一次性任务

apiVersion: batch/v1

kind: Job  //资源类型为Job

metadata:

name: myjob

spec:

parallelism: 2  //设置并行数量

completions: 6  //设置完成Pod总数

template:

metadata:

name: myjob

spec:

containers:

- name: hello

image: busybox

command: ["echo", "hello k8s job!"]

restartPolicy: Never  //对于Job只能设置为Never或OnFailure

启动Job：kubectl apply -f myjob.yml

查看Job状态：

kubectl get job

kubectl get pod --show-all

查看Pod输出：kubectl logs myjob-nfkxk

 

删除Job：kubectl delete -f myjob.yml

Job运行失败时查看kubectl get pod --show-all，restartPolicy: Never会创建多个Pod启动，restartPolicy: OnFailure会自动重启该Pod。

 

定时执行

apiVersion: batch/v2alpha1  //CronJob的apiVersion

kind: CronJob  //资源类型为CronJob

metadata:

name: hello

spec:

schedule: "*/1 * * * *"  //每一分钟启动一次

jobTemplate:

spec:

template:

spec:

containers:

- name: hello

image: busybox

command: ["echo", "hello k8s job!"]

restartPolicy: OnFailure

修改API Server：/etc/kubernetes/manifests/kube-apiserver.yaml

--runtime-config=batch/v2alpha1=true

重启kubelet服务：systemctl restart kubelet.service  //kubelet重启kube-apiserver

查看kube-apiserver版本：kubectl api-versions

创建CronJob：kubectl apply -f cronjob.yml

查看状态：

kubectl get cronjob

kubectl get jobs

kubectl get pod --show-all

kubectl logs hello-1505181720-lf350