分析kubernetes中的事件机制

silenceper · · 533 次点击 · · 开始浏览    
这是一个创建于 的文章,其中的信息可能已经有所发展或是发生改变。

我们通过 kubectl describe [资源] 命令,可以在看到Event输出,并且经常依赖event进行问题定位,从event中可以分析整个POD的运行轨迹,为服务的客观测性提供数据来源,由此可见,event在Kubernetes中起着举足轻重的作用。

event展示

event并不只是kubelet中都有的,关于event的操作被封装在client-go/tools/record包,我们完全可以在写入自定义的event。

现在让我们来一步步揭开event的面纱。

Event定义

其实event也是一个资源对象,并且通过apiserver将event存储在etcd中,所以我们也可以通过 kubectl get event 命令查看对应的event对象。

以下是一个event的yaml文件:

apiVersion: v1
count: 1
eventTime: null
firstTimestamp: "2020-03-02T13:08:22Z"
involvedObject:
  apiVersion: v1
  kind: Pod
  name: example-foo-d75d8587c-xsf64
  namespace: default
  resourceVersion: "429837"
  uid: ce611c62-6c1a-4bd8-9029-136a1adf7de4
kind: Event
lastTimestamp: "2020-03-02T13:08:22Z"
message: Pod sandbox changed, it will be killed and re-created.
metadata:
  creationTimestamp: "2020-03-02T13:08:30Z"
  name: example-foo-d75d8587c-xsf64.15f87ea1df862b64
  namespace: default
  resourceVersion: "479466"
  selfLink: /api/v1/namespaces/default/events/example-foo-d75d8587c-xsf64.15f87ea1df862b64
  uid: 9fe6f72a-341d-4c49-960b-e185982d331a
reason: SandboxChanged
reportingComponent: ""
reportingInstance: ""
source:
  component: kubelet
  host: minikube
type: Normal复制代码


主要字段说明:**

  • involvedObject: 触发event的资源类型
  • lastTimestamp:最后一次触发的时间
  • message:事件说明
  • metadata :event的元信息,name,namespace等
  • reason:event的原因
  • source:上报事件的来源,比如kubelet中的某个节点
  • type:事件类型,Normal或Warning

event字段定义可以看这里:types.go#L5078

接下来我们来看看,整个event是如何下入的。

写入事件

1、这里以kubelet为例,看看是如何进行事件写入的

2、文中代码以Kubernetes 1.17.3为例进行分析

先以一幅图来看下整个的处理流程event处理过程

创建操作事件的客户端:
kubelet/app/server.go#L461

// makeEventRecorder sets up kubeDeps.Recorder if it's nil. It's a no-op otherwise.
func makeEventRecorder(kubeDeps *kubelet.Dependencies, nodeName types.NodeName) {
    if kubeDeps.Recorder != nil {
        return
    }
    //事件广播
    eventBroadcaster := record.NewBroadcaster()
    //创建EventRecorder
    kubeDeps.Recorder = eventBroadcaster.NewRecorder(legacyscheme.Scheme, v1.EventSource{Component: componentKubelet, Host: string(nodeName)})
    //发送event至log输出
    eventBroadcaster.StartLogging(klog.V(3).Infof)
    if kubeDeps.EventClient != nil {
        klog.V(4).Infof("Sending events to api server.")
        //发送event至apiserver
        eventBroadcaster.StartRecordingToSink(&v1core.EventSinkImpl{Interface: kubeDeps.EventClient.Events("")})
    } else {
        klog.Warning("No api server defined - no events will be sent to API server.")
    }
}复制代码

通过 makeEventRecorder 创建了 EventRecorder 实例,这是一个事件广播器,通过它提供了StartLogging和StartRecordingToSink两个事件处理函数,分别将event发送给log和apiserver。
NewRecorder创建了 EventRecorder 的实例,它提供了 EventEventf 等方法供事件记录。

EventBroadcaster

我们来看下EventBroadcaster接口定义:event.go#L113

// EventBroadcaster knows how to receive events and send them to any EventSink, watcher, or log.
type EventBroadcaster interface {
    //
    StartEventWatcher(eventHandler func(*v1.Event)) watch.Interface
    StartRecordingToSink(sink EventSink) watch.Interface
    StartLogging(logf func(format string, args ...interface{})) watch.Interface
    NewRecorder(scheme *runtime.Scheme, source v1.EventSource) EventRecorder

    Shutdown()
}复制代码

具体实现是通过 eventBroadcasterImpl struct来实现了各个方法。

其中StartLogging 和 StartRecordingToSink 其实就是完成了对事件的消费,EventRecorder实现对事件的写入,中间通过channel实现了生产者消费者模型。

EventRecorder

我们先来看下EventRecorder 接口定义:event.go#L88,提供了一下4个方法

// EventRecorder knows how to record events on behalf of an EventSource.
type EventRecorder interface {
    // Event constructs an event from the given information and puts it in the queue for sending.
    // 'object' is the object this event is about. Event will make a reference-- or you may also
    // pass a reference to the object directly.
    // 'type' of this event, and can be one of Normal, Warning. New types could be added in future
    // 'reason' is the reason this event is generated. 'reason' should be short and unique; it
    // should be in UpperCamelCase format (starting with a capital letter). "reason" will be used
    // to automate handling of events, so imagine people writing switch statements to handle them.
    // You want to make that easy.
    // 'message' is intended to be human readable.
    //
    // The resulting event will be created in the same namespace as the reference object.
    Event(object runtime.Object, eventtype, reason, message string)

    // Eventf is just like Event, but with Sprintf for the message field.
    Eventf(object runtime.Object, eventtype, reason, messageFmt string, args ...interface{})

    // PastEventf is just like Eventf, but with an option to specify the event's 'timestamp' field.
    PastEventf(object runtime.Object, timestamp metav1.Time, eventtype, reason, messageFmt string, args ...interface{})

    // AnnotatedEventf is just like eventf, but with annotations attached
    AnnotatedEventf(object runtime.Object, annotations map[string]string, eventtype, reason, messageFmt string, args ...interface{})
}
复制代码

主要参数说明:

  • object 对应event资源定义中的 involvedObject
  • eventtype 对应event资源定义中的type,可选Normal,Warning.
  • reason :事件原因
  • message :事件消息

我们来看下当我们调用 Event(object runtime.Object, eventtype, reason, message string) 的整个过程。
发现最终都调用到了 generateEvent 方法:event.go#L316

func (recorder *recorderImpl) generateEvent(object runtime.Object, annotations map[string]string, timestamp metav1.Time, eventtype, reason, message string) {
    .....
    event := recorder.makeEvent(ref, annotations, eventtype, reason, message)
    event.Source = recorder.source
    go func() {
        // NOTE: events should be a non-blocking operation
        defer utilruntime.HandleCrash()
        recorder.Action(watch.Added, event)
    }()
}复制代码

最终事件在一个 goroutine 中通过调用 recorder.Action 进入处理,这里保证了每次调用event方法都是非阻塞的。
其中 makeEvent 的作用主要是构造了一个event对象,事件name根据InvolvedObject中的name加上时间戳生成:

注意看:对于一些非namespace资源产生的event,event的namespace是default

func (recorder *recorderImpl) makeEvent(ref *v1.ObjectReference, annotations map[string]string, eventtype, reason, message string) *v1.Event {
    t := metav1.Time{Time: recorder.clock.Now()}
    namespace := ref.Namespace
    if namespace == "" {
        namespace = metav1.NamespaceDefault
    }
    return &v1.Event{
        ObjectMeta: metav1.ObjectMeta{
            Name:        fmt.Sprintf("%v.%x", ref.Name, t.UnixNano()),
            Namespace:   namespace,
            Annotations: annotations,
        },
        InvolvedObject: *ref,
        Reason:         reason,
        Message:        message,
        FirstTimestamp: t,
        LastTimestamp:  t,
        Count:          1,
        Type:           eventtype,
    }
}复制代码

进一步跟踪Action方法,apimachinery/blob/master/pkg/watch/mux.go#L188:23

// Action distributes the given event among all watchers.
func (m *Broadcaster) Action(action EventType, obj runtime.Object) {
    m.incoming <- Event{action, obj}
}复制代码

将event写入到了一个channel里面。
注意:
这个Action方式是apimachinery包中的方法,因为实现的sturt recorderImpl
*watch.Broadcaster 作为一个匿名struct,并且在 NewRecorder 进行 Broadcaster 赋值,这个Broadcaster其实就是 eventBroadcasterImpl 中的Broadcaster

到此,基本清楚了event最终被写入到了 Broadcaster 中的 incoming channel中,下面看下是怎么进行消费的。

消费事件

makeEventRecorder 调用的 StartLoggingStartRecordingToSink 其实就是完成了对事件的消费。

  • StartLogging直接将event输出到日志
  • StartRecordingToSink将事件写入到apiserver

两个方法内部都调用了 StartEventWatcher 方法,并且传入一个 eventHandler 方法对event进行处理

func (e *eventBroadcasterImpl) StartEventWatcher(eventHandler func(*v1.Event)) watch.Interface {
    watcher := e.Watch()
    go func() {
        defer utilruntime.HandleCrash()
        for watchEvent := range watcher.ResultChan() {
            event, ok := watchEvent.Object.(*v1.Event)
            if !ok {
                // This is all local, so there's no reason this should
                // ever happen.
                continue
            }
            eventHandler(event)
        }
    }()
    return watcher
}复制代码

其中 watcher.ResultChan 方法就拿到了事件,这里是在一个goroutine中通过func (m *Broadcaster) loop() ==>func (m *Broadcaster) distribute(event Event) 方法调用将event又写入了broadcasterWatcher.result

主要看下 StartRecordingToSink 提供的的eventHandlerrecordToSink 方法:

func recordToSink(sink EventSink, event *v1.Event, eventCorrelator *EventCorrelator, sleepDuration time.Duration) {
    // Make a copy before modification, because there could be multiple listeners.
    // Events are safe to copy like this.
    eventCopy := *event
    event = &eventCopy
    result, err := eventCorrelator.EventCorrelate(event)
    if err != nil {
        utilruntime.HandleError(err)
    }
    if result.Skip {
        return
    }
    tries := 0
    for {
        if recordEvent(sink, result.Event, result.Patch, result.Event.Count > 1, eventCorrelator) {
            break
        }
        tries++
        if tries >= maxTriesPerEvent {
            klog.Errorf("Unable to write event '%#v' (retry limit exceeded!)", event)
            break
        }
        // Randomize the first sleep so that various clients won't all be
        // synced up if the master goes down.
        // 第一次重试增加随机性,防止 apiserver 重启的时候所有的事件都在同一时间发送事件
        if tries == 1 {
            time.Sleep(time.Duration(float64(sleepDuration) * rand.Float64()))
        } else {
            time.Sleep(sleepDuration)
        }
    }
}复制代码

其中event被经过了一个 eventCorrelator.EventCorrelate(event) 方法做预处理,主要是聚合相同的事件(避免产生的事件过多,增加 etcd 和 apiserver 的压力,也会导致查看 pod 事件很不清晰)

下面一个for循环就是在进行重试,最大重试次数是12次,调用 recordEvent 方法才真正将event写入到了apiserver。

事件处理

我们来看下EventCorrelate方法:

// EventCorrelate filters, aggregates, counts, and de-duplicates all incoming events
func (c *EventCorrelator) EventCorrelate(newEvent *v1.Event) (*EventCorrelateResult, error) {
    if newEvent == nil {
        return nil, fmt.Errorf("event is nil")
    }
    aggregateEvent, ckey := c.aggregator.EventAggregate(newEvent)
    observedEvent, patch, err := c.logger.eventObserve(aggregateEvent, ckey)
    if c.filterFunc(observedEvent) {
        return &EventCorrelateResult{Skip: true}, nil
    }
    return &EventCorrelateResult{Event: observedEvent, Patch: patch}, err
}
复制代码

分别调用了 aggregator.EventAggregate logger.eventObservefilterFunc 三个方法,分别作用是:

  1. aggregator.EventAggregate:聚合event,如果在最近 10 分钟出现过 10 个相似的事件(除了 message 和时间戳之外其他关键字段都相同的事件),aggregator 会把它们的 message 设置为 (combined from similar events)+event.Message
  2. logger.eventObserve:它会把相同的事件以及包含 aggregator 被聚合了的相似的事件,通过增加 Count 字段来记录事件发生了多少次。
  3. filterFunc: 这里实现了一个基于令牌桶的限流算法,如果超过设定的速率则丢弃,保证了apiserver的安全。

我们主要来看下aggregator.EventAggregate方法:

func (e *EventAggregator) EventAggregate(newEvent *v1.Event) (*v1.Event, string) {
    now := metav1.NewTime(e.clock.Now())
    var record aggregateRecord
    // eventKey is the full cache key for this event
    //eventKey 是将除了时间戳外所有字段结合在一起
    eventKey := getEventKey(newEvent)
    // aggregateKey is for the aggregate event, if one is needed.
    //aggregateKey 是除了message和时间戳外的字段结合在一起,localKey 是message
    aggregateKey, localKey := e.keyFunc(newEvent)

    // Do we have a record of similar events in our cache?
    e.Lock()
    defer e.Unlock()
    //从cache中根据aggregateKey查询是否存在,如果是相同或者相类似的事件会被放入cache中
    value, found := e.cache.Get(aggregateKey)
    if found {
        record = value.(aggregateRecord)
    }

    //判断上次事件产生的时间是否超过10分钟,如何操作则重新生成一个localKeys集合(集合中存放message)
    maxInterval := time.Duration(e.maxIntervalInSeconds) * time.Second
    interval := now.Time.Sub(record.lastTimestamp.Time)
    if interval > maxInterval {
        record = aggregateRecord{localKeys: sets.NewString()}
    }

    // Write the new event into the aggregation record and put it on the cache
    //将locakKey也就是message放入集合中,如果message相同就是覆盖了
    record.localKeys.Insert(localKey)
    record.lastTimestamp = now
    e.cache.Add(aggregateKey, record)

    // If we are not yet over the threshold for unique events, don't correlate them
    //判断localKeys集合中存放的类似事件是否超过10个,
    if uint(record.localKeys.Len()) < e.maxEvents {
        return newEvent, eventKey
    }

    // do not grow our local key set any larger than max
    record.localKeys.PopAny()

    // create a new aggregate event, and return the aggregateKey as the cache key
    // (so that it can be overwritten.)
    eventCopy := &v1.Event{
        ObjectMeta: metav1.ObjectMeta{
            Name:      fmt.Sprintf("%v.%x", newEvent.InvolvedObject.Name, now.UnixNano()),
            Namespace: newEvent.Namespace,
        },
        Count:          1,
        FirstTimestamp: now,
        InvolvedObject: newEvent.InvolvedObject,
        LastTimestamp:  now,
        //这里会对message加个前缀:(combined from similar events):
        Message:        e.messageFunc(newEvent),
        Type:           newEvent.Type,
        Reason:         newEvent.Reason,
        Source:         newEvent.Source,
    }
    return eventCopy, aggregateKey
}复制代码

aggregator.EventAggregate方法中其实就是判断了通过cache和localKeys判断事件是否相似,如果最近 10 分钟出现过 10 个相似的事件就合并并加上前缀,后续通过logger.eventObserve方法进行count累加,如果message也相同,肯定就是直接count++。

总结

好了,event处理的整个流程基本就是这样,我们可以概括一下,可以结合文中的图对比一起看下:

  1. 创建 EventRecorder 对象,通过其提供的 Event 等方法,创建好event对象
  2. 将创建出来的对象发送给 EventBroadcaster 中的channel中
  3. EventBroadcaster 通过后台运行的goroutine,从管道中取出事件,并广播给提前注册好的handler处理
  4. 当输出log的handler收到事件就直接打印事件
  5. EventSink handler收到处理事件就通过预处理之后将事件发送给apiserver
  6. 其中预处理包含三个动作,1、限流 2、聚合 3、计数
  7. apiserver收到事件处理之后就存储在etcd中

回顾event的整个流程,可以看到event并不是保证100%事件写入(从预处理的过程来看),这样做是为了后端服务etcd的可用性,因为event事件在整个集群中产生是非常频繁的,尤其在服务不稳定的时候,而相比Deployment,Pod等其他资源,又没那么的重要。所以这里做了个取舍。

参考文档:

原文地址:https://silenceper.com/blog/202003/kubernetes-event/

关注"学点程序"公众号,了解更多干货内容 学点程序


有疑问加站长微信联系(非本文作者)

本文来自:掘金

感谢作者:silenceper

查看原文:分析kubernetes中的事件机制

入群交流(和以上内容无关):加入Go大咖交流群,或添加微信:liuxiaoyan-s 备注:入群;或加QQ群:692541889

533 次点击  
加入收藏 微博
暂无回复
添加一条新回复 (您需要 登录 后才能回复 没有账号 ?)
  • 请尽量让自己的回复能够对别人有帮助
  • 支持 Markdown 格式, **粗体**、~~删除线~~、`单行代码`
  • 支持 @ 本站用户;支持表情(输入 : 提示),见 Emoji cheat sheet
  • 图片支持拖拽、截图粘贴等方式上传