print help when -h is present regardless of other flags

* refactor exporter code for better readability and structure
* modified startup log output a little

README.md

@@ -1,3 +1,7 @@
+[![Actions](https://github.com/tlinden/io-exporter/actions/workflows/ci.yaml/badge.svg)](https://github.com/tlinden/io-exporter/actions)
+[![License](https://img.shields.io/badge/license-GPL-blue.svg)](https://github.com/tlinden/io-exporter/blob/master/LICENSE)
+[![Go Report Card](https://goreportcard.com/badge/github.com/tlinden/io-exporter)](https://goreportcard.com/report/github.com/tlinden/io-exporter)
+
 # io-exporter
 
 Report if a given filesystem is operating properly
@@ -14,10 +18,15 @@ specified via commandline.
 ```default
 io-exporter [options] <file>
 Options:
--t --timeout <int>          When should the operation timeout in seconds
--l --label   <label=value>  Add label to exported metric
--h --help                   Show help
--v --version                Show program version
+-t --timeout   <int>          When should the operation timeout in seconds
+-s --sleeptime <int>          Time to sleep between checks (default: 5s)
+-l --label     <label=value>  Add label to exported metric
+-i --internals                Also add labels about resource usage
+-r --read                     Only execute the read test
+-w --write                    Only execute the write test
+-d --debug                    Enable debug log level
+-h --help                     Show help
+-v --version                  Show program version
 ```
 
 ## Output
@@ -31,14 +40,20 @@ io-exporter -l foo=bar -l blah=blubb t/blah
 You'll get such metrics:
 
 ```default
-# HELP io_exporter_io_latency how long does the operation take in seconds
-# TYPE io_exporter_io_latency gauge
-io_exporter_io_latency{file="/tmp/blah",maxwait="1",namespace="debug",pod="foo1"} 0.0001142815
 # HELP io_exporter_io_operation whether io is working on the pvc, 1=ok, 0=fail
 # TYPE io_exporter_io_operation gauge
-io_exporter_io_operation{file="/tmp/blah",maxwait="1",namespace="debug",pod="foo1"} 1
+io_exporter_io_operation{blah="blubb",exectime="1761148383705",file="t/blah",foo="bar",maxwait="1"} 1
+# HELP io_exporter_io_read_latency how long does the read operation take in seconds
+# TYPE io_exporter_io_read_latency gauge
+io_exporter_io_read_latency{blah="blubb",exectime="1761148383705",file="t/blah",foo="bar",maxwait="1"} 0.0040411716
+# HELP io_exporter_io_write_latency how long does the write operation take in seconds
+# TYPE io_exporter_io_write_latency gauge
+io_exporter_io_write_latency{blah="blubb",exectime="1761148383705",file="t/blah",foo="bar",maxwait="1"} 0
 ```
 
+You may  also restrict the exporter  to only test read  (`-r` flag) or
+write (`-w` flag) operation.
+
 ## Installation
 
 There are no released binaries yet.
@@ -76,6 +91,15 @@ Or use the pre-build image:
 docker run -u `id -u $USER` -v ./t:/pvc ghcr.io/tlinden/io-exporter:latest /pvc/testfile
 ```
 
+## Grafana
+
+I provide a [sample dashboard](grafana), which you can add to your grafana or use
+as a starting point to integrate it into your monitoring setup.
+
+It looks like this:
+
+![Screenshot](https://github.com/TLINDEN/io-exporter/blob/main/grafana/screenshot.png)
+
 # Report bugs
 
 [Please open an issue](https://github.com/TLINDEN/io-exporter/issues). Thanks!

blah

@@ -1 +0,0 @@
-AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA

cmd/alloc.go

@@ -1,6 +1,11 @@
 package cmd
 
-import "github.com/ncw/directio"
+import (
+	"bytes"
+	"errors"
+
+	"github.com/ncw/directio"
+)
 
 // aligned allocs used for testing
 type Alloc struct {
@@ -25,3 +30,12 @@ func NewAlloc() *Alloc {
 		readBlock:  directio.AlignedBlock(directio.BlockSize),
 	}
 }
+
+func (alloc *Alloc) Compare() bool {
+	// compare
+	if !bytes.Equal(alloc.writeBlock, alloc.readBlock) {
+		return report(errors.New("read not the same as written"), nil)
+	}
+
+	return true
+}

cmd/config.go

@@ -15,7 +15,7 @@ import (
 )
 
 const (
-	Version = `v0.0.4`
+	Version = `v0.0.8`
 	SLEEP   = 5
 	Usage   = `io-exporter [options] <file>
 Options:
@@ -23,9 +23,15 @@ Options:
 -s --sleeptime <int>          Time to sleep between checks (default: 5s)
 -l --label     <label=value>  Add label to exported metric
 -i --internals                Also add labels about resource usage
+-r --read                     Only execute the read test
+-w --write                    Only execute the write test
 -d --debug                    Enable debug log level
 -h --help                     Show help
 -v --version                  Show program version`
+
+	O_R = iota
+	O_W
+	O_RW
 )
 
 // config via commandline flags
@@ -34,6 +40,8 @@ type Config struct {
 	Showhelp    bool     `koanf:"help"`      // -h
 	Internals   bool     `koanf:"internals"` // -i
 	Debug       bool     `koanf:"debug"`     // -d
+	ReadMode    bool     `koanf:"read"`      // -r
+	WriteMode   bool     `koanf:"write"`     // -w
 	Label       []string `koanf:"label"`     // -v
 	Timeout     int      `koanf:"timeout"`   // -t
 	Port        int      `koanf:"port"`      // -p
@@ -60,6 +68,8 @@ func InitConfig(output io.Writer) (*Config, error) {
 	flagset.BoolP("help", "h", false, "show help")
 	flagset.BoolP("debug", "d", false, "enable debug logs")
 	flagset.BoolP("internals", "i", false, "add internal metrics")
+	flagset.BoolP("read", "r", false, "only execute the read test")
+	flagset.BoolP("write", "w", false, "only execute the write test")
 	flagset.StringArrayP("label", "l", nil, "additional labels")
 	flagset.IntP("timeout", "t", 1, "timeout for file operation in seconds")
 	flagset.IntP("port", "p", 9187, "prometheus metrics port to listen to")
@@ -103,5 +113,10 @@ func InitConfig(output io.Writer) (*Config, error) {
 		conf.Labels = append(conf.Labels, Label{Name: parts[0], Value: parts[1]})
 	}
 
+	if !conf.ReadMode && !conf.WriteMode {
+		conf.ReadMode = true
+		conf.WriteMode = true
+	}
+
 	return conf, nil
 }

cmd/exporter.go

@@ -1,47 +1,114 @@
 package cmd
 
 import (
-	"bytes"
 	"context"
 	"errors"
 	"io"
 	"log/slog"
 	"os"
+	"sync"
 	"time"
 
 	"github.com/ncw/directio"
 )
 
-func die(err error, fd *os.File) bool {
-	slog.Debug("failed to check io", "error", err)
-
-	if fd != nil {
-		if err := fd.Close(); err != nil {
-			slog.Debug("failed to close filehandle", "error", err)
-		}
-	}
-
-	return false
+// our primary container for the io checks
+type Exporter struct {
+	conf    *Config
+	alloc   *Alloc
+	metrics *Metrics
 }
 
-// Calls runcheck() with timeout
-func runExporter(file string, alloc *Alloc, timeout time.Duration) bool {
+type Result struct {
+	result  bool
+	elapsed float64
+}
+
+func NewExporter(conf *Config, alloc *Alloc, metrics *Metrics) *Exporter {
+	return &Exporter{
+		conf:    conf,
+		alloc:   alloc,
+		metrics: metrics,
+	}
+}
+
+// starts the primary go-routine, which will run the io checks for ever
+func (exp *Exporter) RunIOchecks() *sync.WaitGroup {
+	var wg sync.WaitGroup
+	wg.Add(1)
+
+	go func() {
+		for {
+			var res_r, res_w Result
+
+			exp.alloc.Clean()
+
+			if exp.conf.WriteMode {
+				res_w = exp.measure(O_W)
+				slog.Debug("elapsed write time", "elapsed", res_w.elapsed, "result", res_w.result)
+			}
+
+			if exp.conf.ReadMode {
+				res_r = exp.measure(O_R)
+				slog.Debug("elapsed read time", "elapsed", res_r.elapsed, "result", res_r.result)
+			}
+
+			if (exp.conf.WriteMode && exp.conf.ReadMode) && (res_r.result && res_w.result) {
+				if !exp.alloc.Compare() {
+					res_r.result = false
+				}
+			}
+
+			exp.metrics.Set(res_r, res_w)
+
+			time.Sleep(time.Duration(exp.conf.Sleeptime) * time.Second)
+		}
+	}()
+
+	return &wg
+}
+
+// call an io measurement and collect time needed
+func (exp *Exporter) measure(mode int) Result {
+	start := time.Now()
+
+	result := exp.runExporter(mode)
+
+	// ns => s
+	now := time.Now()
+	elapsed := float64(now.Sub(start).Nanoseconds()) / 10000000000
+
+	// makes no sense to measure latency if operation failed
+	if !result {
+		elapsed = 0
+	}
+
+	return Result{elapsed: elapsed, result: result}
+}
+
+// Calls runcheck's with context timeout
+func (exp *Exporter) runExporter(mode int) bool {
 	ctx := context.Background()
-	ctx, cancel := context.WithTimeout(ctx, timeout)
+	ctx, cancel := context.WithTimeout(ctx, time.Duration(exp.conf.Timeout)*time.Second)
 	defer cancel()
 
 	run := make(chan struct{}, 1)
 	var res bool
 
 	go func() {
-		res = runcheck(file, alloc)
+		switch mode {
+		case O_R:
+			res = exp.runcheck_r()
+		case O_W:
+			res = exp.runcheck_w()
+		}
 		run <- struct{}{}
 	}()
 
 	for {
 		select {
 		case <-ctx.Done():
-			return die(ctx.Err(), nil)
+			return report(ctx.Err(), nil)
 		case <-run:
 			return res
 		}
@@ -50,65 +117,66 @@ func runExporter(file string, alloc *Alloc, timeout time.Duration) bool {
 
 // Checks file io on the specified path:
 //
-// - open the file (create if it doesnt exist)
-// - truncate it if it already exists
-// - write some data to it
-// - closes the file
-// - re-opens it for reading
+// - opens it for reading
 // - reads the block
-// - compares if written block is equal to read block
 // - closes file again
 //
 // Returns false if anything failed during that sequence,
 // true otherwise.
-func runcheck(file string, alloc *Alloc) bool {
-	alloc.Clean()
-
-	// write
-	fd, err := directio.OpenFile(file, os.O_RDWR|os.O_TRUNC|os.O_CREATE, 0640)
-	if err != nil {
-		die(err, nil)
-	}
-
-	for i := 0; i < len(alloc.writeBlock); i++ {
-		alloc.writeBlock[i] = 'A'
-	}
-
-	n, err := fd.Write(alloc.writeBlock)
-	if err != nil {
-		return die(err, fd)
-	}
-
-	if n != len(alloc.writeBlock) {
-		return die(errors.New("failed to write block"), fd)
-	}
-
-	if err := fd.Close(); err != nil {
-		return die(err, nil)
-	}
-
+func (exp *Exporter) runcheck_r() bool {
 	// read
-	in, err := directio.OpenFile(file, os.O_RDONLY, 0640)
+	in, err := directio.OpenFile(exp.conf.File, os.O_RDONLY, 0640)
 	if err != nil {
-		die(err, nil)
+		report(err, nil)
 	}
 
-	n, err = io.ReadFull(in, alloc.readBlock)
+	n, err := io.ReadFull(in, exp.alloc.readBlock)
 	if err != nil {
-		return die(err, in)
+		return report(err, in)
 	}
 
-	if n != len(alloc.writeBlock) {
-		return die(errors.New("failed to read block"), fd)
+	if n != len(exp.alloc.writeBlock) {
+		return report(errors.New("failed to read block"), in)
 	}
 
 	if err := in.Close(); err != nil {
-		return die(err, nil)
-	}
-
-	// compare
-	if !bytes.Equal(alloc.writeBlock, alloc.readBlock) {
-		return die(errors.New("read not the same as written"), nil)
+		return report(err, nil)
+	}
+
+	return true
+}
+
+// Checks file io on the specified path:
+//
+// - open the file (create if it doesnt exist)
+// - truncate it if it already exists
+// - write some data to it
+// - closes the file
+//
+// Returns false if anything failed during that sequence,
+// true otherwise.
+func (exp *Exporter) runcheck_w() bool {
+	// write
+	fd, err := directio.OpenFile(exp.conf.File, os.O_RDWR|os.O_TRUNC|os.O_CREATE, 0640)
+	if err != nil {
+		report(err, nil)
+	}
+
+	for i := 0; i < len(exp.alloc.writeBlock); i++ {
+		exp.alloc.writeBlock[i] = 'A'
+	}
+
+	n, err := fd.Write(exp.alloc.writeBlock)
+	if err != nil {
+		return report(err, fd)
+	}
+
+	if n != len(exp.alloc.writeBlock) {
+		return report(errors.New("failed to write block"), fd)
+	}
+
+	if err := fd.Close(); err != nil {
+		return report(err, nil)
 	}
 
 	return true

cmd/metrics.go

@@ -2,6 +2,7 @@ package cmd
 
 import (
 	"fmt"
+	"time"
 
 	"github.com/prometheus/client_golang/prometheus"
 	"github.com/prometheus/client_golang/prometheus/collectors"
@@ -14,15 +15,17 @@ type Label struct {
 
 // simple prometheus wrapper
 type Metrics struct {
-	run      *prometheus.GaugeVec
-	latency  *prometheus.GaugeVec
-	registry *prometheus.Registry
-	values   []string
+	run       *prometheus.GaugeVec
+	latency_r *prometheus.GaugeVec
+	latency_w *prometheus.GaugeVec
+	registry  *prometheus.Registry
+	values    []string
+	mode      int
 }
 
 func NewMetrics(conf *Config) *Metrics {
-	labels := []string{"file", "maxwait"}
-	LabelLen := 2
+	labels := []string{"file", "maxwait", "exectime"}
+	LabelLen := 3
 
 	for _, label := range conf.Labels {
 		labels = append(labels, label.Name)
@@ -36,10 +39,17 @@ func NewMetrics(conf *Config) *Metrics {
 			},
 			labels,
 		),
-		latency: prometheus.NewGaugeVec(
+		latency_r: prometheus.NewGaugeVec(
 			prometheus.GaugeOpts{
-				Name: "io_exporter_io_latency",
-				Help: "how long does the operation take in seconds",
+				Name: "io_exporter_io_read_latency",
+				Help: "how long does the read operation take in seconds",
+			},
+			labels,
+		),
+		latency_w: prometheus.NewGaugeVec(
+			prometheus.GaugeOpts{
+				Name: "io_exporter_io_write_latency",
+				Help: "how long does the write operation take in seconds",
 			},
 			labels,
 		),
@@ -53,7 +63,8 @@ func NewMetrics(conf *Config) *Metrics {
 	if conf.Internals {
 		metrics.registry.MustRegister(
 			metrics.run,
-			metrics.latency,
+			metrics.latency_r,
+			metrics.latency_w,
 
 			// we  might need  to take  care of the  exporter in  terms of
 			// resources, so also report those internals
@@ -65,28 +76,50 @@ func NewMetrics(conf *Config) *Metrics {
 			),
 		)
 	} else {
-		metrics.registry.MustRegister(metrics.run, metrics.latency)
+		metrics.registry.MustRegister(metrics.run, metrics.latency_r, metrics.latency_w)
 	}
 
 	// static labels
 	metrics.values[0] = conf.File
 	metrics.values[1] = fmt.Sprintf("%d", conf.Timeout)
+	metrics.values[2] = fmt.Sprintf("%d", time.Now().UnixMilli())
 
 	// custom labels via -l label=value
 	for idx, label := range conf.Labels {
 		metrics.values[idx+LabelLen] = label.Value
 	}
 
+	switch {
+	case conf.ReadMode && conf.WriteMode:
+		metrics.mode = O_RW
+	case conf.ReadMode:
+		metrics.mode = O_R
+	case conf.WriteMode:
+		metrics.mode = O_W
+	}
+
 	return metrics
 }
 
-func (metrics *Metrics) Set(result bool, elapsed float64) {
+func (metrics *Metrics) Set(result_r, result_w Result) {
 	var res float64
 
-	if result {
-		res = 1
+	switch metrics.mode {
+	case O_RW:
+		if result_r.result && result_w.result {
+			res = 1
+		}
+	case O_R:
+		if result_r.result {
+			res = 1
+		}
+	case O_W:
+		if result_w.result {
+			res = 1
+		}
 	}
 
 	metrics.run.WithLabelValues(metrics.values...).Set(res)
-	metrics.latency.WithLabelValues(metrics.values...).Set(elapsed)
+	metrics.latency_r.WithLabelValues(metrics.values...).Set(result_r.elapsed)
+	metrics.latency_w.WithLabelValues(metrics.values...).Set(result_w.elapsed)
 }

cmd/root.go

@@ -7,11 +7,17 @@ import (
 	"net/http"
 	"os"
 	"strings"
-	"time"
+
+	// enable to debug with roumon
+	//_ "net/http/pprof"
+	// then: roumon -host=localhost -port=9187
 
 	"github.com/prometheus/client_golang/prometheus/promhttp"
 )
 
+// Main program. starts 2 goroutines: our exporter and the http server
+// for  the  prometheus  metrics.  The  exporter  reports  measurement
+// results to prometheus metrics directly
 func Run() {
 	conf, err := InitConfig(os.Stdout)
 	if err != nil {
@@ -23,36 +29,52 @@ func Run() {
 		os.Exit(0)
 	}
 
-	metrics := NewMetrics(conf)
-	alloc := NewAlloc()
-
+	if conf.Showhelp {
+		fmt.Println(Usage)
+		os.Exit(0)
+	}
 	setLogger(os.Stdout, conf.Debug)
 
-	go func() {
-		for {
-			start := time.Now()
+	metrics := NewMetrics(conf)
+	alloc := NewAlloc()
+	exporter := NewExporter(conf, alloc, metrics)
 
-			result := runExporter(conf.File, alloc, time.Duration(conf.Timeout)*time.Second)
-
-			// ns => s
-			now := time.Now()
-			elapsed := float64(now.Sub(start).Nanoseconds()) / 10000000000
-			slog.Debug("elapsed time", "elapsed", elapsed, "result", result)
-
-			metrics.Set(result, elapsed)
-
-			time.Sleep(time.Duration(conf.Sleeptime) * time.Second)
-		}
-	}()
+	wg := exporter.RunIOchecks()
 
 	http.Handle("/metrics", promhttp.HandlerFor(
 		metrics.registry,
 		promhttp.HandlerOpts{},
 	))
 
-	slog.Info("start testing and serving metrics on localhost", "port", conf.Port)
-	slog.Info("test setup", "file", conf.File, "labels", strings.Join(conf.Label, ","))
+	slog.Info(" ╭──")
+	slog.Info(" │ io-exporter starting up", "version", Version)
+	slog.Info(" │ serving metrics", "host", "localhost", "port", conf.Port)
+	slog.Info(" │ test setup", "file", conf.File, "labels", strings.Join(conf.Label, ","))
+	slog.Info(" │ measuring", "read", conf.ReadMode, "write", conf.WriteMode, "timeout(s)", conf.Timeout)
+	slog.Info(" │ debugging", "enabled", conf.Debug)
+	slog.Info(" ╰──")
+
 	if err := http.ListenAndServe(fmt.Sprintf(":%d", conf.Port), nil); err != nil {
 		log.Fatal(err)
 	}
+
+	wg.Wait()
+}
+
+func report(err error, fd *os.File) bool {
+	failure := err.Error()
+	if err.Error() == "context deadline exceeded" {
+		failure = "operation timed out"
+	}
+
+	slog.Error("io error", "error", failure)
+
+	if fd != nil {
+		if err := fd.Close(); err != nil {
+
+			slog.Debug("failed to close filehandle", "error", failure)
+		}
+	}
+
+	return false
 }

grafana/dashboard.js

@@ -0,0 +1,580 @@
+{
+  "annotations": {
+    "list": [
+      {
+        "builtIn": 1,
+        "datasource": {
+          "type": "grafana",
+          "uid": "-- Grafana --"
+        },
+        "enable": true,
+        "hide": true,
+        "iconColor": "rgba(0, 211, 255, 1)",
+        "name": "Annotations & Alerts",
+        "type": "dashboard"
+      }
+    ]
+  },
+  "editable": true,
+  "fiscalYearStartMonth": 0,
+  "graphTooltip": 0,
+  "id": 37,
+  "links": [],
+  "panels": [
+    {
+      "datasource": {
+        "type": "prometheus",
+        "uid": "P5DCFC7561CCDE821"
+      },
+      "fieldConfig": {
+        "defaults": {
+          "color": {
+            "mode": "thresholds"
+          },
+          "custom": {
+            "fillOpacity": 70,
+            "hideFrom": {
+              "legend": false,
+              "tooltip": false,
+              "viz": false
+            },
+            "insertNulls": false,
+            "lineWidth": 0,
+            "spanNulls": false
+          },
+          "mappings": [],
+          "thresholds": {
+            "mode": "absolute",
+            "steps": [
+              {
+                "color": "green",
+                "value": null
+              },
+              {
+                "color": "red",
+                "value": 0
+              },
+              {
+                "color": "green",
+                "value": 1
+              }
+            ]
+          },
+          "unit": "bool_yes_no"
+        },
+        "overrides": []
+      },
+      "gridPos": {
+        "h": 9,
+        "w": 24,
+        "x": 0,
+        "y": 0
+      },
+      "id": 4,
+      "options": {
+        "alignValue": "left",
+        "legend": {
+          "displayMode": "list",
+          "placement": "bottom",
+          "showLegend": true
+        },
+        "mergeValues": true,
+        "rowHeight": 0.9,
+        "showValue": "auto",
+        "tooltip": {
+          "mode": "single",
+          "sort": "none"
+        }
+      },
+      "pluginVersion": "11.3.8+security-01",
+      "targets": [
+        {
+          "editorMode": "code",
+          "expr": "sum(io_exporter_io_operation{container=\"ioexporter\",namespace=\"$namespace\"}) by (pod)",
+          "legendFormat": "{{pod}}",
+          "range": true,
+          "refId": "A"
+        }
+      ],
+      "title": "IO Operation Result",
+      "type": "state-timeline"
+    },
+    {
+      "datasource": {
+        "type": "prometheus",
+        "uid": "P5DCFC7561CCDE821"
+      },
+      "fieldConfig": {
+        "defaults": {
+          "color": {
+            "mode": "palette-classic"
+          },
+          "custom": {
+            "axisBorderShow": false,
+            "axisCenteredZero": false,
+            "axisColorMode": "text",
+            "axisLabel": "",
+            "axisPlacement": "auto",
+            "barAlignment": 0,
+            "barWidthFactor": 0.6,
+            "drawStyle": "line",
+            "fillOpacity": 0,
+            "gradientMode": "none",
+            "hideFrom": {
+              "legend": false,
+              "tooltip": false,
+              "viz": false
+            },
+            "insertNulls": false,
+            "lineInterpolation": "linear",
+            "lineWidth": 1,
+            "pointSize": 5,
+            "scaleDistribution": {
+              "type": "linear"
+            },
+            "showPoints": "auto",
+            "spanNulls": false,
+            "stacking": {
+              "group": "A",
+              "mode": "none"
+            },
+            "thresholdsStyle": {
+              "mode": "off"
+            }
+          },
+          "mappings": [],
+          "thresholds": {
+            "mode": "absolute",
+            "steps": [
+              {
+                "color": "green",
+                "value": null
+              },
+              {
+                "color": "red",
+                "value": 80
+              }
+            ]
+          },
+          "unit": "s"
+        },
+        "overrides": []
+      },
+      "gridPos": {
+        "h": 10,
+        "w": 12,
+        "x": 0,
+        "y": 9
+      },
+      "id": 1,
+      "options": {
+        "legend": {
+          "calcs": [
+            "lastNotNull",
+            "max",
+            "min"
+          ],
+          "displayMode": "table",
+          "placement": "bottom",
+          "showLegend": true
+        },
+        "tooltip": {
+          "mode": "single",
+          "sort": "none"
+        }
+      },
+      "pluginVersion": "11.3.8+security-01",
+      "targets": [
+        {
+          "datasource": {
+            "type": "prometheus",
+            "uid": "P5DCFC7561CCDE821"
+          },
+          "editorMode": "code",
+          "expr": "sum(io_exporter_io_read_latency{container=\"ioexporter\",namespace=\"$namespace\"}) by (pod)",
+          "legendFormat": "{{pod}}",
+          "range": true,
+          "refId": "A"
+        }
+      ],
+      "title": "IO Read Latency",
+      "type": "timeseries"
+    },
+    {
+      "datasource": {
+        "type": "prometheus",
+        "uid": "P5DCFC7561CCDE821"
+      },
+      "fieldConfig": {
+        "defaults": {
+          "color": {
+            "mode": "palette-classic"
+          },
+          "custom": {
+            "axisBorderShow": false,
+            "axisCenteredZero": false,
+            "axisColorMode": "text",
+            "axisLabel": "",
+            "axisPlacement": "auto",
+            "barAlignment": 0,
+            "barWidthFactor": 0.6,
+            "drawStyle": "line",
+            "fillOpacity": 0,
+            "gradientMode": "none",
+            "hideFrom": {
+              "legend": false,
+              "tooltip": false,
+              "viz": false
+            },
+            "insertNulls": false,
+            "lineInterpolation": "linear",
+            "lineWidth": 1,
+            "pointSize": 5,
+            "scaleDistribution": {
+              "type": "linear"
+            },
+            "showPoints": "auto",
+            "spanNulls": false,
+            "stacking": {
+              "group": "A",
+              "mode": "none"
+            },
+            "thresholdsStyle": {
+              "mode": "off"
+            }
+          },
+          "mappings": [],
+          "thresholds": {
+            "mode": "absolute",
+            "steps": [
+              {
+                "color": "green",
+                "value": null
+              },
+              {
+                "color": "red",
+                "value": 80
+              }
+            ]
+          },
+          "unit": "s"
+        },
+        "overrides": []
+      },
+      "gridPos": {
+        "h": 10,
+        "w": 12,
+        "x": 12,
+        "y": 9
+      },
+      "id": 3,
+      "options": {
+        "legend": {
+          "calcs": [
+            "lastNotNull",
+            "max",
+            "min"
+          ],
+          "displayMode": "table",
+          "placement": "bottom",
+          "showLegend": true
+        },
+        "tooltip": {
+          "mode": "single",
+          "sort": "none"
+        }
+      },
+      "pluginVersion": "11.3.8+security-01",
+      "targets": [
+        {
+          "datasource": {
+            "type": "prometheus",
+            "uid": "P5DCFC7561CCDE821"
+          },
+          "editorMode": "code",
+          "expr": "sum(io_exporter_io_write_latency{container=\"ioexporter\",namespace=\"$namespace\"}) by (pod)",
+          "legendFormat": "{{pod}}",
+          "range": true,
+          "refId": "A"
+        }
+      ],
+      "title": "IO Write Latency",
+      "type": "timeseries"
+    },
+    {
+      "datasource": {
+        "type": "prometheus",
+        "uid": "P5DCFC7561CCDE821"
+      },
+      "description": "",
+      "fieldConfig": {
+        "defaults": {
+          "color": {
+            "mode": "palette-classic-by-name"
+          },
+          "custom": {
+            "axisBorderShow": false,
+            "axisCenteredZero": false,
+            "axisColorMode": "text",
+            "axisLabel": "",
+            "axisPlacement": "auto",
+            "barAlignment": 0,
+            "barWidthFactor": 0.6,
+            "drawStyle": "line",
+            "fillOpacity": 55,
+            "gradientMode": "none",
+            "hideFrom": {
+              "legend": false,
+              "tooltip": false,
+              "viz": false
+            },
+            "insertNulls": false,
+            "lineInterpolation": "linear",
+            "lineWidth": 0,
+            "pointSize": 5,
+            "scaleDistribution": {
+              "type": "linear"
+            },
+            "showPoints": "never",
+            "spanNulls": false,
+            "stacking": {
+              "group": "A",
+              "mode": "normal"
+            },
+            "thresholdsStyle": {
+              "mode": "off"
+            }
+          },
+          "mappings": [],
+          "min": 0,
+          "thresholds": {
+            "mode": "absolute",
+            "steps": [
+              {
+                "color": "green",
+                "value": null
+              },
+              {
+                "color": "red",
+                "value": 80
+              }
+            ]
+          },
+          "unit": "s"
+        },
+        "overrides": [
+          {
+            "__systemRef": "hideSeriesFrom",
+            "matcher": {
+              "id": "byNames",
+              "options": {
+                "mode": "exclude",
+                "names": [
+                  "Value"
+                ],
+                "prefix": "All except:",
+                "readOnly": true
+              }
+            },
+            "properties": [
+              {
+                "id": "custom.hideFrom",
+                "value": {
+                  "legend": false,
+                  "tooltip": false,
+                  "viz": true
+                }
+              }
+            ]
+          }
+        ]
+      },
+      "gridPos": {
+        "h": 9,
+        "w": 12,
+        "x": 0,
+        "y": 19
+      },
+      "id": 5,
+      "interval": "1m",
+      "options": {
+        "legend": {
+          "calcs": [
+            "lastNotNull",
+            "mean",
+            "min",
+            "max"
+          ],
+          "displayMode": "table",
+          "placement": "bottom",
+          "showLegend": true
+        },
+        "tooltip": {
+          "mode": "single",
+          "sort": "none"
+        }
+      },
+      "pluginVersion": "11.3.8+security-01",
+      "targets": [
+        {
+          "datasource": {
+            "uid": "$datasource"
+          },
+          "editorMode": "code",
+          "expr": "sum(node_namespace_pod_container:container_cpu_usage_seconds_total:sum_irate{namespace=\"$namespace\", container=\"ioexporter\"}) by (pod)",
+          "format": "time_series",
+          "intervalFactor": 2,
+          "legendFormat": "{{pod}}",
+          "range": true,
+          "refId": "A"
+        }
+      ],
+      "title": "CPU Usage IO Exporter",
+      "type": "timeseries"
+    },
+    {
+      "datasource": {
+        "type": "prometheus",
+        "uid": "P5DCFC7561CCDE821"
+      },
+      "description": "Memory Usage and Limit in Bytes",
+      "fieldConfig": {
+        "defaults": {
+          "color": {
+            "mode": "palette-classic"
+          },
+          "custom": {
+            "axisBorderShow": false,
+            "axisCenteredZero": false,
+            "axisColorMode": "text",
+            "axisLabel": "",
+            "axisPlacement": "auto",
+            "barAlignment": 0,
+            "barWidthFactor": 0.6,
+            "drawStyle": "line",
+            "fillOpacity": 10,
+            "gradientMode": "none",
+            "hideFrom": {
+              "legend": false,
+              "tooltip": false,
+              "viz": false
+            },
+            "insertNulls": false,
+            "lineInterpolation": "linear",
+            "lineWidth": 1,
+            "pointSize": 5,
+            "scaleDistribution": {
+              "type": "linear"
+            },
+            "showPoints": "never",
+            "spanNulls": false,
+            "stacking": {
+              "group": "A",
+              "mode": "none"
+            },
+            "thresholdsStyle": {
+              "mode": "off"
+            }
+          },
+          "links": [],
+          "mappings": [],
+          "thresholds": {
+            "mode": "absolute",
+            "steps": [
+              {
+                "color": "green",
+                "value": null
+              },
+              {
+                "color": "red",
+                "value": 80
+              }
+            ]
+          },
+          "unit": "decbytes"
+        },
+        "overrides": []
+      },
+      "gridPos": {
+        "h": 9,
+        "w": 12,
+        "x": 12,
+        "y": 19
+      },
+      "id": 6,
+      "options": {
+        "alertThreshold": true,
+        "legend": {
+          "calcs": [
+            "mean",
+            "lastNotNull",
+            "max",
+            "min"
+          ],
+          "displayMode": "table",
+          "placement": "bottom",
+          "showLegend": true
+        },
+        "tooltip": {
+          "mode": "multi",
+          "sort": "none"
+        }
+      },
+      "pluginVersion": "11.3.8+security-01",
+      "targets": [
+        {
+          "datasource": {
+            "type": "prometheus",
+            "uid": "P5DCFC7561CCDE821"
+          },
+          "editorMode": "code",
+          "exemplar": true,
+          "expr": "sum(container_memory_working_set_bytes{namespace=\"$namespace\", container=\"ioexporter\"}) by (pod)",
+          "format": "time_series",
+          "interval": "",
+          "intervalFactor": 2,
+          "legendFormat": "{{pod}}",
+          "range": true,
+          "refId": "A"
+        }
+      ],
+      "title": "Average Memory Usage IO Explorer",
+      "type": "timeseries"
+    }
+  ],
+  "preload": false,
+  "schemaVersion": 40,
+  "tags": [],
+  "templating": {
+    "list": [
+      {
+        "current": {
+          "text": "exporter-test",
+          "value": "exporter-test"
+        },
+        "definition": "label_values(io_exporter_io_operation,namespace)",
+        "includeAll": true,
+        "name": "namespace",
+        "options": [],
+        "query": {
+          "qryType": 1,
+          "query": "label_values(io_exporter_io_operation,namespace)",
+          "refId": "PrometheusVariableQueryEditor-VariableQuery"
+        },
+        "refresh": 1,
+        "regex": "",
+        "type": "query"
+      }
+    ]
+  },
+  "time": {
+    "from": "now-6h",
+    "to": "now"
+  },
+  "timepicker": {},
+  "timezone": "browser",
+  "title": "io-test",
+  "uid": "ef1tcv0azbpc0e",
+  "version": 11,
+  "weekStart": ""
+}