温馨提示×

温馨提示×

您好,登录后才能下订单哦!

密码登录×
登录注册×
其他方式登录
点击 登录注册 即表示同意《亿速云用户服务条款》

如何分析Knative Eventing中的Sequence及其4 种使用场景

发布时间:2021-12-28 15:18:18 来源:亿速云 阅读:155 作者:柒染 栏目:云计算

如何分析Knative Eventing中的Sequence及其4 种使用场景,很多新手对此不是很清楚,为了帮助大家解决这个难题,下面小编将为大家详细讲解,有这方面需求的人可以来学习下,希望你能有所收获。

导读:在实际的开发中我们经常会遇到将一条数据需要经过多次处理的场景,称为 Pipeline。那么在 Knative 中是否也提供这样的能力呢?其实从 Knative Eventing 0.7 版本开始,就提供了 Sequence CRD 资源,用于事件处理 Pipeline。下面将为大家详细介绍 Sequence 的定义及在 Knative Eventing 中提供的 4 种使用场景。

Sequence 定义

首先来看一下 Sequence Spec 定义:

apiVersion: messaging.knative.dev/v1alpha1
kind: Sequence
metadata:
  name: test
spec:
  channelTemplate:
    apiVersion: messaging.knative.dev/v1alpha1
    kind: InMemoryChannel
  steps:
    - ref:
        apiVersion: serving.knative.dev/v1alpha1
        kind: Service
        name: test
  reply:
    kind: Broker
    apiVersion: eventing.knative.dev/v1alpha1
    name: test

Sequence Spec 包括 3 个部分:

  1. steps: 在 step 中定义了按照顺序执行的服务,每个服务会对应创建 Subscription;

  2. channelTemplate:指定了使用具体的那个 Channel;

  3. reply:(可选)定义了将最后一个 step 服务结果转发到的目标服务。

    Sequence 都是适合哪些具体应用场景呢?我们上面也提到了事件处理的 Pipeline。那么在实际场景应用中究竟以什么样的形式体现呢? 现在我们揭晓一下 Sequence 在 Knative Eventing 中提供的如下 4 种使用场景:

  • 直接访问 Service;

  • 面向事件处理;

  • 级联 Sequence;

  • 面向 Broker/Trigger。

直接访问 Service 场景

事件源产生的事件直接发送给 Sequence 服务, Sequence 接收到事件之后顺序调用 Service 服务对事件进行处理:

如何分析Knative Eventing中的Sequence及其4 种使用场景

创建 Knative Service

这里我们创建 3 个 Knative Service 用于事件处理。每个 Service 接收到事件之后会打印当前的事件处理信息。

apiVersion: serving.knative.dev/v1alpha1
kind: Service
metadata:
  name: first
spec:
  template:
    spec:
      containers:
        - image: registry.cn-hangzhou.aliyuncs.com/knative-sample/probable-summer:2656f39a7fcb6afd9fc79e7a4e215d14d651dc674f38020d1d18c6f04b220700
          env:
            - name: STEP
              value: "0"

---
apiVersion: serving.knative.dev/v1alpha1
kind: Service
metadata:
  name: second
spec:
  template:
    spec:
      containers:
        - image: registry.cn-hangzhou.aliyuncs.com/knative-sample/probable-summer:2656f39a7fcb6afd9fc79e7a4e215d14d651dc674f38020d1d18c6f04b220700
          env:
            - name: STEP
              value: "1"
---
apiVersion: serving.knative.dev/v1alpha1
kind: Service
metadata:
  name: third
spec:
  template:
    spec:
      containers:
        - image: registry.cn-hangzhou.aliyuncs.com/knative-sample/probable-summer:2656f39a7fcb6afd9fc79e7a4e215d14d651dc674f38020d1d18c6f04b220700
          env:
            - name: STEP
              value: "2"
---

创建 Sequence

创建顺序调用 first->second->third Service 的 Sequence。

apiVersion: messaging.knative.dev/v1alpha1
kind: Sequence
metadata:
  name: sequence
spec:
  channelTemplate:
    apiVersion: messaging.knative.dev/v1alpha1
    kind: InMemoryChannel
  steps:
    - ref:
        apiVersion: serving.knative.dev/v1alpha1
        kind: Service
        name: first
    - ref:
        apiVersion: serving.knative.dev/v1alpha1
        kind: Service
        name: second
    - ref:
        apiVersion: serving.knative.dev/v1alpha1
        kind: Service
        name: third

创建数据源

创建 CronJobSource 数据源,每隔 1 分钟发送一条事件消息 {"message": "Hello world!"} 到 Sequence 服务。

apiVersion: sources.eventing.knative.dev/v1alpha1
kind: CronJobSource
metadata:
  name: cronjob-source
spec:
  schedule: "*/1 * * * *"
  data: '{"message": "Hello world!"}'
  sink:
    apiVersion: messaging.knative.dev/v1alpha1
    kind: Sequence
    name: sequence

示例结果

如何分析Knative Eventing中的Sequence及其4 种使用场景 如何分析Knative Eventing中的Sequence及其4 种使用场景 如何分析Knative Eventing中的Sequence及其4 种使用场景 如何分析Knative Eventing中的Sequence及其4 种使用场景

面向事件处理场景

事件源产生的事件直接发送给 Sequence 服务, Sequence 接收到事件之后顺序调用 Service 服务对事件进行处理,处理之后的最终结果会调用 event-display Service 显示: 如何分析Knative Eventing中的Sequence及其4 种使用场景

创建 Knative Service

同上创建 3 个 Knative Service 用于事件处理:

apiVersion: serving.knative.dev/v1alpha1
kind: Service
metadata:
  name: first
spec:
  template:
    spec:
      containers:
        - image: registry.cn-hangzhou.aliyuncs.com/knative-sample/probable-summer:2656f39a7fcb6afd9fc79e7a4e215d14d651dc674f38020d1d18c6f04b220700
          env:
            - name: STEP
              value: "0"

---
apiVersion: serving.knative.dev/v1alpha1
kind: Service
metadata:
  name: second
spec:
  template:
    spec:
      containers:
        - image: registry.cn-hangzhou.aliyuncs.com/knative-sample/probable-summer:2656f39a7fcb6afd9fc79e7a4e215d14d651dc674f38020d1d18c6f04b220700
          env:
            - name: STEP
              value: "1"
---
apiVersion: serving.knative.dev/v1alpha1
kind: Service
metadata:
  name: third
spec:
  template:
    spec:
      containers:
        - image: registry.cn-hangzhou.aliyuncs.com/knative-sample/probable-summer:2656f39a7fcb6afd9fc79e7a4e215d14d651dc674f38020d1d18c6f04b220700
          env:
            - name: STEP
              value: "2"
---

创建 Sequence

创建顺序调用 first->second->third Service 的 Sequence,将处理结果通过 reply 发送给 event-display

apiVersion: messaging.knative.dev/v1alpha1
kind: Sequence
metadata:
  name: sequence
spec:
  channelTemplate:
    apiVersion: messaging.knative.dev/v1alpha1
    kind: InMemoryChannel
  steps:
    - ref:
        apiVersion: serving.knative.dev/v1alpha1
        kind: Service
        name: first
    - ref:
        apiVersion: serving.knative.dev/v1alpha1
        kind: Service
        name: second
    - ref:
        apiVersion: serving.knative.dev/v1alpha1
        kind: Service
        name: third
  reply:
    kind: Service
    apiVersion: serving.knative.dev/v1alpha1
    name: event-display

创建结果显示 Service

创建 event-display Service, 用于接收最终的结果信息。

apiVersion: serving.knative.dev/v1alpha1
kind: Service
metadata:
  name: event-display
spec:
  template:
    spec:
      containers:
        - image: registry.cn-hangzhou.aliyuncs.com/knative-release/event_display:bf45b3eb1e7fc4cb63d6a5a6416cf696295484a7662e0cf9ccdf5c080542c21d

创建数据源

创建 CronJobSource 数据源,每隔 1 分钟发送一条事件消息 {"message": "Hello world!"} 到 Sequence 服务。

apiVersion: sources.eventing.knative.dev/v1alpha1
kind: CronJobSource
metadata:
  name: cronjob-source
spec:
  schedule: "*/1 * * * *"
  data: '{"message": "Hello world!"}'
  sink:
    apiVersion: messaging.knative.dev/v1alpha1
    kind: Sequence
    name: sequence

示例结果

如何分析Knative Eventing中的Sequence及其4 种使用场景 如何分析Knative Eventing中的Sequence及其4 种使用场景

级联 Sequence 场景

Sequence 更高级的地方还在于支持级联处理: Sequence By Sequence,这样可以进行多次 Sequence 处理,满足复杂事件处理场景需求。 如何分析Knative Eventing中的Sequence及其4 种使用场景

创建 Knative Service

创建 6 个 Knative Service 用于事件处理, 前 3 个用于第 1 个 Sequence,后 3 个用于第 2 个 Sequence。

apiVersion: serving.knative.dev/v1alpha1
kind: Service
metadata:
  name: first
spec:
  template:
    spec:
      containers:
        - image: registry.cn-hangzhou.aliyuncs.com/knative-sample/probable-summer:2656f39a7fcb6afd9fc79e7a4e215d14d651dc674f38020d1d18c6f04b220700
          env:
            - name: STEP
              value: "0"

---
apiVersion: serving.knative.dev/v1alpha1
kind: Service
metadata:
  name: second
spec:
  template:
    spec:
      containers:
        - image: registry.cn-hangzhou.aliyuncs.com/knative-sample/probable-summer:2656f39a7fcb6afd9fc79e7a4e215d14d651dc674f38020d1d18c6f04b220700
          env:
            - name: STEP
              value: "1"
---
apiVersion: serving.knative.dev/v1alpha1
kind: Service
metadata:
  name: third
spec:
  template:
    spec:
      containers:
        - image: registry.cn-hangzhou.aliyuncs.com/knative-sample/probable-summer:2656f39a7fcb6afd9fc79e7a4e215d14d651dc674f38020d1d18c6f04b220700
          env:
            - name: STEP
              value: "2"
---
apiVersion: serving.knative.dev/v1alpha1
kind: Service
metadata:
  name: fourth
spec:
  template:
    spec:
      containers:
        - image: registry.cn-hangzhou.aliyuncs.com/knative-sample/probable-summer:2656f39a7fcb6afd9fc79e7a4e215d14d651dc674f38020d1d18c6f04b220700
          env:
            - name: STEP
              value: "3"

---
apiVersion: serving.knative.dev/v1alpha1
kind: Service
metadata:
  name: fifth
spec:
  template:
    spec:
      containers:
        - image: registry.cn-hangzhou.aliyuncs.com/knative-sample/probable-summer:2656f39a7fcb6afd9fc79e7a4e215d14d651dc674f38020d1d18c6f04b220700
          env:
            - name: STEP
              value: "4"
---
apiVersion: serving.knative.dev/v1alpha1
kind: Service
metadata:
  name: sixth
spec:
  template:
    spec:
      containers:
        - image: registry.cn-hangzhou.aliyuncs.com/knative-sample/probable-summer:2656f39a7fcb6afd9fc79e7a4e215d14d651dc674f38020d1d18c6f04b220700
          env:
            - name: STEP
              value: "5"
---

创建第 1 个 Sequence

使用 first->second->third Service 用于第 1 个 Sequence 调用处理,将执行结果发送给第 2 个 Sequence。

apiVersion: messaging.knative.dev/v1alpha1
kind: Sequence
metadata:
  name: first-sequence
spec:
  channelTemplate:
    apiVersion: messaging.knative.dev/v1alpha1
    kind: InMemoryChannel
  steps:
    - ref:
        apiVersion: serving.knative.dev/v1alpha1
        kind: Service
        name: first
    - ref:
        apiVersion: serving.knative.dev/v1alpha1
        kind: Service
        name: second
    - ref:
        apiVersion: serving.knative.dev/v1alpha1
        kind: Service
        name: third
  reply:
    kind: Sequence
    apiVersion: messaging.knative.dev/v1alpha1
    name: second-sequence

创建第 2 个 Sequence

使用 fourth->fifth->sixth Service 用于第 2 个 Sequence 调用处理,将执行结果发送给 event-display

apiVersion: messaging.knative.dev/v1alpha1
kind: Sequence
metadata:
  name: second-sequence
spec:
  channelTemplate:
    apiVersion: messaging.knative.dev/v1alpha1
    kind: InMemoryChannel
  steps:
    - ref:
        apiVersion: serving.knative.dev/v1alpha1
        kind: Service
        name: fourth
    - ref:
        apiVersion: serving.knative.dev/v1alpha1
        kind: Service
        name: fifth
    - ref:
        apiVersion: serving.knative.dev/v1alpha1
        kind: Service
        name: sixth
  reply:
    kind: Service
    apiVersion: serving.knative.dev/v1alpha1
    name: event-display

创建结果显示 Service

apiVersion: serving.knative.dev/v1alpha1
kind: Service
metadata:
  name: event-display
spec:
  template:
    spec:
      containerers:
        - image: registry.cn-hangzhou.aliyuncs.com/knative-release/event_display:bf45b3eb1e7fc4cb63d6a5a6416cf696295484a7662e0cf9ccdf5c080542c21d

创建数据源指向第 1 个 Sequence

apiVersion: sources.eventing.knative.dev/v1alpha1
kind: CronJobSource
metadata:
  name: cronjob-source
spec:
  schedule: "*/1 * * * *"
  data: '{"message": "Hello world!"}'
  sink:
    apiVersion: messaging.knative.dev/v1alpha1
    kind: Sequence
    name: first-sequence

示例结果

如何分析Knative Eventing中的Sequence及其4 种使用场景 如何分析Knative Eventing中的Sequence及其4 种使用场景

Broker/Trigger 场景

事件源 cronjobsource 向 Broker 发送事件,通过 Trigger 将这些事件发送到由 3 个 Service 调用的 Sequence 中。Sequence 处理完之后将结果事件发送给 Broker,并最终由另一个 Trigger 发送给 event-display Service 显示事件结果:

如何分析Knative Eventing中的Sequence及其4 种使用场景

创建 Knative Service

同上创建 3 个 Knative Service,用于 Sequence 中服务处理。

apiVersion: serving.knative.dev/v1alpha1
kind: Service
metadata:
  name: first
spec:
  template:
    spec:
      containers:
      - image: registry.cn-hangzhou.aliyuncs.com/knative-sample/probable-summer:2656f39a7fcb6afd9fc79e7a4e215d14d651dc674f38020d1d18c6f04b220700
        env:
        - name: STEP
          value: "0"

---
apiVersion: serving.knative.dev/v1alpha1
kind: Service
metadata:
  name: second
spec:
  template:
    spec:
      containers:
      - image: registry.cn-hangzhou.aliyuncs.com/knative-sample/probable-summer:2656f39a7fcb6afd9fc79e7a4e215d14d651dc674f38020d1d18c6f04b220700
        env:
        - name: STEP
          value: "1"
---
apiVersion: serving.knative.dev/v1alpha1
kind: Service
metadata:
  name: third
spec:
  template:
    spec:
      containers:
      - image: registry.cn-hangzhou.aliyuncs.com/knative-sample/probable-summer:2656f39a7fcb6afd9fc79e7a4e215d14d651dc674f38020d1d18c6f04b220700
        env:
        - name: STEP
          value: "2"

---

创建 Sequence

创建 Sequence,这里依次顺序执行 first->second->third 这 3 个服务。将最终处理的结果发送到 broker-test 中。

apiVersion: messaging.knative.dev/v1alpha1
kind: Sequence
metadata:
  name: sequence
spec:
  channelTemplate:
    apiVersion: messaging.knative.dev/v1alpha1
    kind: InMemoryChannel
  steps:
    - ref:
        apiVersion: serving.knative.dev/v1alpha1
        kind: Service
        name: first
    - ref:
        apiVersion: serving.knative.dev/v1alpha1
        kind: Service
        name: second
    - ref:
        apiVersion: serving.knative.dev/v1alpha1
        kind: Service
        name: third
  reply:
    kind: Broker
    apiVersion: eventing.knative.dev/v1alpha1
    name: default

创建事件源指向 Broker

创建 CronjobSource,它将每隔 1 分钟发送一条 {"message": "Hello world!"} 消息到 broker-test 中。

apiVersion: sources.eventing.knative.dev/v1alpha1
kind: CronJobSource
metadata:
  name: cronjob-source
spec:
  schedule: "*/1 * * * *"
  data: '{"message": "Hello world!"}'
  sink:
    apiVersion: eventing.knative.dev/v1alpha1
    kind: Broker
    name: default

创建 Broker

创建默认 Broker:

kubectl label namespace default knative-eventing-injection=enabled

创建 Trigger  指向 Sequence

创建订阅事件类型为 dev.knative.cronjob.event 的 Trigger, 用于 Sequence 进行消费处理。

apiVersion: eventing.knative.dev/v1alpha1
kind: Trigger
metadata:
  name: sequence-trigger
spec:
  filter:
    sourceAndType:
      type: dev.knative.cronjob.event
  subscriber:
    ref:
      apiVersion: messaging.knative.dev/v1alpha1
      kind: Sequence
      name: sequence

创建结果订阅 Trigger

创建订阅 samples.http.mod3 的事件类型 Trigger,将 Sequence 执行的结果发送给 event-display Service 进行显示。

apiVersion: eventing.knative.dev/v1alpha1
kind: Trigger
metadata:
  name: display-trigger
spec:
  filter:
    sourceAndType:
      type: samples.http.mod3
  subscriber:
    ref:
      apiVersion: serving.knative.dev/v1alpha1
      kind: Service
      name: event-display
---
apiVersion: serving.knative.dev/v1alpha1
kind: Service
metadata:
  name: event-display
spec:
  template:
    spec:
      containers:
        - image: registry.cn-hangzhou.aliyuncs.com/knative-release/event_display:bf45b3eb1e7fc4cb63d6a5a6416cf696295484a7662e0cf9ccdf5c080542c21d
---

<a name="4ed65929-3"></a>

示例结果

如何分析Knative Eventing中的Sequence及其4 种使用场景 如何分析Knative Eventing中的Sequence及其4 种使用场景

看完上述内容是否对您有帮助呢?如果还想对相关知识有进一步的了解或阅读更多相关文章,请关注亿速云行业资讯频道,感谢您对亿速云的支持。

向AI问一下细节

免责声明:本站发布的内容(图片、视频和文字)以原创、转载和分享为主,文章观点不代表本网站立场,如果涉及侵权请联系站长邮箱:is@yisu.com进行举报,并提供相关证据,一经查实,将立刻删除涉嫌侵权内容。

AI