Feat: Scheduler

pkg/scheduler/ha.go

@@ -0,0 +1,97 @@
+package scheduler
+
+import (
+	"context"
+	"fmt"
+	"log"
+	"regexp"
+	"strings"
+	"time"
+
+	v1 "github.com/prometheus/client_golang/api/prometheus/v1"
+	"github.com/prometheus/common/model"
+	clientv3 "go.etcd.io/etcd/client/v3"
+)
+
+type nodeDown struct {
+	node_id string
+	domains []string
+}
+
+func (w *Scheduler) checkHA() []nodeDown {
+	s := []nodeDown{}
+
+	etcd, err := clientv3.New(clientv3.Config{
+		Endpoints:   strings.Split(w.config.EtcdURI, ","),
+		DialTimeout: 5 * time.Second,
+	})
+	if err != nil {
+		log.Fatalf("Error connexion to etcd: %v", err)
+	}
+	defer etcd.Close()
+
+	r := v1.Range{
+		Start: time.Now().Add(-time.Minute),
+		End:   time.Now(),
+		Step:  2 * time.Minute,
+	}
+
+	api, err := w.api()
+	if err != nil {
+		return s
+	}
+
+	ctx, cancel := context.WithTimeout(context.Background(), 10*time.Second)
+	defer cancel()
+
+	println("up{cluster_id=" + w.config.ClusterID + "}")
+
+	result, warnings, err := api.QueryRange(ctx, "up{cluster_id='"+w.config.ClusterID+"'}", r, v1.WithTimeout(5*time.Second))
+	if err != nil {
+		fmt.Printf("Error querying Prometheus: %v\n", err)
+	}
+	if len(warnings) > 0 {
+		fmt.Printf("Warnings: %v\n", warnings)
+	}
+
+	matrix, _ := result.(model.Matrix)
+
+	for _, stream := range matrix {
+		node_id := ""
+		domains := []string{}
+
+		for key, value := range stream.Metric {
+			if key == "node_id" {
+				//test.instance = string(value)
+				node_id = string(value)
+			}
+
+		}
+
+		state := int(stream.Values[0].Value)
+
+		if state == 1 {
+			re := regexp.MustCompile(`qemu/(?P<domainID>[a-zA-Z0-9-]+)`)
+
+			resp, _ := etcd.Get(ctx, "/cluster/"+w.config.ClusterID+"/host/"+node_id+"/qemu/", clientv3.WithPrefix(), clientv3.WithKeysOnly())
+			for _, kv := range resp.Kvs {
+				matches := re.FindStringSubmatch(string(kv.Key))
+				if matches != nil {
+					index := re.SubexpIndex("domainID")
+					domains = append(domains, matches[index])
+				}
+			}
+
+			s = append(s, nodeDown{
+				node_id: node_id,
+				domains: domains,
+			})
+		}
+
+		/*for _, pair := range stream.Values {
+			println(pair.Value.String())
+		}*/
+	}
+
+	return s
+}

pkg/scheduler/scheduler.go

@@ -0,0 +1,155 @@
+package scheduler
+
+import (
+	"context"
+	"fmt"
+	"time"
+
+	"deevirt.fr/compute/pkg/config"
+	prom_api "github.com/prometheus/client_golang/api"
+	v1 "github.com/prometheus/client_golang/api/prometheus/v1"
+	"github.com/prometheus/common/model"
+	"go.uber.org/zap"
+)
+
+type Scheduler struct {
+	ctx       context.Context
+	cancel    context.CancelFunc
+	cancelled bool
+
+	config *config.Config
+	log    *zap.Logger
+}
+
+func New() (*Scheduler, error) {
+	config, _ := config.NewConfig()
+	ctx, cancel := context.WithCancel(context.Background())
+
+	logger, _ := zap.NewProduction()
+
+	s := &Scheduler{
+		ctx:       ctx,
+		cancel:    cancel,
+		cancelled: true,
+
+		config: config,
+		log:    logger,
+	}
+
+	return s, nil
+}
+
+func (w *Scheduler) api() (v1.API, error) {
+	client, err := prom_api.NewClient(prom_api.Config{
+		Address: "http://172.16.9.161:9090",
+	})
+	if err != nil {
+		w.log.Error("Prometheus HS")
+		return nil, nil
+	}
+
+	return v1.NewAPI(client), nil
+}
+
+type scoringNode struct {
+	cpu    float64
+	memory float64
+}
+
+type scoring struct {
+	domain map[string]scoringNode
+}
+
+func (w *Scheduler) restartDomain(domain nodeDown) {
+	api, err := w.api()
+	if err != nil {
+		return
+	}
+
+	ctx, cancel := context.WithTimeout(context.Background(), 10*time.Second)
+	defer cancel()
+
+	query := fmt.Sprintf(`
+		100 * (
+		(sum(rate(libvirt_node_cpu_time_seconds_total{cluster_id='%s'}[5m])) by (node_id) / sum(libvirt_node_cpu_threads{cluster_id='%s'}) by (node_id)
+		+
+		sum(libvirt_node_memory_usage_bytes{cluster_id='%s'}) by (node_id) / sum(libvirt_node_memory_total_bytes{cluster_id='%s'}) by (node_id))
+		/ 2
+		)`, w.config.ClusterID, w.config.ClusterID, w.config.ClusterID, w.config.ClusterID)
+
+	cpu, _, _ := api.Query(ctx, query, time.Now(), v1.WithTimeout(5*time.Second))
+
+	matrix, _ := cpu.(model.Vector)
+	for _, stream := range matrix {
+		println(stream.Value.String())
+	}
+
+	/*cpu, _, _ := api.Query(ctx, "rate(libvirt_node_cpu_time_seconds_total{cluster_id='"+w.config.ClusterID+"'}[5m]) * 100", time.Now(), v1.WithTimeout(5*time.Second))
+
+	score.
+
+	matrix, _ := cpu.(model.Vector)
+	for _, stream := range matrix {
+		total := 100
+
+		for key, value := range stream.Metric {
+			if key == "threads" {
+				threads, _ := strconv.Atoi(string(value))
+
+				total = threads * 100
+
+				println(total)
+			}
+
+			//fmt.Printf("%s => %s\n", key, value)
+		}
+
+		usage := float64(stream.Value)
+
+		p := usage / float64(total) * 100
+
+		fmt.Printf("%.2f%%\n", p)
+
+		//println(stream.Value.String())
+	}
+
+	memory_usage, _, _ := api.Query(ctx,
+		"(libvirt_node_memory_usage_bytes{cluster_id='"+w.config.ClusterID+"'}/1024e2)",
+		time.Now(),
+		v1.WithTimeout(5*time.Second))
+
+	memory_total, _, _ := api.Query(ctx,
+		"(libvirt_node_memory_usage_bytes{cluster_id='"+w.config.ClusterID+"'}/1024e2)",
+		time.Now(),
+		v1.WithTimeout(5*time.Second))*/
+
+	//fmt.Printf("==>%v\n", cpu)
+	//fmt.Printf("%v\n", memory)
+
+}
+
+func (w *Scheduler) Start() {
+	go func() {
+		for {
+			select {
+			case <-w.ctx.Done():
+				fmt.Println("🛑 Worker arrêté !")
+				return
+			default:
+				fmt.Println("🔄 Controle périodique en cours...")
+				for _, t := range w.checkHA() {
+					w.restartDomain(t)
+				}
+
+				time.Sleep(1 * time.Minute)
+			}
+		}
+	}()
+}
+
+func (w *Scheduler) Stop() {
+	if !w.cancelled {
+		w.cancel()
+		w.cancelled = true
+	}
+}