packets that were dropped on the network interface."},
+		{Name: TxPackets, Help: "The total number of packets transmitted on the network interface."},
+		{Name: TxDropped, Help: "The total number of transmitted packets that were dropped on the network interface."},
+	}
 )
 
 func init() {
@@ -60,7 +69,7 @@ func (s *ProcNetdev) CollectOnce() (map[string]map[uint32]uint64, error) {
 func collect(nslist []*nettop.Entity) (map[string]map[uint32]uint64, error) {
 	resMap := make(map[string]map[uint32]uint64)
 	for _, m := range NetdevMetrics {
-		resMap[m] = make(map[uint32]uint64)
+		resMap[m.Name] = make(map[uint32]uint64)
 	}
 
 	netdev := getAllNetdev(nslist)

pkg/exporter/probe/procnetstat/procnetstat.go

@@ -60,38 +60,40 @@ const (
 )
 
 var (
-	TCPExtMetrics = []string{TCPListenDrops,
-		TCPListenOverflows,
-		TCPSynRetrans,
-		TCPFastRetrans,
-		TCPRetransFail,
-		TCPTimeouts,
-		TCPAbortOnClose,
-		TCPAbortOnMemory,
-		TCPAbortOnTimeout,
-		TCPAbortOnLinger,
-		TCPAbortOnData,
-		TCPAbortFailed,
-		TCPACKSkippedSynRecv,
-		TCPACKSkippedPAWS,
-		TCPACKSkippedSeq,
-		TCPACKSkippedFinWait2,
-		TCPACKSkippedTimeWait,
-		TCPACKSkippedChallenge,
-		TCPRcvQDrop,
-		TCPMemoryPressures,
-		TCPMemoryPressuresChrono,
-		PAWSActive,
-		PAWSEstab,
-		EmbryonicRsts,
-		TCPWinProbe,
-		TCPKeepAlive,
-		TCPMTUPFail,
-		TCPMTUPSuccess,
-		TCPZeroWindowDrop,
-		TCPBacklogDrop,
-		PFMemallocDrop,
-		TCPWqueueTooBig}
+	TCPExtMetrics = []probe.LegacyMetric{
+		{Name: TCPListenDrops, Help: "The total number of TCP connection requests that were dropped because the listen queue was full."},
+		{Name: TCPListenOverflows, Help: "The total number of times the TCP listen queue has overflown."},
+		{Name: TCPSynRetrans, Help: "The total number of SYN packets that were retransmitted."},
+		{Name: TCPFastRetrans, Help: "The total number of fast retransmissions made by TCP."},
+		{Name: TCPRetransFail, Help: "The total number of failed retransmissions in TCP."},
+		{Name: TCPTimeouts, Help: "The total number of TCP timeouts."},
+		{Name: TCPAbortOnClose, Help: "The number of TCP connections that were aborted on close."},
+		{Name: TCPAbortOnMemory, Help: "The number of TCP connections that were aborted due to memory allocation failures."},
+		{Name: TCPAbortOnTimeout, Help: "The number of TCP connections that were aborted due to timeouts."},
+		{Name: TCPAbortOnLinger, Help: "The number of TCP connections that were aborted due to linger timeouts."},
+		{Name: TCPAbortOnData, Help: "The number of TCP connections that were aborted due to data-related issues."},
+		{Name: TCPAbortFailed, Help: "The number of attempts to abort TCP connections that failed."},
+		{Name: TCPACKSkippedSynRecv, Help: "The number of ACKs skipped while in SYN_RECV state."},
+		{Name: TCPACKSkippedPAWS, Help: "The number of ACKs skipped due to PAWS (Protection Against Wrapped Sequence numbers)."},
+		{Name: TCPACKSkippedSeq, Help: "The number of ACKs skipped due to sequence number issues."},
+		{Name: TCPACKSkippedFinWait2, Help: "The number of ACKs skipped while in FIN_WAIT_2 state."},
+		{Name: TCPACKSkippedTimeWait, Help: "The number of ACKs skipped while in TIME_WAIT state."},
+		{Name: TCPACKSkippedChallenge, Help: "The number of ACKs skipped due to challenges in the communication."},
+		{Name: TCPRcvQDrop, Help: "The total number of received packets that were dropped due to queue overflow."},
+		{Name: TCPMemoryPressures, Help: "The total number of occasions where the TCP stack experienced memory pressure."},
+		{Name: TCPMemoryPressuresChrono, Help: "Chronological count of TCP memory pressure events."},
+		{Name: PAWSActive, Help: "Indicates whether the PAWS mechanism is active."},
+		{Name: PAWSEstab, Help: "The number of established connections utilizing PAWS."},
+		{Name: EmbryonicRsts, Help: "The number of embryonic (half-open) connections that were reset."},
+		{Name: TCPWinProbe, Help: "The total number of window probes sent to check for window size."},
+		{Name: TCPKeepAlive, Help: "The total number of TCP keepalive packets sent."},
+		{Name: TCPMTUPFail, Help: "The total number of MTU (Maximum Transmission Unit) probe failures."},
+		{Name: TCPMTUPSuccess, Help: "The total number of successful MTU (Maximum Transmission Unit) discoveries."},
+		{Name: TCPZeroWindowDrop, Help: "The total number of packets dropped due to a zero window condition."},
+		{Name: TCPBacklogDrop, Help: "The total number of packets dropped from the TCP backlog queue."},
+		{Name: PFMemallocDrop, Help: "The total number of packets dropped due to PF_MEMALLOC allocations failing."},
+		{Name: TCPWqueueTooBig, Help: "The total number of TCP send queue drops due to the queue being too large."},
+	}
 )
 
 func init() {
@@ -129,7 +131,7 @@ func collect(nslist []*nettop.Entity) (map[string]map[uint32]uint64, error) {
 	resMap := make(map[string]map[uint32]uint64)
 
 	for _, stat := range TCPExtMetrics {
-		resMap[stat] = make(map[uint32]uint64)
+		resMap[stat.Name] = make(map[uint32]uint64)
 	}
 
 	for _, et := range nslist {
@@ -141,13 +143,13 @@ func collect(nslist []*nettop.Entity) (map[string]map[uint32]uint64, error) {
 
 		extstats := stats[ProtocolTCPExt]
 		for _, stat := range TCPExtMetrics {
-			if _, ok := extstats[stat]; ok {
-				data, err := strconv.ParseUint(extstats[stat], 10, 64)
+			if _, ok := extstats[stat.Name]; ok {
+				data, err := strconv.ParseUint(extstats[stat.Name], 10, 64)
 				if err != nil {
 					log.Errorf("%s failed parse stat %s, pid: %d err: %v", probeName, stat, et.GetPid(), err)
 					continue
 				}
-				resMap[stat][uint32(et.GetNetns())] += data
+				resMap[stat.Name][uint32(et.GetNetns())] += data
 			}
 		}
 	}

pkg/exporter/probe/procsnmp/procsnmp.go

@@ -82,11 +82,53 @@ const (
 )
 
 var (
-	TCPStatMetrcis = []string{TCPActiveOpens, TCPPassiveOpens, TCPRetransSegs, TCPAttemptFails, TCPEstabResets, TCPCurrEstab, TCPInSegs, TCPOutSegs, TCPInErrs, TCPOutRsts}
-	UDPStatMetrics = []string{UDPInDatagrams, UDPNoPorts, UDPInErrors, UDPOutDatagrams, UDPRcvbufErrors, UDPSndbufErrors, UDPInCsumErrors, UDPIgnoredMulti}
-	IPMetrics      = []string{IPForwarding, IPDefaultTTL, IPInReceives, IPInHdrErrors, IPInAddrErrors, IPForwDatagrams, IPInUnknownProtos, IPInDiscards, IPInDelivers, IPOutRequests, IPOutDiscards, IPOutNoRoutes, IPReasmTimeout, IPReasmReqds, IPReasmOKs, IPReasmFails, IPFragOKs, IPFragFails, IPFragCreates}
+	TCPStatMetrcis = []probe.LegacyMetric{
+		{Name: TCPActiveOpens, Help: "The number of active TCP connections opened."},
+		{Name: TCPPassiveOpens, Help: "The number of passive TCP connections opened (i.e., connections established by accepting incoming connections)."},
+		{Name: TCPRetransSegs, Help: "The total number of segments that have been retransmitted."},
+		{Name: TCPAttemptFails, Help: "The number of failed attempts to establish a TCP connection."},
+		{Name: TCPEstabResets, Help: "The number of established TCP connections that were reset."},
+		{Name: TCPCurrEstab, Help: "The current number of established TCP connections."},
+		{Name: TCPInSegs, Help: "The total number of TCP segments received."},
+		{Name: TCPOutSegs, Help: "The total number of TCP segments sent."},
+		{Name: TCPInErrs, Help: "The total number of erroneous packets received on TCP."},
+		{Name: TCPOutRsts, Help: "The total number of TCP segments sent with the RST flag set."},
+	}
+
+	UDPStatMetrics = []probe.LegacyMetric{
+		{Name: UDPInDatagrams, Help: "The total number of UDP datagrams received."},
+		{Name: UDPNoPorts, Help: "The total number of UDP datagrams received for which there was no port at the destination."},
+		{Name: UDPInErrors, Help: "The total number of erroneous received UDP packets."},
+		{Name: UDPOutDatagrams, Help: "The total number of UDP datagrams sent."},
+		{Name: UDPRcvbufErrors, Help: "The total number of UDP datagrams dropped due to socket receive buffer errors."},
+		{Name: UDPSndbufErrors, Help: "The total number of UDP datagrams dropped due to socket send buffer errors."},
+		{Name: UDPInCsumErrors, Help: "The total number of UDP datagrams received with a checksum error."},
+		{Name: UDPIgnoredMulti, Help: "The total number of received UDP multicast packets that were ignored."},
+	}
+
+	IPMetrics = []probe.LegacyMetric{
+		{Name: IPForwarding, Help: "Indicates whether IP forwarding is enabled (1 for enabled, 0 for disabled)."},
+		{Name: IPDefaultTTL, Help: "The default time-to-live (TTL) value for IP packets."},
+		{Name: IPInReceives, Help: "The total number of IP packets received."},
+		{Name: IPInHdrErrors, Help: "The total number of received IP packets that had a header error."},
+		{Name: IPInAddrErrors, Help: "The total number of received IP packets that were discarded due to address errors."},
+		{Name: IPForwDatagrams, Help: "The total number of IP packets forwarded by this machine."},
+		{Name: IPInUnknownProtos, Help: "The total number of received IP packets for which the protocol is not known."},
+		{Name: IPInDiscards, Help: "The total number of received IP packets that were discarded."},
+		{Name: IPInDelivers, Help: "The total number of delivered IP packets."},
+		{Name: IPOutRequests, Help: "The total number of IP packets sent out."},
+		{Name: IPOutDiscards, Help: "The total number of outgoing IP packets that were discarded."},
+		{Name: IPOutNoRoutes, Help: "The total number of outgoing IP packets for which no route could be found."},
+		{Name: IPReasmTimeout, Help: "The total number of times that IP reassembly timed out."},
+		{Name: IPReasmReqds, Help: "The total number of IP reassembly requests made."},
+		{Name: IPReasmOKs, Help: "The total number of successful IP reassembly operations."},
+		{Name: IPReasmFails, Help: "The total number of failed IP reassembly operations."},
+		{Name: IPFragOKs, Help: "The total number of IP packets that were fragmented successfully."},
+		{Name: IPFragFails, Help: "The total number of IP packets that failed to fragment."},
+		{Name: IPFragCreates, Help: "The total number of IP fragments created."},
+	}
 
-	metricsMap = map[string][]string{
+	metricsMap = map[string][]probe.LegacyMetric{
 		TCP: TCPStatMetrcis,
 		UDP: UDPStatMetrics,
 		IP:  IPMetrics,
@@ -165,7 +207,7 @@ func collect() (map[string]map[string]map[uint32]uint64, error) {
 	for proto, metricsList := range metricsMap {
 		res[proto] = make(map[string]map[uint32]uint64)
 		for _, metrics := range metricsList {
-			res[proto][metrics] = make(map[uint32]uint64)
+			res[proto][metrics.Name] = make(map[uint32]uint64)
 		}
 	}
 

pkg/exporter/probe/procsock/procsock.go

@@ -28,8 +28,15 @@ const (
 )
 
 var (
-	TCPSockStatMetrics = []string{TCPSockInuse, TCPSockOrphan, TCPSockTimewait, TCPSockeAlloc, TCPSockeMem}
-	probeName          = "sock"
+	TCPSockStatMetrics = []probe.LegacyMetric{
+		{Name: TCPSockInuse, Help: "The total number of TCP sockets currently in use."},
+		{Name: TCPSockOrphan, Help: "The total number of orphaned TCP sockets."},
+		{Name: TCPSockTimewait, Help: "The total number of TCP sockets in the TIME_WAIT state."},
+		{Name: TCPSockeAlloc, Help: "The total number of TCP sockets allocated."},
+		{Name: TCPSockeMem, Help: "The total amount of memory allocated for TCP sockets."},
+	}
+
+	probeName = "sock"
 )
 
 func init() {
@@ -70,7 +77,7 @@ type tcpsockstat struct {
 func collect() (resMap map[string]map[uint32]uint64, err error) {
 	resMap = make(map[string]map[uint32]uint64)
 	for _, stat := range TCPSockStatMetrics {
-		resMap[stat] = map[uint32]uint64{}
+		resMap[stat.Name] = map[uint32]uint64{}
 	}
 
 	// for _, nslogic := range nslist {

pkg/exporter/probe/procsoftnet/procsoftnet.go

@@ -25,7 +25,10 @@ const (
 )
 
 var (
-	softnetMetrics = []string{SNProcessed, SNDropped}
+	softnetMetrics = []probe.LegacyMetric{
+		{Name: SNProcessed, Help: "The total number of packets processed by the softnet layer"},
+		{Name: SNDropped, Help: "The total number of packets dropped by the softnet layer"},
+	}
 )
 
 func init() {
@@ -63,7 +66,7 @@ func collect(nslist []*nettop.Entity) (map[string]map[uint32]uint64, error) {
 	resMap := make(map[string]map[uint32]uint64)
 
 	for _, m := range softnetMetrics {
-		resMap[m] = map[uint32]uint64{}
+		resMap[m.Name] = map[uint32]uint64{}
 	}
 
 	for _, ns := range nslist {
@@ -72,7 +75,7 @@ func collect(nslist []*nettop.Entity) (map[string]map[uint32]uint64, error) {
 			continue
 		}
 		for _, m := range softnetMetrics {
-			resMap[m][uint32(ns.GetNetns())] = stat[m]
+			resMap[m.Name][uint32(ns.GetNetns())] = stat[m.Name]
 		}
 	}
 	return resMap, nil