1、Job

1.1 理论基础

Job是马上运行的一次性任务， job会创建一个或者多个 Pods，并确保指定数量的 Pods 成功终止。 随着 Pods 成功结束，Job 跟踪记录成功完成的 Pods 个数。 当数量达到指定的成功个数阈值时，任务（即 Job）结束。 删除 Job 的操作会清除所创建的全部 Pods。所以Job类型Pod的**restartPolicy**通常是**Never**或者**OnFailure**。
一种简单的使用场景下，你会创建一个 Job 对象以便以一种可靠的方式运行某 Pod 直到完成。 当第一个 Pod 失败或者被删除（比如因为节点硬件失效或者重启）时，Job 对象会启动一个新的 Pod。

1.2 命令行创建Job

  1. [root@clientvm ~]# kubectl create job -h
  2. Create a job with the specified name.
  3. Usage:
  4.   kubectl create job NAME --image=image [--from=cronjob/name] -- [COMMAND] [args...] [options]
  6. Examples:
  7.   # Create a job
  8.   kubectl create job my-job --image=busybox
  10.   # Create a job with command
  11.   kubectl create job my-job --image=busybox -- date
  13.   # Create a job from a CronJob named "a-cronjob"
  14.   kubectl create job test-job --from=cronjob/a-cronjob

操作范例;

  1. [root@clientvm ~]# kubectl create job my-job -n mytest --image=busybox -- echo hello
  2. job.batch/my-job created
  4. [root@clientvm ~]# kubectl get pod -n mytest
  5. NAME           READY   STATUS              RESTARTS   AGE
  6. my-job-4twr8   0/1     ContainerCreating   0          3s
  8. [root@clientvm ~]# kubectl get pod -n mytest
  9. NAME           READY   STATUS      RESTARTS   AGE
  10. my-job-4twr8   0/1     Completed   0          5s
  12. [root@clientvm ~]# kubectl logs my-job-4twr8 -n mytest
  13. hello
  15. [root@clientvm ~]# kubectl get jobs.batch -n mytest
  16. NAME     COMPLETIONS   DURATION   AGE
  17. my-job   1/1           4s         70s
  19. ##删除Pod会自动创建
  20. [root@clientvm ~]# kubectl delete pod my-job-jp794 -n mytest
  21. pod "my-job-jp794" deleted
  23. [root@clientvm ~]#
  24. [root@clientvm ~]# kubectl get pod -n mytest
  25. NAME           READY   STATUS    RESTARTS   AGE
  26. my-job-c72bq   1/1     Running   0          46s

1.3 使用yaml创建job

生成yaml文件

  1. [root@clientvm ~]# kubectl create job my-job -n mytest --image=busybox --dry-run=client -o yaml -- sleep 200
  2. apiVersion: batch/v1
  3. kind: Job
  4. metadata:
  5.   creationTimestamp: null
  6.   name: my-job
  7.   namespace: mytest
  8. spec:
  9.   template:
  10.     metadata:
  11.       creationTimestamp: null
  12.     spec:
  13.       containers:
  14.       - command:
  15.         - sleep
  16.         - "200"
  17.         image: busybox
  18.         name: my-job
  19.         resources: {}
  20.       restartPolicy: Never
  21. status: {}

创建Job

  1. [root@clientvm ~]# kubectl apply -f job.yaml
  2. job.batch/my-job created
  4. [root@clientvm ~]# kubectl get jobs.batch -n mytest
  5. NAME     COMPLETIONS   DURATION   AGE
  6. my-job   0/1           38s        38s
  8. [root@clientvm ~]# kubectl get pod -n mytest
  9. NAME           READY   STATUS    RESTARTS   AGE
  10. my-job-n2z45   1/1     Running   0          49s

1.4 设置并行与完成次数

completions：设置完成次数
parallelism： 设置并行

  1. [root@clientvm ~]# kubectl explain job.spec
  2. KIND:     Job
  3. VERSION:  batch/v1
  4. ......
  5.    completions  <integer>
  6.      Specifies the desired number of successfully finished pods the job should
  7.      be run with. 
  8.    parallelism  <integer>
  9.      Specifies the maximum desired number of pods the job should run at any
  10.      given time.

2、CronJob

Cron Job 创建基于时间调度的 Jobs。
一个 CronJob 对象就像 crontab (cron table) 文件中的一行。 它用 Cron 格式进行编写， 并周期性地在给定的调度时间执行 Job。

CronJobs 对于创建周期性的、反复重复的任务很有用，例如执行数据备份或者发送邮件。 CronJobs 也可以用来计划在指定时间来执行的独立任务，例如计划当集群看起来很空闲时 执行某个 Job。

2.1 命令行创建CronJob

  1. [root@clientvm ~]# kubectl create cronjob -h
  2. Create a cronjob with the specified name.
  3. Usage:
  4.   kubectl create cronjob NAME --image=image --schedule='0/5 * * * ?' -- [COMMAND] [args...] [flags] [options]
  6. Aliases:
  7. cronjob, cj
  9. Examples:
  10.   # Create a cronjob
  11.   kubectl create cronjob my-job --image=busybox --schedule="*/1 * * * *"
  13.   # Create a cronjob with command
  14.   kubectl create cronjob my-job --image=busybox --schedule="*/1 * * * *" -- date

范例：

  1. [root@clientvm ~]# kubectl create cronjob my-job -n mytest --image=busybox --schedule="*/1 * * * *" -- date
  2. cronjob.batch/my-job created
  4. [root@clientvm ~]# kubectl get cronjobs.batch -n mytest
  5. NAME     SCHEDULE      SUSPEND   ACTIVE   LAST SCHEDULE   AGE
  6. my-job   */1 * * * *   False     0        <none>          12s
  7. [root@clientvm ~]# kubectl get cronjobs.batch -n mytest
  8. NAME     SCHEDULE      SUSPEND   ACTIVE   LAST SCHEDULE   AGE
  9. my-job   */1 * * * *   False     1        10s             63s
  10. [root@clientvm ~]#
  11. [root@clientvm ~]#
  12. [root@clientvm ~]# kubectl get pod -n mytest
  13. NAME                      READY   STATUS              RESTARTS   AGE
  14. my-job-1607481720-5dvr2   0/1     ContainerCreating   0          17s
  16. [root@clientvm ~]# kubectl get pod -n mytest
  17. NAME                      READY   STATUS      RESTARTS   AGE
  18. my-job-1607481720-5dvr2   0/1     Completed   0          35s
  20. [root@clientvm ~]# kubectl get pod -n mytest
  21. NAME                      READY   STATUS      RESTARTS   AGE
  22. my-job-1607481720-5dvr2   0/1     Completed   0          80s
  23. my-job-1607481780-sszc5   0/1     Completed   0          20s
  25. [root@clientvm ~]# kubectl logs my-job-1607481720-5dvr2 -n mytest
  26. Wed Dec  9 02:42:26 UTC 2020
  28. [root@clientvm ~]# kubectl describe cronjobs.batch -n mytest  my-job
  29. Name:                          my-job
  30. Namespace:                     mytest
  31. Labels:                        <none>
  32. Annotations:                   <none>
  33. Schedule:                      */1 * * * *
  34. ......
  35. Events:
  36.   Type    Reason            Age    From                Message
  37.   ----    ------            ----   ----                -------
  38.   Normal  SuccessfulCreate  2m12s  cronjob-controller  Created job my-job-1607481720
  39.   Normal  SawCompletedJob   111s   cronjob-controller  Saw completed job: my-job-1607481720, status: Complete
  40.   Normal  SuccessfulCreate  71s    cronjob-controller  Created job my-job-1607481780
  41.   Normal  SawCompletedJob   51s    cronjob-controller  Saw completed job: my-job-1607481780, status: Complete
  42.   Normal  SuccessfulCreate  11s    cronjob-controller  Created job my-job-1607481840
  44. ## 删除
  45. [root@clientvm ~]# kubectl delete cronjobs.batch -n mytest  my-job
  46. cronjob.batch "my-job" deleted

2.2 使用yaml创建CronJob

生成配置文件范例

  1. [root@clientvm ~]# kubectl create cronjob my-job -n mytest --image=busybox --dry-run=client -o yaml --schedule="*/1 * * * *"  -- date
  2. apiVersion: batch/v1beta1
  3. kind: CronJob
  4. metadata:
  5.   creationTimestamp: null
  6.   name: my-job
  7.   namespace: mytest
  8. spec:
  9.   jobTemplate:
  10.     metadata:
  11.       creationTimestamp: null
  12.       name: my-job
  13.     spec:
  14.       template:
  15.         metadata:
  16.           creationTimestamp: null
  17.         spec:
  18.           containers:
  19.           - command:
  20.             - date
  21.             image: busybox
  22.             name: my-job
  23.             resources: {}
  24.           restartPolicy: OnFailure
  25.   schedule: '*/1 * * * *'
  26. status: {}

创建CronJob

  1. [root@clientvm ~]# kubectl apply -f cronjob.yaml
  2. cronjob.batch/my-job created
  4. [root@clientvm ~]# kubectl get cronjobs.batch -n mytest
  5. NAME     SCHEDULE      SUSPEND   ACTIVE   LAST SCHEDULE   AGE
  6. my-job   */1 * * * *   False     0        <none>          7s