Feat/processes

Readme.md

@@ -10,6 +10,7 @@ Noderig metrics:
 - Load
 - Disk
 - Net
+- Processes
 - External collectors
 
 ## Status
@@ -40,6 +41,7 @@ Flags:
       --mem uint8         memory metrics level (default 1)
       --net uint8         network metrics level (default 1)
       --load uint8        load metrics level (default 1)
+      --process uint8   processes metrics level (default 0)
   -c  --collectors string external collectors directory (default "./collectors")
   -k  --keep-for uint     keep collectors data for the given number of fetch (default 3)
 ```
@@ -119,6 +121,12 @@ Noderig have some built-in collectors.
 <tr><td>os.net.dropped{direction=out,iface=eth0}</td><td>iface out drop count (drops)</td></tr>
 </table>
 
+### Process
+<table>
+<tr><td>0</td><td></td><td>disabled metrics</td></tr>
+<tr><td rowspan="2">1</td><td>os.process.up{name=ntpd}</td><td>is process running (boolean)</td></tr>
+</table>
+
 ## Configuration
 
 Noderig can read a simple default [config file](config.yaml).
@@ -140,11 +148,12 @@ Noderig have some built-in collectors. They could be configured by a log level.
 You can also defined custom collectors, in an scollector way. (see: http://bosun.org/scollector/external-collectors)
 
 ```yaml
-cpu: 1  # CPU collector level     (Optional, default: 1)
-mem: 1  # Memory collector level  (Optional, default: 1)
-load: 1 # Load collector level    (Optional, default: 1)
-disk: 1 # Disk collector level    (Optional, default: 1)
-net: 1  # Network collector level (Optional, default: 1)
+cpu: 1      # CPU collector level     (Optional, default: 1)
+mem: 1      # Memory collector level  (Optional, default: 1)
+load: 1     # Load collector level    (Optional, default: 1)
+disk: 1     # Disk collector level    (Optional, default: 1)
+net: 1      # Network collector level (Optional, default: 1)
+process: 0  # Process collector level (Optional, default: 0)
 ```
 
 #### Collectors Modules
@@ -175,6 +184,13 @@ disk-opts:
     - sda3
 ```
 
+```yaml
+process-opts:
+  whitelist: # Give a filtering list of processes to watch
+    - httpd
+    - noderig
+```
+
 #### Parameters
 
 Noderig can be customized through some parameters.

cmd/root.go

@@ -9,7 +9,7 @@ import (
 
 	"net/http"
 
-	log "github.com/Sirupsen/logrus"
+	log "github.com/sirupsen/logrus"
 	"github.com/spf13/cobra"
 	"github.com/spf13/viper"
 
@@ -29,6 +29,7 @@ func init() {
 	RootCmd.Flags().Uint8("mem", 1, "memory metrics level")
 	RootCmd.Flags().Uint8("disk", 1, "disk metrics level")
 	RootCmd.Flags().Uint8("net", 1, "network metrics level")
+	RootCmd.Flags().Uint8("process", 0, "process metrics level")
 	RootCmd.Flags().Uint64("period", 1000, "default collection period")
 	RootCmd.Flags().StringP("collectors", "c", "./collectors", "external collectors directory")
 	RootCmd.Flags().Uint64P("keep-for", "k", 3, "keep collectors data for the given number of fetch")
@@ -101,6 +102,9 @@ var RootCmd = &cobra.Command{
 		disk := collectors.NewDisk(uint(viper.GetInt("period")), uint8(viper.GetInt("disk")), viper.Get("disk-opts"))
 		cs = append(cs, disk)
 
+		process := collectors.NewProcess(uint(viper.GetInt("period")), uint8(viper.GetInt("process")), viper.Sub("process-opts"))
+		cs = append(cs, process)
+
 		// Load external collectors
 		cpath := viper.GetString("collectors")
 		cdir, err := os.Open(cpath)

cmd/version.go

@@ -7,7 +7,7 @@ import (
 )
 
 var (
-	version = "2.3.1"
+	version = "2.3.2"
 	githash = "HEAD"
 )
 

collectors/collector.go

@@ -14,7 +14,7 @@ import (
 	"sync"
 	"time"
 
-	log "github.com/Sirupsen/logrus"
+	log "github.com/sirupsen/logrus"
 	"github.com/shirou/gopsutil/cpu"
 )
 

collectors/cpu.go

@@ -7,7 +7,7 @@ import (
 	"sync"
 	"time"
 
-	log "github.com/Sirupsen/logrus"
+	log "github.com/sirupsen/logrus"
 	"github.com/shirou/gopsutil/cpu"
 	"github.com/shirou/gopsutil/host"
 )

collectors/disk.go

@@ -7,7 +7,7 @@ import (
 	"sync"
 	"time"
 
-	log "github.com/Sirupsen/logrus"
+	log "github.com/sirupsen/logrus"
 	"github.com/shirou/gopsutil/disk"
 )
 

collectors/load.go

@@ -6,7 +6,7 @@ import (
 	"sync"
 	"time"
 
-	log "github.com/Sirupsen/logrus"
+	log "github.com/sirupsen/logrus"
 	"github.com/shirou/gopsutil/load"
 )
 

collectors/memory.go

@@ -6,7 +6,7 @@ import (
 	"sync"
 	"time"
 
-	log "github.com/Sirupsen/logrus"
+	log "github.com/sirupsen/logrus"
 	"github.com/shirou/gopsutil/mem"
 )
 

collectors/net.go

@@ -6,7 +6,7 @@ import (
 	"sync"
 	"time"
 
-	log "github.com/Sirupsen/logrus"
+	log "github.com/sirupsen/logrus"
 	"github.com/shirou/gopsutil/net"
 )
 

collectors/process.go

@@ -0,0 +1,101 @@
+package collectors
+
+import (
+	"bytes"
+	"fmt"
+	"strings"
+	"sync"
+	"time"
+
+	"github.com/mitchellh/go-ps"
+	log "github.com/sirupsen/logrus"
+	"github.com/spf13/viper"
+)
+
+// Process collector
+type Process struct {
+	whitelist []string
+	mutex     sync.RWMutex
+	sensision bytes.Buffer
+	level     uint8
+}
+
+// NewProcess collector
+func NewProcess(period uint, level uint8, opts *viper.Viper) *Process {
+	p := &Process{
+		level: level,
+	}
+
+	if opts != nil {
+		p.whitelist = opts.GetStringSlice("whitelist")
+	}
+	log.Debugf("Process whitelist: %+v", p.whitelist)
+
+	if p.level == 0 || len(p.whitelist) == 0 {
+		return p
+	}
+
+	log.Debugf("Process whitelist: %+v", p.whitelist)
+
+	tick := time.Tick(time.Duration(period) * time.Millisecond)
+	go func() {
+		for range tick {
+			if err := p.scrape(); err != nil {
+				log.Error(err)
+			}
+		}
+	}()
+
+	return p
+}
+
+// Metrics delivers metrics.
+func (p *Process) Metrics() *bytes.Buffer {
+	p.mutex.RLock()
+	defer p.mutex.RUnlock()
+
+	var res bytes.Buffer
+	res.Write(p.sensision.Bytes())
+	return &res
+}
+
+func (p *Process) scrape() error {
+	processes, err := ps.Processes()
+	if err != nil {
+		return err
+	}
+	now := time.Now()
+
+	res := map[string]bool{}
+
+	for _, processName := range p.whitelist {
+		res[processName] = false
+		for _, process := range processes {
+			if strings.Contains(process.Executable(), processName) {
+				res[processName] = true
+				continue
+			}
+		}
+	}
+
+	// protect consistency
+	p.mutex.Lock()
+	defer p.mutex.Unlock()
+
+	p.sensision.Reset()
+
+	for pcs, ok := range res {
+		if _, err := p.sensision.WriteString(processMetric(now, pcs, ok)); err != nil {
+			return err
+		}
+	}
+
+	return nil
+}
+
+func processMetric(t time.Time, processName string, up bool) string {
+	if up {
+		return fmt.Sprintf("%d// os.process.up{name=%s} true\n", t.UnixNano()/1000, processName)
+	}
+	return fmt.Sprintf("%d// os.process.up{name=%s} false\n", t.UnixNano()/1000, processName)
+}

config.yaml

@@ -3,8 +3,13 @@ mem: 1
 load: 1
 disk: 1
 net: 1
+process: 0
 disk-opts:
   names:
     - sda1
+process-opts:
+  whitelist:
+    - noderig
+    - ntpd
 #flushPath: /tmp/noderig
 #listen: none

glide.yaml

@@ -1,6 +1,6 @@
 package: github.com/ovh/noderig
 import:
-- package: github.com/Sirupsen/logrus
+- package: github.com/sirupsen/logrus
   version: ^0.11.0
 - package: github.com/spf13/cobra
 - package: github.com/spf13/viper

noderig.go

@@ -11,7 +11,7 @@
 package main
 
 import (
-	log "github.com/Sirupsen/logrus"
+	log "github.com/sirupsen/logrus"
 
 	"github.com/ovh/noderig/cmd"
 )