Modularize the whole thing (#16)

* Use channels, add a metric router, split up configuration and use extended version of Influx line protocol internally

* Use central timer for collectors and router. Add expressions to router

* Add expression to router config

* Update entry points

* Start with README

* Update README for CCMetric

* Formatting

* Update README.md

* Add README for MultiChanTicker

* Add README for MultiChanTicker

* Update README.md

* Add README to metric router

* Update main README

* Remove SinkEntity type

* Update README for sinks

* Update go files

* Update README for receivers

* Update collectors README

* Update collectors README

* Use seperate page per collector

* Fix for tempstat page

* Add docs for customcmd collector

* Add docs for ipmistat collector

* Add docs for topprocs collector

* Update customCmdMetric.md

* Use seconds when calculating LIKWID metrics

* Add IB metrics ib_recv_pkts and ib_xmit_pkts

* Drop domain part of host name

* Updated to latest stable version of likwid

* Define source code dependencies in Makefile

* Add GPFS / IBM Spectrum Scale collector

* Add vet and staticcheck make targets

* Add vet and staticcheck make targets

* Avoid go vet warning:
struct field tag `json:"..., omitempty"` not compatible with reflect.StructTag.Get: suspicious space in struct tag value
struct field tag `json:"...", omitempty` not compatible with reflect.StructTag.Get: key:"value" pairs not separated by spaces

* Add sample collector to README.md

* Add CPU frequency collector

* Avoid staticcheck warning: redundant return statement

* Avoid staticcheck warning: unnecessary assignment to the blank identifier

* Simplified code

* Add CPUFreqCollectorCpuinfo
a metric collector to measure the current frequency of the CPUs
as obtained from /proc/cpuinfo
Only measure on the first hyperthread

* Add collector for NFS clients

* Move publication of metrics into Flush() for NatsSink

* Update GitHub actions

* Refactoring

* Avoid vet warning: Println arg list ends with redundant newline

* Avoid vet warning struct field commands has json tag but is not exported

* Avoid vet warning: return copies lock value.

* Corrected typo

* Refactoring

* Add go sources in internal/...

* Bad separator in Makefile

* Fix Infiniband collector

Co-authored-by: Holger Obermaier <40787752+ho-ob@users.noreply.github.com>
This commit is contained in:
Thomas Gruber
2022-01-25 15:37:43 +01:00
committed by GitHub
parent 222862af32
commit 200af84c54
60 changed files with 2596 additions and 1105 deletions

View File

@@ -1,35 +1,44 @@
This folder contains the receivers for the cc-metric-collector.
# CCMetric receivers
# `metricReceiver.go`
The base class/configuration is located in `metricReceiver.go`.
This folder contains the ReceiveManager and receiver implementations for the cc-metric-collector.
# Receivers
* `natsReceiver.go`: Receives metrics from the Nats transport system in Influx line protocol encoding. The database name is used as subscription subject for the NATS messages. It uses https://github.com/nats-io/nats.go
# Configuration
# Installation
Nothing to do, all receivers are pure Go code
# Receiver configuration
The configuration file for the receivers is a list of configurations. The `type` field in each specifies which receiver to initialize.
```json
"receiver": {
[
{
"type": "nats",
"address": "nats://my-url"
"address": "nats://my-url",
"port" : "4222",
"database": "testcluster"
},
}
]
```
## `nats`
The receiver connects to `address` and `port` and subscribes itself for all messages with topic `database`. The default port is `4222`.
## Type `nats`
```json
{
"type": "nats",
"address": "<nats-URI or hostname>",
"port" : "<portnumber>",
"database": "<subscribe topic>"
}
```
The `nats` receiver subscribes to the topic `database` and listens on `address` and `port` for metrics in the InfluxDB line protocol.
# Contributing own receivers
A receiver contains three functions and is derived from the type `Receiver` (in `metricReceiver.go`):
* `Init(config ReceiverConfig) error`
* `Start() error`
* `Close()`
* `Name() string`
* `SetSink(sink chan ccMetric.CCMetric)`
The data structures should be set up in `Init()` like opening a file or server connection. The `Start()` function should either start a go routine or issue some other asynchronous mechanism for receiving metrics. The `Close()` function should tear down anything created in `Init()`.
Finally, the receiver needs to be registered in the `metric-collector.go`. There is a list of receivers called `Receivers` which is a map (string -> pointer to receiver). Add a new entry with a descriptive name and the new receiver.
Finally, the receiver needs to be registered in the `receiveManager.go`. There is a list of receivers called `AvailableReceivers` which is a map (`receiver_type_string` -> `pointer to Receiver interface`). Add a new entry with a descriptive name and the new receiver.

View File

@@ -2,30 +2,41 @@ package receivers
import (
// "time"
s "github.com/ClusterCockpit/cc-metric-collector/sinks"
lp "github.com/ClusterCockpit/cc-metric-collector/internal/ccMetric"
influx "github.com/influxdata/line-protocol"
)
type ReceiverConfig struct {
Addr string `json:"address"`
Port string `json:"port"`
Database string `json:"database"`
Type string `json:"type"`
Addr string `json:"address"`
Port string `json:"port"`
Database string `json:"database"`
Organization string `json:"organization,omitempty"`
Type string `json:"type"`
}
type Receiver struct {
type receiver struct {
name string
addr string
port string
database string
organization string
sink s.SinkFuncs
sink chan lp.CCMetric
}
type ReceiverFuncs interface {
Init(config ReceiverConfig, sink s.SinkFuncs) error
type Receiver interface {
Init(config ReceiverConfig) error
Start()
Close()
Name() string
SetSink(sink chan lp.CCMetric)
}
func (r *receiver) Name() string {
return r.name
}
func (r *receiver) SetSink(sink chan lp.CCMetric) {
r.sink = sink
}
func Tags2Map(metric influx.Metric) map[string]string {

View File

@@ -2,56 +2,68 @@ package receivers
import (
"errors"
s "github.com/ClusterCockpit/cc-metric-collector/sinks"
lp "github.com/influxdata/line-protocol"
"fmt"
lp "github.com/ClusterCockpit/cc-metric-collector/internal/ccMetric"
influx "github.com/influxdata/line-protocol"
nats "github.com/nats-io/nats.go"
"log"
"time"
)
type NatsReceiverConfig struct {
Addr string `json:"address"`
Port string `json:"port"`
Database string `json:"database"`
}
type NatsReceiver struct {
Receiver
receiver
nc *nats.Conn
handler *lp.MetricHandler
parser *lp.Parser
handler *influx.MetricHandler
parser *influx.Parser
meta map[string]string
config ReceiverConfig
}
var DefaultTime = func() time.Time {
return time.Unix(42, 0)
}
func (r *NatsReceiver) Init(config ReceiverConfig, sink s.SinkFuncs) error {
if len(config.Addr) == 0 ||
len(config.Port) == 0 ||
len(config.Database) == 0 {
func (r *NatsReceiver) Init(config ReceiverConfig) error {
r.name = "NatsReceiver"
r.config = config
if len(r.config.Addr) == 0 ||
len(r.config.Port) == 0 ||
len(r.config.Database) == 0 {
return errors.New("Not all configuration variables set required by NatsReceiver")
}
r.addr = config.Addr
r.meta = map[string]string{"source": r.name}
r.addr = r.config.Addr
if len(r.addr) == 0 {
r.addr = nats.DefaultURL
}
r.port = config.Port
r.port = r.config.Port
if len(r.port) == 0 {
r.port = "4222"
}
log.Print("Init NATS Receiver")
nc, err := nats.Connect(r.addr)
log.Print("[NatsReceiver] INIT")
uri := fmt.Sprintf("%s:%s", r.addr, r.port)
nc, err := nats.Connect(uri)
if err == nil {
r.database = config.Database
r.sink = sink
r.database = r.config.Database
r.nc = nc
} else {
log.Print(err)
r.nc = nil
return err
}
r.handler = lp.NewMetricHandler()
r.parser = lp.NewParser(r.handler)
r.handler = influx.NewMetricHandler()
r.parser = influx.NewParser(r.handler)
r.parser.SetTimeFunc(DefaultTime)
return err
}
func (r *NatsReceiver) Start() {
log.Print("Start NATS Receiver")
log.Print("[NatsReceiver] START")
r.nc.Subscribe(r.database, r._NatsReceive)
}
@@ -59,9 +71,13 @@ func (r *NatsReceiver) _NatsReceive(m *nats.Msg) {
metrics, err := r.parser.Parse(m.Data)
if err == nil {
for _, m := range metrics {
y, err := lp.New(m.Name(), Tags2Map(m), Fields2Map(m), m.Time())
if err == nil {
r.sink.Write(y)
y := lp.FromInfluxMetric(m)
for k, v := range r.meta {
y.AddMeta(k, v)
}
//y, err := lp.New(m.Name(), Tags2Map(m), r.meta, Fields2Map(m), m.Time())
if r.sink != nil {
r.sink <- y
}
}
}
@@ -69,7 +85,7 @@ func (r *NatsReceiver) _NatsReceive(m *nats.Msg) {
func (r *NatsReceiver) Close() {
if r.nc != nil {
log.Print("Close NATS Receiver")
log.Print("[NatsReceiver] CLOSE")
r.nc.Close()
}
}

153
receivers/receiveManager.go Normal file
View File

@@ -0,0 +1,153 @@
package receivers
import (
"encoding/json"
lp "github.com/ClusterCockpit/cc-metric-collector/internal/ccMetric"
"log"
"os"
"sync"
)
var AvailableReceivers = map[string]Receiver{
"nats": &NatsReceiver{},
}
type receiveManager struct {
inputs []Receiver
output chan lp.CCMetric
done chan bool
wg *sync.WaitGroup
config []ReceiverConfig
}
type ReceiveManager interface {
Init(wg *sync.WaitGroup, receiverConfigFile string) error
AddInput(rawConfig json.RawMessage) error
AddOutput(output chan lp.CCMetric)
Start()
Close()
}
func (rm *receiveManager) Init(wg *sync.WaitGroup, receiverConfigFile string) error {
rm.inputs = make([]Receiver, 0)
rm.output = nil
rm.done = make(chan bool)
rm.wg = wg
rm.config = make([]ReceiverConfig, 0)
configFile, err := os.Open(receiverConfigFile)
if err != nil {
log.Print(err.Error())
return err
}
defer configFile.Close()
jsonParser := json.NewDecoder(configFile)
var rawConfigs []json.RawMessage
err = jsonParser.Decode(&rawConfigs)
if err != nil {
log.Print(err.Error())
return err
}
for _, raw := range rawConfigs {
log.Print("[ReceiveManager] ", string(raw))
rm.AddInput(raw)
// if _, found := AvailableReceivers[k.Type]; !found {
// log.Print("[ReceiveManager] SKIP Config specifies unknown receiver 'type': ", k.Type)
// continue
// }
// r := AvailableReceivers[k.Type]
// err = r.Init(k)
// if err != nil {
// log.Print("[ReceiveManager] SKIP Receiver ", k.Type, " cannot be initialized: ", err.Error())
// continue
// }
// rm.inputs = append(rm.inputs, r)
}
return nil
}
func (rm *receiveManager) Start() {
rm.wg.Add(1)
for _, r := range rm.inputs {
log.Print("[ReceiveManager] START ", r.Name())
r.Start()
}
log.Print("[ReceiveManager] STARTED\n")
// go func() {
// for {
//ReceiveManagerLoop:
// select {
// case <- rm.done:
// log.Print("ReceiveManager done\n")
// rm.wg.Done()
// break ReceiveManagerLoop
// default:
// for _, c := range rm.inputs {
//ReceiveManagerInputLoop:
// select {
// case <- rm.done:
// log.Print("ReceiveManager done\n")
// rm.wg.Done()
// break ReceiveManagerInputLoop
// case p := <- c:
// log.Print("ReceiveManager: ", p)
// rm.output <- p
// default:
// }
// }
// }
// }
// }()
// for _, r := range rm.inputs {
// r.Close()
// }
}
func (rm *receiveManager) AddInput(rawConfig json.RawMessage) error {
var config ReceiverConfig
err := json.Unmarshal(rawConfig, &config)
if err != nil {
log.Print("[ReceiveManager] SKIP ", config.Type, " JSON config error: ", err.Error())
log.Print(err.Error())
return err
}
if _, found := AvailableReceivers[config.Type]; !found {
log.Print("[ReceiveManager] SKIP ", config.Type, " unknown receiver: ", err.Error())
return err
}
r := AvailableReceivers[config.Type]
err = r.Init(config)
if err != nil {
log.Print("[ReceiveManager] SKIP ", r.Name(), " initialization failed: ", err.Error())
return err
}
rm.inputs = append(rm.inputs, r)
rm.config = append(rm.config, config)
return nil
}
func (rm *receiveManager) AddOutput(output chan lp.CCMetric) {
rm.output = output
for _, r := range rm.inputs {
r.SetSink(rm.output)
}
}
func (rm *receiveManager) Close() {
for _, r := range rm.inputs {
log.Print("[ReceiveManager] CLOSE ", r.Name())
r.Close()
}
rm.wg.Done()
log.Print("[ReceiveManager] CLOSE\n")
log.Print("[ReceiveManager] EXIT\n")
}
func New(wg *sync.WaitGroup, receiverConfigFile string) (ReceiveManager, error) {
r := &receiveManager{}
err := r.Init(wg, receiverConfigFile)
if err != nil {
return nil, err
}
return r, err
}