Deployment可以管理在线业务，能够让应用永不宕机，是Kubernetes最实用最常用的一个对象。它实现了Kubernetes一个相当重要的功能：水平扩展/收缩。这个功能是编排系统的必备能力。
如果只用Pod，一些熟悉比如restartPolicy也可以管理异常重启，但是太过简单。而且它只能保证容器正常工作，如果从外部删除了这个容器，那么对容器的控制也无法操作。
除此之外，还有高可用、多实例等操作，所以还需要一个对象来管理在线业务的Pod。
Replication Controller（复制控制器，RC）和ReplicaSet（复制集，RS）是两种部署Pod的方式，但是在实际生产中往往用的都是更高级的Deployment，所以这里只做简要介绍。
RC主要用于规定Pod副本数的期望值，Pod出现失败、删除、终止时，RC会自动替换掉，即使只有一个Pod也应该声明RC。
一个Replication Controller的示例：
ReplicaSet在RC的基础上增加了标签选择器，是一个更高级的概念。
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
|
apiVersion: apps/v1
kind: ReplicaSet
metadata:
name: nginx-set
labels:
app: nginx
spec:
replicas: 3
selector:
matchLabels:
app: nginx
template:
metadata:
labels:
app: nginx
spec:
containers:
- name: nginx
image: nginx:1.7.9
|
相比于直接创建一个Pod，更好的方式就是用一个replicas=1的Deployment。这样即使节点故障了，Kubernetes还可以从健康的节点自动创建一个Pod。
实际上，Deployment是ReplicaSet的母集，Deployment控制的就是ReplicaSet对象，而不是Pod。Pod的个数通过ReplicaSet管理。
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
|
apiVersion: apps/v1
kind: Deployment
metadata:
labels:
app: ngx-dep
name: ngx-dep
spec:
replicas: 3
selector:
matchLabels:
app: ngx-dep
template:
metadata:
labels:
app: ngx-dep
spec:
containers:
- image: nginx:alpine
name: nginx
ports:
- containerPort: 80
|
一个Deployment api，对应N个Pods，实际上Deployment就是更高级的ReplicaSet：
1
2
3
|
$ kubectl get rs
NAME DESIRED CURRENT READY AGE
nginx-deployment-7759cfdc55 3 3 3 4m59s
|
可以看到Pods被自动调度了
1
2
3
4
5
|
$ kubectl get pods -o wide
NAME READY STATUS RESTARTS AGE IP NODE NOMINATED NODE READINESS GATES
nginx-deployment-7759cfdc55-66j5g 1/1 Running 0 5m46s 10.244.1.77 master <none> <none>
nginx-deployment-7759cfdc55-77wh5 1/1 Running 0 5m46s 10.244.1.78 master <none> <none>
nginx-deployment-7759cfdc55-snj68 1/1 Running 0 5m46s 10.244.0.27 engine-macmini <none> <none>
|
相比Job多了replicas、selector字段
在Kubernetes集群中运行多少个Pod实例，这里的话k8s会创建两个Pod实例
并且Kubernetes会持续监视Pods，如果发现个数不一样，就会新建，会一直保持Pod个数跟配置文件的一样
selector的作用是选出要被Deployment管理的Pod对象，matchLabels里的label要跟下面Pod的label一样，比如这里都是ngx-dep
这里之所以要写两次一样的，是因为下面的Pod除了被Deployment调度，还可能被其他API对象管理，比如负载均衡的Service等。之前的Job是离线对象，Pod基本都是一次性的，所以一个Job绑一个Pod即可。
1
2
3
|
$ kubectl get deploy
NAME READY UP-TO-DATE AVAILABLE AGE
nginx-deployment 3/3 3 3 18m
|
1
2
3
4
5
|
kubectl get pods
NAME READY STATUS RESTARTS AGE
nginx-deployment-f4488db8f-5vkhr 1/1 Running 0 116s
nginx-deployment-f4488db8f-bl9xv 1/1 Running 0 97s
nginx-deployment-f4488db8f-tflhd 1/1 Running 0 2m14s
|
如果删掉第一个：
1
2
|
$ kubectl delete pod nginx-deployment-f4488db8f-5vkhr
pod "nginx-deployment-f4488db8f-5vkhr" deleted
|
Kubernetes会自动创建一个Pod
1
2
3
4
5
|
$ kubectl get pods
NAME READY STATUS RESTARTS AGE
nginx-deployment-f4488db8f-bl9xv 1/1 Running 0 2m21s
nginx-deployment-f4488db8f-pq7rk 1/1 Running 0 14s
nginx-deployment-f4488db8f-tflhd 1/1 Running 0 2m58s
|
还可以随时调整Pod的数量：
1
2
3
|
# 把ngx-dep调整到5个
$ kubectl scale --replicas=2 deploy nginx-deployment
deployment.apps/nginx-deployment scaled
|
不过更好的方式是修改yaml的配置再重新apply
如果对应用进行更新，比如更新image或者label，就会出发Deployment的滚动更新。
1
2
3
|
$ kubectl set image deployment nginx-deployment nginx=nginx:1.9.1
deployment.apps/nginx-deployment image updated
|
会重新起新版本的对应数量的pod
1
2
3
4
|
$ kubectl get rs
NAME DESIRED CURRENT READY AGE
nginx-deployment-7759cfdc55 0 0 0 18m
nginx-deployment-f4488db8f 3 3 3 73s
|
看看Deployment的详细信息
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
|
$ kubectl describe deploy nginx-deployment
Name: nginx-deployment
Namespace: k8s-learning
CreationTimestamp: Tue, 06 Dec 2022 02:05:01 +0800
Labels: app=nginx
Annotations: deployment.kubernetes.io/revision: 2
kubernetes.io/change-cause: kubectl set image deployment nginx-deployment nginx=nginx:1.9.1 --record=true
Selector: app=nginx
Replicas: 2 desired | 2 updated | 2 total | 2 available | 0 unavailable
StrategyType: RollingUpdate
MinReadySeconds: 0
RollingUpdateStrategy: 25% max unavailable, 25% max surge
Pod Template:
Labels: app=nginx
Containers:
nginx:
Image: nginx:1.9.1
Port: 80/TCP
Host Port: 0/TCP
Environment: <none>
Mounts: <none>
Volumes: <none>
Conditions:
Type Status Reason
---- ------ ------
Progressing True NewReplicaSetAvailable
Available True MinimumReplicasAvailable
OldReplicaSets: <none>
NewReplicaSet: nginx-deployment-f4488db8f (2/2 replicas created)
Events:
Type Reason Age From Message
---- ------ ---- ---- -------
Normal ScalingReplicaSet 21m deployment-controller Scaled up replica set nginx-deployment-7759cfdc55 to 3
Normal ScalingReplicaSet 5m deployment-controller Scaled up replica set nginx-deployment-f4488db8f to 1
Normal ScalingReplicaSet 4m42s deployment-controller Scaled down replica set nginx-deployment-7759cfdc55 to 2 from 3
Normal ScalingReplicaSet 4m42s deployment-controller Scaled up replica set nginx-deployment-f4488db8f to 2 from 1
Normal ScalingReplicaSet 4m23s deployment-controller Scaled down replica set nginx-deployment-7759cfdc55 to 1 from 2
Normal ScalingReplicaSet 4m23s deployment-controller Scaled up replica set nginx-deployment-f4488db8f to 3 from 2
Normal ScalingReplicaSet 4m22s deployment-controller Scaled down replica set nginx-deployment-7759cfdc55 to 0 from 1
Normal ScalingReplicaSet 86s deployment-controller Scaled down replica set nginx-deployment-f4488db8f to 2 from 3
|
1
|
kubectl delete deployment <dep_name>
|
Deployment不会关心这些Pod在哪些节点上运行，只关心Pod的数量等。
然而一些应用跟节点是有关的，比如网络类应用、监控类应用、日志类应用、安全类应用，需要在每个节点都进行部署。
DaemonSet的目的是在集群的每个符合匹配要求的节点上都运行且仅运行一个Pod，就像看门狗一样。
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
|
apiVersion: apps/v1
kind: DaemonSet
metadata:
name: redis-ds
labels:
app: redis-ds
spec:
selector:
matchLabels:
name: redis-ds
template:
metadata:
labels:
name: redis-ds
spec:
containers:
- image: redis:5-alpine
name: redis
ports:
- containerPort: 6379
|
跟Deployment相比其实就少了replicas字段
1
2
3
|
$ kubectl get ds
NAME DESIRED CURRENT READY UP-TO-DATE AVAILABLE NODE SELECTOR AGE
redis-ds 2 2 1 2 1 <none> 10s
|
1
2
3
4
|
$ kubectl get pod -o wide
NAME READY STATUS RESTARTS AGE IP NODE NOMINATED NODE READINESS GATES
redis-ds-g57tq 1/1 Running 0 44s 192.168.219.115 master <none> <none>
redis-ds-kq5wn 1/1 Running 0 44s 192.168.244.76 engine-macmini <none> <none>
|
会自己查找集群的节点，然后部署