1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22

// Lock required for leader election
	rl, err := resourcelock.New(
               //在config會指定說要新增哪一種resource loack
               c.ComponentConfig.Generic.LeaderElection.ResourceLock,
		//在config會指定說resource lock要綁在哪一個namespace
		c.ComponentConfig.Generic.LeaderElection.ResourceNamespace,
               //在config會指定說resource lock的名稱
		c.ComponentConfig.Generic.LeaderElection.ResourceName,
               //帶入建立configmap,endpoint client object
		c.LeaderElectionClient.CoreV1(),
               //帶入建立leases client object
		c.LeaderElectionClient.CoordinationV1(),
                //Resource lock 內容
		resourcelock.ResourceLockConfig{
                //Resource lock 內容 ID 目前會以hostname_uuid
			Identity:      id,
                //EventRecorder 會紀錄 resource lock 事件
			EventRecorder: c.EventRecorder,
		})
	if err != nil {
		klog.Fatalf("error creating lock: %v", err)
	}

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20

type Interface interface {
	// Get returns the LeaderElectionRecord
	Get(ctx context.Context) (*LeaderElectionRecord, []byte, error)

	// Create attempts to create a LeaderElectionRecord
	Create(ctx context.Context, ler LeaderElectionRecord) error

	// Update will update and existing LeaderElectionRecord
	Update(ctx context.Context, ler LeaderElectionRecord) error

	// RecordEvent is used to record events
	RecordEvent(string)

	// Identity will return the locks Identity
	Identity() string

	// Describe is used to convert details on current resource lock
	// into a string
	Describe() string
}

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20

type ConfigMapLock struct {
	// ConfigMapMeta should contain a Name and a Namespace of a
	// ConfigMapMeta object that the LeaderElector will attempt to lead.
	ConfigMapMeta metav1.ObjectMeta
	Client        corev1client.ConfigMapsGetter
	LockConfig    ResourceLockConfig
	cm            *v1.ConfigMap
}

...

type LeaseLock struct {
	// LeaseMeta should contain a Name and a Namespace of a
	// LeaseMeta object that the LeaderElector will attempt to lead.
	LeaseMeta  metav1.ObjectMeta
	Client     coordinationv1client.LeasesGetter
	LockConfig ResourceLockConfig
	lease      *coordinationv1.Lease
}
...

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47

// Manufacture will create a lock of a given type according to the input parameters
func New(lockType string, ns string, name string, coreClient corev1.CoreV1Interface, coordinationClient coordinationv1.CoordinationV1Interface, rlc ResourceLockConfig) (Interface, error) {
	endpointsLock := &EndpointsLock{
		EndpointsMeta: metav1.ObjectMeta{
			Namespace: ns,
			Name:      name,
		},
		Client:     coreClient,
		LockConfig: rlc,
	}
	configmapLock := &ConfigMapLock{
		ConfigMapMeta: metav1.ObjectMeta{
			Namespace: ns,
			Name:      name,
		},
		Client:     coreClient,
		LockConfig: rlc,
	}
	leaseLock := &LeaseLock{
		LeaseMeta: metav1.ObjectMeta{
			Namespace: ns,
			Name:      name,
		},
		Client:     coordinationClient,
		LockConfig: rlc,
	}
	switch lockType {
	case EndpointsResourceLock:
		return endpointsLock, nil
	case ConfigMapsResourceLock:
		return configmapLock, nil
	case LeasesResourceLock:
		return leaseLock, nil
	case EndpointsLeasesResourceLock:
		return &MultiLock{
			Primary:   endpointsLock,
			Secondary: leaseLock,
		}, nil
	case ConfigMapsLeasesResourceLock:
		return &MultiLock{
			Primary:   configmapLock,
			Secondary: leaseLock,
		}, nil
	default:
		return nil, fmt.Errorf("Invalid lock-type %s", lockType)
	}
}

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20

// Try and become the leader and start cloud controller manager loops
        //使用 leaderelection package 裡的 RunOrDie function 會按照你設定的 LeaseDuration, RenewDeadline , RetryPeriod 三個時間嘗試成為領導人。
	leaderelection.RunOrDie(context.TODO(), 
            leaderelection.LeaderElectionConfig{
		Lock:          rl,
		LeaseDuration: c.ComponentConfig.Generic.LeaderElection.LeaseDuration.Duration,
		RenewDeadline: c.ComponentConfig.Generic.LeaderElection.RenewDeadline.Duration,
		RetryPeriod:   c.ComponentConfig.Generic.LeaderElection.RetryPeriod.Duration,
		Callbacks: leaderelection.LeaderCallbacks{
                        // 在 controller manger 裡面就是執行一個run 的function 當作controller 的進入點
                        // 也可以理解為controller manger 程式的進入點
			OnStartedLeading: run,
			OnStoppedLeading: func() {
				klog.Fatalf("leaderelection lost")
			},
		},
		WatchDog: electionChecker,
		Name:     "cloud-controller-manager",
	})

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16

// RunOrDie starts a client with the provided config or panics if the config
// fails to validate. RunOrDie blocks until leader election loop is
// stopped by ctx or it has stopped holding the leader lease
func RunOrDie(ctx context.Context, lec LeaderElectionConfig) {
        //這裡很簡單的去new一個LeaderElector物件，主要是檢查使用者輸入的config有沒有問題。
	le, err := NewLeaderElector(lec)
	if err != nil {
		panic(err)
	}
        //todo還不是很清楚watch dog的作用
	if lec.WatchDog != nil {
		lec.WatchDog.SetLeaderElection(le)
	}
        //主要在這個run function 等等會來解密這個function 做了什麼事
	le.Run(ctx)
}

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21

// Run starts the leader election loop. Run will not return
// before leader election loop is stopped by ctx or it has
// stopped holding the leader lease
func (le *LeaderElector) Run(ctx context.Context) {
        //kubernetes 內置處理 crash 的方法本篇不討論。
	defer runtime.HandleCrash()
        //當使用者觸發了 context cancel 時會呼叫 OnStoppedLeading 方法。
	defer func() {
		le.config.Callbacks.OnStoppedLeading()
	}()
        //在這裡會試圖取得會是創建 resource lock,沒有拿到資源鎖的話就會卡在這個function裡面喔！
	if !le.acquire(ctx) {
		return // ctx signalled done
	}
	ctx, cancel := context.WithCancel(ctx)
	defer cancel()
        //啟動一個 goroutine 去執行 controller manger 的進入點 run function
	go le.config.Callbacks.OnStartedLeading(ctx)
        //執行 resource lock 的續約動作。
	le.renew(ctx)
}

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31

// acquire loops calling tryAcquireOrRenew and returns true immediately when tryAcquireOrRenew succeeds.
// Returns false if ctx signals done.
func (le *LeaderElector) acquire(ctx context.Context) bool {
	ctx, cancel := context.WithCancel(ctx)
	defer cancel()
	succeeded := false
	desc := le.config.Lock.Describe()
	klog.Infof("attempting to acquire leader lease %v...", desc)
        // kubernetes封裝了一個類似timmer的function比較，細節不去探討。
        // 我們只要知道他是一個定時觸發的function就好。
	wait.JitterUntil(func() {
                //嘗試獲取 resource lock 或是續約resource lock
		succeeded = le.tryAcquireOrRenew(ctx)
                //不知道實際用途，追code感覺是有人定義Callbacks.OnNewLeader的話會callback告知 leader換了。
		le.maybeReportTransition()
                // 獲取 resource lock 失敗，會等下次timmer到了再嘗試獲取租約。
		if !succeeded {
			klog.V(4).Infof("failed to acquire lease %v", desc)
			return
		}
                //獲取租約成功紀錄事件
		le.config.Lock.RecordEvent("became leader")
                //獲取租約成功紀錄事件
		le.metrics.leaderOn(le.config.Name)
                //獲取租約成功紀錄事件
		klog.Infof("successfully acquired lease %v", desc)
                //只要獲取成功後就不用再跑timmer，（這個Timmer用意就是讓沒有拿到resource lock的人可以定時去拿拿看）
		cancel()
	}, le.config.RetryPeriod, JitterFactor, true, ctx.Done())
	return succeeded
}

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41

// renew loops calling tryAcquireOrRenew and returns immediately when tryAcquireOrRenew fails or ctx signals done.
func (le *LeaderElector) renew(ctx context.Context) {
	ctx, cancel := context.WithCancel(ctx)
	defer cancel()
        // kubernetes封裝了一個類似timmer的function比較，細節不去探討。
        // 可能有人會問跟wait.JitterUntil有什麼區別，這邊之後會另開戰場去討論（先挖洞給自己跳）
	wait.Until(func() {
		timeoutCtx, timeoutCancel := context.WithCancel(ctx)
		defer timeoutCancel()
                // PollImmediateUntil 會等 ConditionFunc 執行完畢
                // 若是 ConditionFunc 執行失敗(是false)就會等 RetryPeriod 時間在執行一次。
                // 如果 ConditionFunc 執行的結果是error或是true的話就不會重新嘗試 ConditionFunc 。
		err := wait.PollImmediateUntil(le.config.RetryPeriod, func() (bool, error) {
                // ConditionFunc
                // 嘗試建立resource lock 或是更新resource lock 租約
			return le.tryAcquireOrRenew(timeoutCtx), nil
		}, timeoutCtx.Done())
                //不知道實際用途，追code感覺是有人定義Callbacks.OnNewLeader的話會callback告知 leader換了。
		le.maybeReportTransition()
                // 拿到 resource lock 以configmaplock 為例 desc就會等於 configmaplock namespace/configmap name
		desc := le.config.Lock.Describe()
                // 這裡代表更新了租約成功就跳出去了，等待下一次timmer的呼叫。
		if err == nil {
			klog.V(5).Infof("successfully renewed lease %v", desc)
			return
		}
                //租約更新失敗紀錄
		le.config.Lock.RecordEvent("stopped leading")
                //租約更新失敗紀錄
		le.metrics.leaderOff(le.config.Name)
                //租約更新失敗紀錄
		klog.Infof("failed to renew lease %v: %v", desc, err)
                //再也不透過timmer定期更新租約
		cancel()
	}, le.config.RetryPeriod, ctx.Done())

	// if we hold the lease, give it up
	if le.config.ReleaseOnCancel {
		le.release()
	}
}