Tekton初体验

发表于 | 分类于
字数统计: 1.2k 字 | 阅读时长 ≈ 5 分钟

前言

  • 简介

    1
    2
    Tekton是一个强大、灵活的构建CI/CD流水线系统的开源框架，允许开发者构建、测试和发布应用。
    Tekton是云原生的，通过CRD让用户快速灵活定义流水线。

  • 使用tekton的好处

    1
    2
    3
    4
    5
    可定制：tekton实体是完全可定制的，从而具有高度的灵活性。
    可重复使用：tekton实体完全可移植，因此一旦定义组织内任何人都可以使用给定的管道并重用其构造块。快速构建复杂的管道而无需重新造轮子。
    可扩展：可使用tekton目录中的预制组件快色创建新的并展开现有管道。
    标准化：tekton在k8s集群中作为扩展安装并运行。
    缩放性：与集群进行缩放，无需重新定义资源分配或对管道的任何其他修改。

  • 组件

    1
    2
    3
    4
    5
    6
    7
    tekton pipelines: tekton的基础，定义了一组CRD用于定义pipeline
    tekton triggers: 允许基于event实例化pipeline，如git的PR请求
    tekton cli: 提供命令行工具和tekton交互
    tekton dashboard: 图形化界面展示pipeline信息
    tekton catalog: 高质量，社区贡献的pipeline仓库
    tekton hub: 图形化界面访问tekton catalog
    tekton operator: 在k8s上安装、移除、更新tekton组件

  • 基本概念

1
2
3
4
5
6
7
Tekton最重要的五个概念：Task、TaskRun、Pipeline、PipelineRun、PipelineResource

Task: tekton中的最小单元，代表一个任务模版，包含多个步骤。每个操作定义为task中的一个step
Pipeline: 多个task组成的有向无环图，定义了流水线的模版
PipelineRun: pipeline真正执行时需要定义一个pipelineRun作为流水线的实例，生成一条流水线记录
TaskRun: task真正执行的实例，记录任务状态
PipelineResource: 流水线执行过程中需要的资源信息。

安装

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
# 官方文档
https://tekton.dev/docs/installation/pipelines/

# 能访问国外网，如gcr.io
kubectl apply --filename https://storage.googleapis.com/tekton-releases/pipeline/latest/release.yaml


# 查看安装的资源
[root@master ~]# kubectl get pod -n tekton-pipelines-resolvers
NAME                                                 READY   STATUS    RESTARTS   AGE
tekton-pipelines-remote-resolvers-5b6d9f9d55-gqst5   1/1     Running   0          11m
[root@master ~]# kubectl get pod -n tekton-pipelines
NAME                                           READY   STATUS    RESTARTS   AGE
tekton-pipelines-controller-6754f4c756-8t5gp   1/1     Running   0          11m
tekton-pipelines-webhook-649f49588-tztsk       1/1     Running   0          11m
[root@master ~]# 
[root@master ~]# kubectl get crd | grep tekton
clustertasks.tekton.dev                               2023-01-24T09:04:04Z
customruns.tekton.dev                                 2023-01-24T09:04:04Z
pipelineresources.tekton.dev                          2023-01-24T09:04:04Z
pipelineruns.tekton.dev                               2023-01-24T09:04:04Z
pipelines.tekton.dev                                  2023-01-24T09:04:04Z
resolutionrequests.resolution.tekton.dev              2023-01-24T09:04:04Z
runs.tekton.dev                                       2023-01-24T09:04:04Z
taskruns.tekton.dev                                   2023-01-24T09:04:04Z
tasks.tekton.dev                                      2023-01-24T09:04:04Z
verificationpolicies.tekton.dev                       2023-01-24T09:04:04Z

安装Dashboard

  • yaml安装
1
2
3
4
5
6
7
8
9
kubectl apply --filename https://storage.googleapis.com/tekton-releases/dashboard/latest/release.yaml
# 这是默认安装只读权限的dashboard，安装读写权限使用 release-full.yaml

# 查看资源
[root@master ~]# kubectl get pod -n tekton-pipelines
NAME                                           READY   STATUS    RESTARTS   AGE
tekton-dashboard-664cb4fd64-54ww8              1/1     Running   0          110s
tekton-pipelines-controller-6754f4c756-8t5gp   1/1     Running   0          17m
tekton-pipelines-webhook-649f49588-tztsk       1/1     Running   0          17m
  • ingress暴露服务
    1
    2
    3
    4
    5
    6
    7
    8
    9
    10
    11
    12
    13
    14
    15
    16
    17
    18
    apiVersion: networking.k8s.io/v1
    kind: Ingress
    metadata:
      name: tekon-dashboard
      namespace: tekton-pipelines
    spec:
      ingressClassName: liykingress
      rules:
      - host: tekton.key1024.cn
        http:
          paths:
          - backend:
              service:
                name: tekton-dashboard
                port:
                  number: 9097
            path: /
            pathType: Prefix

创建任务

  • 创建简单task 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
# 创建个最简单task，使用alpine的镜像启动一个容器执行指定脚步
# hello-world.yaml 文件内容, 定义名称为 hello 的 task
apiVersion: tekton.dev/v1beta1
kind: Task
metadata:
  name: hello
spec:
  steps:
    - name: echo
      image: alpine
      script: |
        #!/bin/sh
        echo "Hello World"
        
# 部署：kubectl apply -f hello-world.yaml
1
2
3
4
5
6
7
8
9
10
11
# 创建taskrun来执行hello这个task
#hello-world-run.yaml
apiVersion: tekton.dev/v1beta1
kind: TaskRun
metadata:
  name: hello-task-run
spec:
  taskRef:
    name: hello
    
# 部署：kubectl apply -f hello-world-run.yaml
1
2
3
4
5
6
7
# 查看任务
[root@master task]# kubectl get task
NAME    AGE
hello   3m52s
[root@master task]# kubectl get taskrun
NAME             SUCCEEDED   REASON      STARTTIME   COMPLETIONTIME
hello-task-run   True        Succeeded   79s         26s

  • 创建pipeline调用task
    1
    2
    3
    4
    5
    6
    7
    8
    9
    10
    11
    12
    13
    14
    # 创建名称goodbye的task
    # goodbye-world.yaml
    apiVersion: tekton.dev/v1beta1
    kind: Task
    metadata:
      name: goodbye
    spec:
      steps:
        - name: goodbye
          image: ubuntu
          script: |
            #!/bin/sh
            echo "Goodbye World!"
    # 部署：kubectl apply -f goodbye-world.yaml
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
# 创建并运行1个包含上述2个task的pipeline
#hello-goodbye-pipeline.yaml
apiVersion: tekton.dev/v1beta1
kind: Pipeline
metadata:
  name: hello-goodbye
spec:
  tasks:
    - name: hello
      taskRef:
        name: hello
    - name: goodbye
      runAfter:
        - hello
      taskRef:
        name: goodbye
# 部署：kubectl apply -f hello-goodbye-pipeline.yaml
1
2
3
4
5
6
7
8
9
10
# 创建pipelineRun对象来实例化一个pipeline，即运行这个pipeline
#hello-goodbye-pipeline-run.yaml
apiVersion: tekton.dev/v1beta1
kind: PipelineRun
metadata:
  name: hello-goodbye-run
spec:
  pipelineRef:
    name: hello-goodbye
# 部署：kubectl apply -f hello-goodbye-pipeline-run.yaml

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
# 上述文件合并一起
# multi-task-pipeline.yaml
apiVersion: tekton.dev/v1beta1
kind: Task
metadata:
  name: hello
spec:
  steps:
    - name: echo
      image: alpine
      script: |
        #!/usr/bin/env sh
        echo "Hello World"
---
apiVersion: tekton.dev/v1beta1
kind: Task
metadata:
  name: goodbye
spec:
  steps:
    - name: goodbye
      image: ubuntu
      script: |
        #!/usr/bin/env sh
        echo "Goodbye World!"
---
apiVersion: tekton.dev/v1beta1
kind: Pipeline
metadata:
  name: multiple-task-pipeline
spec:
  tasks:
    - name: hello
      taskRef:
        name: hello
    - name: goodbye
      runAfter:
        - hello
      taskRef:
        name: goodbye
  • 简化task任务的pipeline
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
# 上面定义的task比较累赘，可以在pipeline中内嵌定义task
# simple-multi-task-pipeline.yaml
apiVersion: tekton.dev/v1beta1
kind: Pipeline
metadata:
  name: simple-multi-task-pipeline
spec:
  tasks:
    - name: hello
      taskSpec:
        steps:
        - image: alpine
          script: |
            #!/bin/sh
            echo "Hello world!"
 
    - name: goodbye
      runAfter:
        - hello
      taskSpec:
        steps:
        - image: alpine
          script: |
            #!/bin/sh
            echo "Goodbye world!"

# 执行可定义pipelineRun通过kubectl来执行；也可以在tekton dashboard上触发执行（需有读写权限的dashboard）
1
2
3
4
5
6
7
8
#simple-multi-task-pipeline-run.yaml
apiVersion: tekton.dev/v1beta1
kind: PipelineRun
metadata:
  name: simple-multi-task-pipeline-run
spec:
  pipelineRef:
    name: simple-multi-task-pipeline
-------------本文结束感谢您的阅读-------------
原创技术分享，感谢您的支持。