6.6 KiB
Kubernetes 101
体验Kubernetes最简单的方法是跑一个nginx容器,然后使用kubectl操作该容器。Kubernetes提供了一个类似于docker run
的命令kubectl run
,可以方便的创建一个容器(实际上创建的是一个由deployment来管理的Pod):
$ kubectl run --image=nginx nginx-app --port=80
deployment "nginx-app" created
$ kubectl get pods
NAME READY STATUS RESTARTS AGE
nginx-app-4028413181-cnt1i 1/1 Running 0 52s
等到容器变成Running后,就可以各种kubectl
命令来操作它了,比如
kubectl get
- 类似于docker ps
,查询资源列表kubectl describe
- 类似于docker inspect
,获取资源的详细信息kubectl logs
- 类似于docker logs
,获取容器的日志kubectl exec
- 类似于docker exec
,在容器内执行一个命令
$ kubectl get pods
NAME READY STATUS RESTARTS AGE
nginx-app-4028413181-cnt1i 1/1 Running 0 6m
$ kubectl exec nginx-app-4028413181-cnt1i ps aux
USER PID %CPU %MEM VSZ RSS TTY STAT START TIME COMMAND
root 1 0.0 0.5 31736 5108 ? Ss 00:19 0:00 nginx: master process nginx -g daemon off;
nginx 5 0.0 0.2 32124 2844 ? S 00:19 0:00 nginx: worker process
root 18 0.0 0.2 17500 2112 ? Rs 00:25 0:00 ps aux
$ kubectl describe pod nginx-app-4028413181-cnt1i
Name: nginx-app-4028413181-cnt1i
Namespace: default
Node: boot2docker/192.168.64.12
Start Time: Tue, 06 Sep 2016 08:18:41 +0800
Labels: pod-template-hash=4028413181
run=nginx-app
Status: Running
IP: 172.17.0.3
Controllers: ReplicaSet/nginx-app-4028413181
Containers:
nginx-app:
Container ID: docker://4ef989b57d0a7638ad9c5bbc22e16d5ea5b459281c77074fc982eba50973107f
Image: nginx
Image ID: docker://sha256:4efb2fcdb1ab05fb03c9435234343c1cc65289eeb016be86193e88d3a5d84f6b
Port: 80/TCP
State: Running
Started: Tue, 06 Sep 2016 08:19:30 +0800
Ready: True
Restart Count: 0
Environment Variables: <none>
Conditions:
Type Status
Initialized True
Ready True
PodScheduled True
Volumes:
default-token-9o8ks:
Type: Secret (a volume populated by a Secret)
SecretName: default-token-9o8ks
QoS Tier: BestEffort
Events:
FirstSeen LastSeen Count From SubobjectPath Type Reason Message
--------- -------- ----- ---- ------------- -------- ------ -------
8m 8m 1 {default-scheduler } Normal Scheduled Successfully assigned nginx-app-4028413181-cnt1i to boot2docker
8m 8m 1 {kubelet boot2docker} spec.containers{nginx-app} Normal Pulling pulling image "nginx"
7m 7m 1 {kubelet boot2docker} spec.containers{nginx-app} Normal Pulled Successfully pulled image "nginx"
7m 7m 1 {kubelet boot2docker} spec.containers{nginx-app} Normal Created Created container with docker id 4ef989b57d0a
7m 7m 1 {kubelet boot2docker} spec.containers{nginx-app} Normal Started Started container with docker id 4ef989b57d0a
$ kubectl logs nginx-app-4028413181-cnt1i
127.0.0.1 - - [06/Sep/2016:00:27:13 +0000] "GET / HTTP/1.0 " 200 612 "-" "-" "-"
127.0.0.1 - - [06/Sep/2016:00:27:15 +0000] "GET / HTTP/1.0 " 200 612 "-" "-" "-"
使用yaml定义Pod
上面是通过kubectl run
来启动了第一个Pod,但是kubectl run
并不能支持所有的功能。在Kubernetes中,更经常使用yaml文件来定义资源,并通过kubectl create -f file.yaml
来创建资源。比如,一个简单的nginx Pod可以定义为:
apiVersion: v1
kind: Pod
metadata:
name: nginx
labels:
app: nginx
spec:
containers:
- name: nginx
image: nginx
ports:
- containerPort: 80
前面提到,kubectl run
并不是直接创建一个Pod,而是先创建一个Deployment资源(replicas=1),再由Deployment来自动创建Pod,这等价于这样一个配置:
apiVersion: extensions/v1beta1
kind: Deployment
metadata:
labels:
run: nginx-app
name: nginx-app
namespace: default
spec:
replicas: 1
selector:
matchLabels:
run: nginx-app
strategy:
rollingUpdate:
maxSurge: 1
maxUnavailable: 1
type: RollingUpdate
template:
metadata:
labels:
run: nginx-app
spec:
containers:
- image: nginx
name: nginx-app
ports:
- containerPort: 80
protocol: TCP
dnsPolicy: ClusterFirst
restartPolicy: Always
使用Volume
Pod的生命周期通常比较短,只要出现了异常,就会创建一个新的Pod来代替它。那容器产生的数据呢?容器内的数据会随着Pod消亡而自动消失。Volume就是为了持久化容器数据而生,比如可以为redis容器指定一个hostPath来存储redis数据:
apiVersion: v1
kind: Pod
metadata:
name: redis
spec:
containers:
- name: redis
image: redis
volumeMounts:
- name: redis-persistent-storage
mountPath: /data/redis
volumes:
- name: redis-persistent-storage
hostPath:
path: /data/
Kubernetes volume支持非常多的插件,可以根据实际需要来选择:
- emptyDir
- hostPath
- gcePersistentDisk
- awsElasticBlockStore
- nfs
- iscsi
- flocker
- glusterfs
- rbd
- cephfs
- gitRepo
- secret
- persistentVolumeClaim
- downwardAPI
- azureFileVolume
- vsphereVolume
使用Service
前面虽然创建了Pod,但是在kubernetes中,Pod的IP地址会随着Pod的重启而变化,并不建议直接拿Pod的IP来交互。那如何来访问这些Pod提供的服务呢?使用Service。Service为一组Pod(通过labels来选择)提供一个统一的入口,并为它们提供负载均衡和自动服务发现。比如,可以为前面的nginx-app
创建一个service:
$ kubectl expose deployment nginx-app --type=NodePort --port=80 --target-port=80
service "nginx-app" exposed
$ kubectl describe service nginx-app
Name: nginx-app
Namespace: default
Labels: run=nginx-app
Selector: run=nginx-app
Type: NodePort
IP: 10.0.0.66
Port: <unset> 80/TCP
NodePort: <unset> 30772/TCP
Endpoints: 172.17.0.3:80
Session Affinity: None
No events.
这样,在cluster内部就可以通过http://10.0.0.66
和http://node-ip:30772
来访问nginx-app。而在cluster外面,只能通过http://node-ip:30772
来访问。