ClusterCockpit/cc-metric-collector
1
0
Fork 0
mirror of https://github.com/ClusterCockpit/cc-metric-collector.git synced 2025-08-14 14:52:58 +02:00
Code Issues Packages Projects Releases Wiki Activity

Formatting

Browse Source
This commit is contained in:
Thomas Roehl
2021-12-21 14:04:31 +01:00
parent 988cea381e
commit a6feb16ec1
30 changed files with 775 additions and 793 deletions
Download Patch File Download Diff File Expand all files Collapse all files

183
collectors/collectorManager.go
View File

@@ -1,16 +1,15 @@
package collectors
import (
lp "github.com/ClusterCockpit/cc-metric-collector/internal/ccMetric"
"sync"
"time"
"log"
"os"
"encoding/json"
mct "github.com/ClusterCockpit/cc-metric-collector/internal/multiChanTicker"
"encoding/json"
lp "github.com/ClusterCockpit/cc-metric-collector/internal/ccMetric"
mct "github.com/ClusterCockpit/cc-metric-collector/internal/multiChanTicker"
"log"
"os"
"sync"
"time"
)
var AvailableCollectors = map[string]MetricCollector{
"likwid": &LikwidCollector{},
"loadavg": &LoadavgCollector{},
@@ -27,15 +26,14 @@ var AvailableCollectors = map[string]MetricCollector{
"ipmistat": &IpmiCollector{},
}
type collectorManager struct {
collectors []MetricCollector
output chan lp.CCMetric
done chan bool
ticker mct.MultiChanTicker
duration time.Duration
wg *sync.WaitGroup
config map[string]json.RawMessage
collectors []MetricCollector
output chan lp.CCMetric
done chan bool
ticker mct.MultiChanTicker
duration time.Duration
wg *sync.WaitGroup
config map[string]json.RawMessage
}
type CollectorManager interface {
@@ -45,96 +43,95 @@ type CollectorManager interface {
Close()
}
func (cm *collectorManager) Init(ticker mct.MultiChanTicker, duration time.Duration, wg *sync.WaitGroup, collectConfigFile string) error {
cm.collectors = make([]MetricCollector, 0)
cm.output = nil
cm.done = make(chan bool)
cm.wg = wg
cm.ticker = ticker
cm.duration = duration
configFile, err := os.Open(collectConfigFile)
if err != nil {
log.Print(err.Error())
return err
}
defer configFile.Close()
jsonParser := json.NewDecoder(configFile)
cm.collectors = make([]MetricCollector, 0)
cm.output = nil
cm.done = make(chan bool)
cm.wg = wg
cm.ticker = ticker
cm.duration = duration
configFile, err := os.Open(collectConfigFile)
if err != nil {
log.Print(err.Error())
return err
}
defer configFile.Close()
jsonParser := json.NewDecoder(configFile)
err = jsonParser.Decode(&cm.config)
if err != nil {
log.Print(err.Error())
return err
}
for k, cfg := range cm.config {
log.Print(k, " ", cfg)
if _, found := AvailableCollectors[k]; !found {
log.Print("[CollectorManager] SKIP unknown collector ", k)
continue
}
c := AvailableCollectors[k]
err = c.Init(cfg)
if err != nil {
log.Print("[CollectorManager] Collector ", k, "initialization failed: ", err.Error())
continue
}
cm.collectors = append(cm.collectors, c)
}
return nil
log.Print(err.Error())
return err
}
for k, cfg := range cm.config {
log.Print(k, " ", cfg)
if _, found := AvailableCollectors[k]; !found {
log.Print("[CollectorManager] SKIP unknown collector ", k)
continue
}
c := AvailableCollectors[k]
err = c.Init(cfg)
if err != nil {
log.Print("[CollectorManager] Collector ", k, "initialization failed: ", err.Error())
continue
}
cm.collectors = append(cm.collectors, c)
}
return nil
}
func (cm *collectorManager) Start() {
cm.wg.Add(1)
tick := make(chan time.Time)
cm.ticker.AddChannel(tick)
go func() {
for {
CollectorManagerLoop:
select {
case <- cm.done:
for _, c := range cm.collectors {
c.Close()
}
cm.wg.Done()
log.Print("[CollectorManager] DONE\n")
break CollectorManagerLoop
case t := <- tick:
for _, c := range cm.collectors {
CollectorManagerInputLoop:
select {
case <- cm.done:
for _, c := range cm.collectors {
c.Close()
}
cm.wg.Done()
log.Print("[CollectorManager] DONE\n")
break CollectorManagerInputLoop
default:
log.Print("[CollectorManager] ", c.Name(), " ", t)
c.Read(cm.duration, cm.output)
}
}
}
}
log.Print("[CollectorManager] EXIT\n")
}()
log.Print("[CollectorManager] STARTED\n")
cm.wg.Add(1)
tick := make(chan time.Time)
cm.ticker.AddChannel(tick)
go func() {
for {
CollectorManagerLoop:
select {
case <-cm.done:
for _, c := range cm.collectors {
c.Close()
}
cm.wg.Done()
log.Print("[CollectorManager] DONE\n")
break CollectorManagerLoop
case t := <-tick:
for _, c := range cm.collectors {
CollectorManagerInputLoop:
select {
case <-cm.done:
for _, c := range cm.collectors {
c.Close()
}
cm.wg.Done()
log.Print("[CollectorManager] DONE\n")
break CollectorManagerInputLoop
default:
log.Print("[CollectorManager] ", c.Name(), " ", t)
c.Read(cm.duration, cm.output)
}
}
}
}
log.Print("[CollectorManager] EXIT\n")
}()
log.Print("[CollectorManager] STARTED\n")
}
func (cm *collectorManager) AddOutput(output chan lp.CCMetric) {
cm.output = output
cm.output = output
}
func (cm *collectorManager) Close() {
cm.done <- true
log.Print("[CollectorManager] CLOSE")
cm.done <- true
log.Print("[CollectorManager] CLOSE")
}
func New(ticker mct.MultiChanTicker, duration time.Duration, wg *sync.WaitGroup, collectConfigFile string) (CollectorManager, error) {
cm := &collectorManager{}
err := cm.Init(ticker, duration, wg, collectConfigFile)
if err != nil {
return nil, err
}
return cm, err
cm := &collectorManager{}
err := cm.Init(ticker, duration, wg, collectConfigFile)
if err != nil {
return nil, err
}
return cm, err
}

2
collectors/cpustatMetric.go
View File

@@ -24,7 +24,7 @@ type CpustatCollector struct {
func (m *CpustatCollector) Init(config json.RawMessage) error {
m.name = "CpustatCollector"
m.setup()
m.meta = map[string]string{"source" : m.name, "group" : "CPU"}
m.meta = map[string]string{"source": m.name, "group": "CPU"}
if len(config) > 0 {
err := json.Unmarshal(config, &m.config)
if err != nil {

2
collectors/customCmdMetric.go
View File

@@ -32,7 +32,7 @@ type CustomCmdCollector struct {
func (m *CustomCmdCollector) Init(config json.RawMessage) error {
var err error
m.name = "CustomCmdCollector"
m.meta = map[string]string{"source" : m.name, "group" : "Custom"}
m.meta = map[string]string{"source": m.name, "group": "Custom"}
if len(config) > 0 {
err = json.Unmarshal(config, &m.config)
if err != nil {

2
collectors/diskstatMetric.go
View File

@@ -27,7 +27,7 @@ type DiskstatCollector struct {
func (m *DiskstatCollector) Init(config json.RawMessage) error {
var err error
m.name = "DiskstatCollector"
m.meta = map[string]string{"source" : m.name, "group" : "Disk"}
m.meta = map[string]string{"source": m.name, "group": "Disk"}
m.setup()
if len(config) > 0 {
err = json.Unmarshal(config, &m.config)

2
collectors/infinibandMetric.go
View File

@@ -54,7 +54,7 @@ func (m *InfinibandCollector) Init(config json.RawMessage) error {
m.name = "InfinibandCollector"
m.use_perfquery = false
m.setup()
m.meta = map[string]string{"source" : m.name, "group" : "Network"}
m.meta = map[string]string{"source": m.name, "group": "Network"}
m.tags = map[string]string{"type": "node"}
if len(config) > 0 {
err = json.Unmarshal(config, &m.config)

10
collectors/ipmiMetric.go
View File

@@ -31,7 +31,7 @@ type IpmiCollector struct {
func (m *IpmiCollector) Init(config json.RawMessage) error {
m.name = "IpmiCollector"
m.setup()
m.meta = map[string]string{"source" : m.name, "group" : "IPMI"}
m.meta = map[string]string{"source": m.name, "group": "IPMI"}
if len(config) > 0 {
err := json.Unmarshal(config, &m.config)
if err != nil {
@@ -85,7 +85,7 @@ func (m *IpmiCollector) readIpmiTool(cmd string, output chan lp.CCMetric) {
y, err := lp.New(name, map[string]string{"type": "node"}, m.meta, map[string]interface{}{"value": v}, time.Now())
if err == nil {
y.AddMeta("unit", unit)
y.AddMeta("unit", unit)
output <- y
}
}
@@ -112,9 +112,9 @@ func (m *IpmiCollector) readIpmiSensors(cmd string, output chan lp.CCMetric) {
name := strings.ToLower(strings.Replace(lv[1], " ", "_", -1))
y, err := lp.New(name, map[string]string{"type": "node"}, m.meta, map[string]interface{}{"value": v}, time.Now())
if err == nil {
if len(lv) > 4 {
y.AddMeta("unit", lv[4])
}
if len(lv) > 4 {
y.AddMeta("unit", lv[4])
}
output <- y
}
}

30
collectors/likwidMetric.go
View File

@@ -27,21 +27,21 @@ import (
type MetricScope int
const (
METRIC_SCOPE_HWTHREAD = iota
METRIC_SCOPE_SOCKET
METRIC_SCOPE_NUMA
METRIC_SCOPE_NODE
METRIC_SCOPE_HWTHREAD = iota
METRIC_SCOPE_SOCKET
METRIC_SCOPE_NUMA
METRIC_SCOPE_NODE
)
func (ms MetricScope) String() string {
return []string{"Head", "Shoulder", "Knee", "Toe"}[ms]
return []string{"Head", "Shoulder", "Knee", "Toe"}[ms]
}
type LikwidCollectorMetricConfig struct {
Name string `json:"name"`
Calc string `json:"calc"`
Scope MetricScope `json:"socket_scope"`
Publish bool `json:"publish"`
Name string `json:"name"`
Calc string `json:"calc"`
Scope MetricScope `json:"socket_scope"`
Publish bool `json:"publish"`
}
type LikwidCollectorEventsetConfig struct {
@@ -127,13 +127,13 @@ func (m *LikwidCollector) Init(config json.RawMessage) error {
}
}
m.setup()
m.meta = map[string]string{"source" : m.name, "group" : "PerfCounter"}
m.meta = map[string]string{"source": m.name, "group": "PerfCounter"}
cpulist := CpuList()
m.cpulist = make([]C.int, len(cpulist))
slist := getSocketCpus()
m.sock2tid = make(map[int]int)
// m.numa2tid = make(map[int]int)
// m.numa2tid = make(map[int]int)
for i, c := range cpulist {
m.cpulist[i] = C.int(c)
if sid, found := slist[m.cpulist[i]]; found {
@@ -264,7 +264,7 @@ func (m *LikwidCollector) Read(interval time.Duration, output chan lp.CCMetric)
for sid, tid := range m.sock2tid {
y, err := lp.New(metric.Name,
map[string]string{"type": "socket",
"type-id": fmt.Sprintf("%d", int(sid))},
"type-id": fmt.Sprintf("%d", int(sid))},
m.meta,
map[string]interface{}{"value": m.mresults[i][tid][metric.Name]},
time.Now())
@@ -276,7 +276,7 @@ func (m *LikwidCollector) Read(interval time.Duration, output chan lp.CCMetric)
for tid, cpu := range m.cpulist {
y, err := lp.New(metric.Name,
map[string]string{"type": "cpu",
"type-id": fmt.Sprintf("%d", int(cpu))},
"type-id": fmt.Sprintf("%d", int(cpu))},
m.meta,
map[string]interface{}{"value": m.mresults[i][tid][metric.Name]},
time.Now())
@@ -295,7 +295,7 @@ func (m *LikwidCollector) Read(interval time.Duration, output chan lp.CCMetric)
for sid, tid := range m.sock2tid {
y, err := lp.New(metric.Name,
map[string]string{"type": "socket",
"type-id": fmt.Sprintf("%d", int(sid))},
"type-id": fmt.Sprintf("%d", int(sid))},
m.meta,
map[string]interface{}{"value": m.gmresults[tid][metric.Name]},
time.Now())
@@ -307,7 +307,7 @@ func (m *LikwidCollector) Read(interval time.Duration, output chan lp.CCMetric)
for tid, cpu := range m.cpulist {
y, err := lp.New(metric.Name,
map[string]string{"type": "cpu",
"type-id": fmt.Sprintf("%d", int(cpu))},
"type-id": fmt.Sprintf("%d", int(cpu))},
m.meta,
map[string]interface{}{"value": m.gmresults[tid][metric.Name]},
time.Now())

2
collectors/loadavgMetric.go
View File

@@ -32,7 +32,7 @@ func (m *LoadavgCollector) Init(config json.RawMessage) error {
return err
}
}
m.meta = map[string]string{"source" : m.name, "group" : "LOAD"}
m.meta = map[string]string{"source": m.name, "group": "LOAD"}
m.tags = map[string]string{"type": "node"}
m.load_matches = []string{"load_one", "load_five", "load_fifteen"}
m.proc_matches = []string{"proc_run", "proc_total"}

8
collectors/lustreMetric.go
View File

@@ -37,7 +37,7 @@ func (m *LustreCollector) Init(config json.RawMessage) error {
}
m.setup()
m.tags = map[string]string{"type": "node"}
m.meta = map[string]string{"source" : m.name, "group" : "Lustre"}
m.meta = map[string]string{"source": m.name, "group": "Lustre"}
m.matches = map[string]map[string]int{"read_bytes": {"read_bytes": 6, "read_requests": 1},
"write_bytes": {"write_bytes": 6, "write_requests": 1},
"open": {"open": 1},
@@ -90,9 +90,9 @@ func (m *LustreCollector) Read(interval time.Duration, output chan lp.CCMetric)
if err == nil {
y, err := lp.New(name, m.tags, m.meta, map[string]interface{}{"value": x}, time.Now())
if err == nil {
if strings.Contains(name, "byte") {
y.AddMeta("unit", "Byte")
}
if strings.Contains(name, "byte") {
y.AddMeta("unit", "Byte")
}
output <- y
}
}

2
collectors/memstatMetric.go
View File

@@ -35,7 +35,7 @@ func (m *MemstatCollector) Init(config json.RawMessage) error {
return err
}
}
m.meta = map[string]string{"source" : m.name, "group" : "Memory", "unit": "kByte"}
m.meta = map[string]string{"source": m.name, "group": "Memory", "unit": "kByte"}
m.stats = make(map[string]int64)
m.matches = make(map[string]string)
m.tags = map[string]string{"type": "node"}

10
collectors/metricCollector.go
View File

@@ -1,6 +1,7 @@
package collectors
import (
"encoding/json"
"errors"
lp "github.com/ClusterCockpit/cc-metric-collector/internal/ccMetric"
influx "github.com/influxdata/line-protocol"
@@ -9,7 +10,6 @@ import (
"strconv"
"strings"
"time"
"encoding/json"
)
type MetricCollector interface {
@@ -21,10 +21,10 @@ type MetricCollector interface {
}
type metricCollector struct {
output chan lp.CCMetric
name string
init bool
meta map[string]string
output chan lp.CCMetric
name string
init bool
meta map[string]string
}
func (c *metricCollector) Name() string {

14
collectors/netstatMetric.go
View File

@@ -25,7 +25,7 @@ type NetstatCollector struct {
func (m *NetstatCollector) Init(config json.RawMessage) error {
m.name = "NetstatCollector"
m.setup()
m.meta = map[string]string{"source" : m.name, "group" : "Memory"}
m.meta = map[string]string{"source": m.name, "group": "Memory"}
m.matches = map[int]string{
1: "bytes_in",
9: "bytes_out",
@@ -75,12 +75,12 @@ func (m *NetstatCollector) Read(interval time.Duration, output chan lp.CCMetric)
if err == nil {
y, err := lp.New(name, tags, m.meta, map[string]interface{}{"value": int(float64(v) * 1.0e-3)}, time.Now())
if err == nil {
switch {
case strings.Contains(name, "byte"):
y.AddMeta("unit", "Byte")
case strings.Contains(name, "pkt"):
y.AddMeta("unit", "Packets")
}
switch {
case strings.Contains(name, "byte"):
y.AddMeta("unit", "Byte")
case strings.Contains(name, "pkt"):
y.AddMeta("unit", "Packets")
}
output <- y
}
}

10
collectors/nvidiaMetric.go
View File

@@ -4,8 +4,8 @@ import (
"encoding/json"
"errors"
"fmt"
"github.com/NVIDIA/go-nvml/pkg/nvml"
lp "github.com/ClusterCockpit/cc-metric-collector/internal/ccMetric"
"github.com/NVIDIA/go-nvml/pkg/nvml"
"log"
"time"
)
@@ -32,7 +32,7 @@ func (m *NvidiaCollector) Init(config json.RawMessage) error {
var err error
m.name = "NvidiaCollector"
m.setup()
m.meta = map[string]string{"source" : m.name, "group" : "Nvidia"}
m.meta = map[string]string{"source": m.name, "group": "Nvidia"}
if len(config) > 0 {
err = json.Unmarshal(config, &m.config)
if err != nil {
@@ -91,14 +91,14 @@ func (m *NvidiaCollector) Read(interval time.Duration, output chan lp.CCMetric)
_, skip = stringArrayContains(m.config.ExcludeMetrics, "mem_total")
y, err := lp.New("mem_total", tags, m.meta, map[string]interface{}{"value": t}, time.Now())
if err == nil && !skip {
y.AddMeta("unit", "MByte")
y.AddMeta("unit", "MByte")
output <- y
}
f := float64(meminfo.Used) / (1024 * 1024)
_, skip = stringArrayContains(m.config.ExcludeMetrics, "fb_memory")
y, err = lp.New("fb_memory", tags, m.meta, map[string]interface{}{"value": f}, time.Now())
if err == nil && !skip {
y.AddMeta("unit", "MByte")
y.AddMeta("unit", "MByte")
output <- y
}
}
@@ -108,7 +108,7 @@ func (m *NvidiaCollector) Read(interval time.Duration, output chan lp.CCMetric)
_, skip = stringArrayContains(m.config.ExcludeMetrics, "temp")
y, err := lp.New("temp", tags, m.meta, map[string]interface{}{"value": float64(temp)}, time.Now())
if err == nil && !skip {
y.AddMeta("unit", "degC")
y.AddMeta("unit", "degC")
output <- y
}
}

14
collectors/tempMetric.go
View File

@@ -5,12 +5,12 @@ import (
"fmt"
lp "github.com/ClusterCockpit/cc-metric-collector/internal/ccMetric"
"io/ioutil"
"log"
"os"
"path/filepath"
"strconv"
"strings"
"time"
"log"
)
const HWMON_PATH = `/sys/class/hwmon`
@@ -28,7 +28,7 @@ type TempCollector struct {
func (m *TempCollector) Init(config json.RawMessage) error {
m.name = "TempCollector"
m.setup()
m.meta = map[string]string{"source" : m.name, "group" : "IPMI", "unit": "degC"}
m.meta = map[string]string{"source": m.name, "group": "IPMI", "unit": "degC"}
if len(config) > 0 {
err := json.Unmarshal(config, &m.config)
if err != nil {
@@ -90,10 +90,10 @@ func (m *TempCollector) Read(interval time.Duration, output chan lp.CCMetric) {
break
}
}
mname := strings.Replace(name, " ", "_", -1)
if !strings.Contains(mname, "temp") {
mname = fmt.Sprintf("temp_%s", mname)
}
mname := strings.Replace(name, " ", "_", -1)
if !strings.Contains(mname, "temp") {
mname = fmt.Sprintf("temp_%s", mname)
}
buffer, err := ioutil.ReadFile(string(file))
if err != nil {
continue
@@ -102,7 +102,7 @@ func (m *TempCollector) Read(interval time.Duration, output chan lp.CCMetric) {
if err == nil {
y, err := lp.New(strings.ToLower(mname), tags, m.meta, map[string]interface{}{"value": int(float64(x) / 1000)}, time.Now())
if err == nil {
log.Print("[", m.name, "] ",y)
log.Print("[", m.name, "] ", y)
output <- y
}
}

2
collectors/topprocsMetric.go
View File

@@ -28,7 +28,7 @@ func (m *TopProcsCollector) Init(config json.RawMessage) error {
var err error
m.name = "TopProcsCollector"
m.tags = map[string]string{"type": "node"}
m.meta = map[string]string{"source" : m.name, "group" : "TopProcs"}
m.meta = map[string]string{"source": m.name, "group": "TopProcs"}
if len(config) > 0 {
err = json.Unmarshal(config, &m.config)
if err != nil {