Deployment
1. 简介
例如如下的部署文件：
apiVersion: apps/v1
kind: Deployment
metadata:
name: nginx-deployment
labels:
app: nginx
spec:
replicas: 3
selector:
matchLabels:
app: nginx
template:
metadata:
labels:
app: nginx
spec:
containers:
- name: nginx
image: nginx:1.7.9
ports:
- containerPort: 80
当我们在 Kubernetes 集群中创建上述 Deployment 对象时，它不只会创建 Deployment 资源，还会创建另外的 ReplicaSet 以及三个 Pod 对象：
$ kubectl get deployments.apps
NAME READY UP-TO-DATE AVAILABLE AGE
nginx-deployment 3/3 3 3 6m55s
$ kubectl get replicasets.apps
NAME DESIRED CURRENT READY AGE
nginx-deployment-76bf4969df 3 3 3 7m27s
$ kubectl get pods
NAME READY STATUS RESTARTS AGE
nginx-deployment-76bf4969df-58gxj 1/1 Running 0 7m42s
nginx-deployment-76bf4969df-9jgk9 1/1 Running 0 7m42s
nginx-deployment-76bf4969df-m4pkg 1/1 Running 0 7m43s
每一个 Deployment 都会和它的依赖组成以下的拓扑结构:
graph TD;
A(Deployment) --> B(ReplicaSet);
B(ReplicaSet) --> C(Pod1);
B(ReplicaSet) --> D(Pod2);
B(ReplicaSet) --> E(Pod3);
2. DeploymentController
所有的 Deployment 对象都是由 Controller Manager 集群中的 DeploymentController 进行管理。
DeploymentController 作为管理 Deployment 资源的控制器，会在启动时通过 Informer 监听三种不同资源的通知：Pod、ReplicaSet 和 Deployment，这三种资源的变动都会触发 DeploymentController 中的回调。
graph TD;
A(Deployment Informer) --> |Add/Update/Delete| D(DeploymentController);
B(ReplicaSet Informer) --> |Add/Update/Delete| D(DeploymentController);
C(Pod Informer) --> |Delete| D(DeploymentController);
StatefulSet
1. 简介
StatefulSet 是用于管理有状态应用的工作负载对象，不仅能管理 Pod 的对象，还它能够保证这些 Pod 的顺序性和唯一性。
例如如下的部署文件：
apiVersion: apps/v1
kind: StatefulSet
metadata:
name: web
spec:
serviceName: "nginx"
replicas: 2
selector:
matchLabels:
app: nginx
template:
metadata:
labels:
app: nginx
spec:
containers:
- name: nginx
image: k8s.gcr.io/nginx-slim:0.8
volumeMounts:
- name: www
mountPath: /usr/share/nginx/html
volumeClaimTemplates:
- metadata:
name: www
spec:
accessModes: [ "ReadWriteOnce" ]
resources:
requests:
storage: 1Gi
如果我们在 Kubernetes 集群中创建如上所示的 StatefulSet 对象，会得到以下结果：
$ kubectl get statefulsets.apps
kNAME READY AGE
web 2/2 2m27s
$ kubectl get pods
NAME READY STATUS RESTARTS AGE
web-0 1/1 Running 0 2m31s
web-1 1/1 Running 0 105s
$ kubectl get persistentvolumes
NAME CAPACITY ACCESS MODES RECLAIM POLICY STATUS CLAIM STORAGECLASS REASON AGE
pvc-19ef374f-39d1-11e9-b870-9efb418608da 1Gi RWO Delete Bound default/www-web-1 do-block-storage 21m
pvc-fe53d5f7-39d0-11e9-b870-9efb418608da 1Gi RWO Delete Bound default/www-web-0 do-block-storage 21m
$ kubectl get persistentvolumeclaims
NAME STATUS VOLUME CAPACITY ACCESS MODES STORAGECLASS AGE
www-web-0 Bound pvc-fe53d5f7-39d0-11e9-b870-9efb418608da 1Gi RWO do-block-storage 21m
www-web-1 Bound pvc-19ef374f-39d1-11e9-b870-9efb418608da 1Gi RWO do-block-storage 21m
两个 Pod 对象名中包含了它们的序列号，该序列号会在 StatefulSet 存在的时间内保持不变，哪怕 Pod 被重启或者重新调度，也不会出现任何的改变。
graph TD;
A(StatefulSet) --> B(Pod1);
A(StatefulSet) --> C(Pod2);
B(Pod1) --> D(PersistentVolume1);
D(PersistentVolume1) --> E(PersistentVolumeClaim1);
C(Pod2) --> F(PersistentVolume2);
F(PersistentVolume2) --> G(PersistentVolumeClaim2);
2. StatefulSetController
StatefulSetController 会同时从 PodInformer 和 ReplicaSetInformer 中接受增删改事件并将事件推送到队列中。
graph TD;
A(Pod Informer) --> |Add/Update/Delete| C(StatefulSetController);
B(ReplicaSet Informer) --> |Add/Update/Delete| C(StatefulSetController);
C(StatefulSetController) --> |Add StatefulSet| D(queue);
DaemonSet
1. 简介
DaemonSet 可以保证集群中所有的或者部分的节点都能够运行同一份 Pod 副本，每当有新的节点被加入到集群时，Pod 就会在目标的节点上启动，如果节点被从集群中剔除，节点上的 Pod 也会被垃圾收集器清除；DaemonSet 的作用就像是计算机中的守护进程，它能够运行集群存储、日志收集和监控等『守护进程』，这些服务一般是集群中必备的基础服务。
例如如下的部署文件：
apiVersion: apps/v1
kind: DaemonSet
metadata:
name: fluentd-elasticsearch
namespace: kube-system
spec:
selector:
matchLabels:
name: fluentd-elasticsearch
template:
metadata:
labels:
name: fluentd-elasticsearch
spec:
containers:
- name: fluentd-elasticsearch
image: k8s.gcr.io/fluentd-elasticsearch:1.20
volumeMounts:
- name: varlog
mountPath: /var/log
- name: varlibdockercontainers
mountPath: /var/lib/docker/containers
readOnly: true
volumes:
- name: varlog
hostPath:
path: /var/log
- name: varlibdockercontainers
hostPath:
path: /var/lib/docker/containers
在 Kubernetes 集群的 kube-system 命名空间中创建 DaemonSet 资源并在所有的节点上创建新的 Pod：
$ kubectl get daemonsets.apps fluentd-elasticsearch --namespace kube-system
NAME DESIRED CURRENT READY UP-TO-DATE AVAILABLE NODE SELECTOR AGE
fluentd-elasticsearch 1 1 1 1 1 <none> 19h
$ kubectl get pods --namespace kube-system --label name=fluentd-elasticsearch
NAME READY STATUS RESTARTS AGE
fluentd-elasticsearch-kvtwj 1/1 Running 0 19h
2. DaemonSetsController
用于管理 DaemonSet 的控制器是 DaemonSetsController，该控制器会监听 DaemonSet、ControllerRevision [版本信息]、Pod 和 Node 资源的变动。
graph LR;
DaemonSetsController -.-> Worker;
DaemonSetsController -.-> Worker;
DaemonSetsController -.-> Worker;
Queue --> Worker;
Queue --> Worker;
Queue --> Worker;
DaemonSet Informer --> Queue;
ControllerRevision --> Queue;
Pod Informer --> Queue;
Node Informer --> Queue;
