[horus] Downtime function test results completed (#442)

apache · Oct 9, 2024 · def734b · def734b
1 parent 6118649
commit def734b
Show file tree

Hide file tree

Showing 6 changed files with 14 additions and 30 deletions.
diff --git a/app/horus/base/db/db.go b/app/horus/base/db/db.go
@@ -149,7 +149,7 @@ func GetRecoveryNodeDataInfoDate(day int) ([]NodeDataInfo, error) {
 
 func GetRestartNodeDataInfoDate() ([]NodeDataInfo, error) {
 	var ndi []NodeDataInfo
-	session := db.Where("restart = 0 and repair = 0 and module_name = ?", "node_down")
+	session := db.Where("restart = 0 and repair = 0 and module_name = ?", "nodeDown")
 	err := session.Find(&ndi)
 	return ndi, err
 }

diff --git a/app/horus/cmd/main.go b/app/horus/cmd/main.go
@@ -93,11 +93,13 @@ func main() {
 		return nil
 	})
 	group.Add(func() error {
-		klog.Info("horus node recovery manager start success.")
-		err := horus.RecoveryManager(ctx)
-		if err != nil {
-			klog.Errorf("horus node recovery manager start failed err:%v", err)
-			return err
+		if c.CustomModular.Enabled {
+			klog.Info("horus node recovery manager start success.")
+			err := horus.RecoveryManager(ctx)
+			if err != nil {
+				klog.Errorf("horus node recovery manager start failed err:%v", err)
+				return err
+			}
 		}
 		return nil
 	})

diff --git a/app/horus/core/horuser/node_downtime.go b/app/horus/core/horuser/node_downtime.go
@@ -119,6 +119,7 @@ func (h *Horuser) DownTimeNodes(clusterName, addr string) {
 	}
 
 	msg := fmt.Sprintf("\n【%s】\n【集群：%v】\n【已达到宕机临界点：%v】", h.cc.NodeDownTime.DingTalk.Title, clusterName, len(WithDownNodeIPs))
+
 	newfound := 0
 
 	for nodeName, _ := range WithDownNodeIPs {

diff --git a/app/horus/core/horuser/node_drain.go b/app/horus/core/horuser/node_drain.go
@@ -16,9 +16,7 @@
 package horuser
 
 import (
-	"context"
 	"fmt"
-	corev1 "k8s.io/api/core/v1"
 	v1 "k8s.io/apimachinery/pkg/apis/meta/v1"
 	"k8s.io/klog/v2"
 )
@@ -54,33 +52,16 @@ func (h *Horuser) Drain(nodeName, clusterName string) (err error) {
 				break
 			}
 		}
-		klog.Errorf("node Drain evict pod result items:%d count:%v nodeName:%v\n clusterName:%v\n podName:%v\n podNamespace:%v\n", items+1, count, nodeName, clusterName, pods.Name, pods.Namespace)
 		if ds {
 			continue
 		}
+		klog.Errorf("node Drain evict pod result items:%d count:%v nodeName:%v\n clusterName:%v\n podName:%v\n podNamespace:%v\n", items+1, count, nodeName, clusterName, pods.Name, pods.Namespace)
+
 		err = h.Evict(pods.Name, pods.Namespace, clusterName)
 		if err != nil {
 			klog.Errorf("node Drain evict pod err:%v items:%d count:%v nodeName:%v\n clusterName:%v\n podName:%v\n podNamespace:%v\n", err, items+1, count, nodeName, clusterName, pods.Name, pods.Namespace)
 			return err
 		}
-		err = h.Finalizer(clusterName, pods.Name, pods.Namespace)
-		if err != nil {
-			klog.Errorf("node Drain finalizer pod err:%v items:%d count:%v nodeName:%v\n clusterName:%v\n podName:%v\n podNamespace:%v\n", err, items+1, count, nodeName, clusterName, pods.Name, pods.Namespace)
-			return err
-		}
-
-		var oldPod *corev1.Pod
-		var _ = h.Terminating(clusterName, oldPod)
-		newPod, _ := kubeClient.CoreV1().Pods(oldPod.Namespace).Get(context.Background(), oldPod.Name, v1.GetOptions{})
-		if newPod == nil {
-			return err
-		}
-		if newPod.UID != oldPod.UID {
-			return err
-		}
-		if newPod.DeletionTimestamp.IsZero() {
-			return err
-		}
 	}
 	return nil
 }
diff --git a/app/horus/core/horuser/node_restart.go b/app/horus/core/horuser/node_restart.go
@@ -69,7 +69,7 @@ func (h *Horuser) TryRestart(node db.NodeDataInfo) {
 	klog.Infof("RestartMarker result success:%v", success)
 
 	if success {
-		msg := fmt.Sprintf("\n【等待宕机节点腾空后重启】\n【节点:%v】\n【日期:%v】\n【集群:%v】\n", node.NodeName, node.FirstDate, node.ClusterName)
+		msg := fmt.Sprintf("\n【宕机节点等待腾空后重启】\n【节点:%v】\n【日期:%v】\n【集群:%v】\n", node.NodeName, node.FirstDate, node.ClusterName)
 		alerter.DingTalkSend(h.cc.NodeDownTime.DingTalk, msg)
 
 		cmd := exec.Command("/bin/bash", "core/horuser/restart.sh", node.NodeIP, h.cc.NodeDownTime.AllSystemUser, h.cc.NodeDownTime.AllSystemPassword)

diff --git a/manifests/horus/horus.yaml b/manifests/horus/horus.yaml
@@ -25,7 +25,7 @@ kubeMultiple:
   cluster: config.1
 
 promMultiple:
-  cluster: http://192.168.15.134:31974
+  cluster: http://192.168.15.133:31974
 
 nodeRecovery:
   dayNumber: 1
@@ -64,7 +64,7 @@ customModular:
     title: "自定义通知"
 
 nodeDownTime:
-  enabled: true
+  enabled: false
   intervalSecond: 15
   promQueryTimeSecond: 60
   abnormalityQL: