【k8s资源调度-StatefulSet】
k8s的statefulset 资源调度
·
1、部署对象StatefulSet资源（无状态应用）
StatefulSet针对的是有状态应用，有状态应用会对我们的当前pod的网络、文件系统等有关联。
2、配置文件如下
StatefulSet资源的配置文件粗略如下，如下的配置信息包含了数据卷，目前还未涉及到数据卷的信息，可以在创建StatefulSet的时候把数据卷的信息先删除。如果不删除这个数据卷的信息，创建的StatefulSet会卡在没有数据卷的信息导致无法ready。
---
apiversion: v1
kind: Service
metadata:
name: nginx
labels:
app: nginx
spec:
ports:
- port: 80
name: web
clusterIP: None
selector:
app: nginx
---
apiVersion: apps/v1 # StatefulSet 资源的版本
kind: StatefulSet # StatefulSet 类型的资源
metadata: # 元数据
name: web # StatefulSet的名字
spec:
serviceName: "nginx" # 使用哪个service服务来管理dns
replicas: 2 # 期望副本数
selector: # 选择器
matchLabels: # 按照标签匹配
app: nginx # 选择器名字
template: # pod 的模板
metadata: # 元数据
labels: # pod的标签
app: nginx # 标签值是：app=nginx
spec: # pod的期望信息
containers: # pod的容器信息
- name: nginx # pod的名字
image: nginx:1.20 # 容器镜像信息
ports: # 端口
- containerPort: 80 # 容器的端口
name: web # 容器内端口的名字
# 以下信息可以不先不使用 -----------------------------------------------
volumeMounts: # 加载一个数据卷
- name: www # 加载的数据卷的名字是www
mountPath: /usr/share/nginx/html # 加载到容器中的哪个目录
volumeclaimTemplates: # 数据卷模板
metadata: #数据卷描述
name: www #数据卷的名称
annotations: #数据卷的注解
volume.alpha.kubernetes.io/storage-class:anything
spec: #数据卷的规约
accessModes:["ReadWriteOnce"] # 访问模式
resources:
requests:
storage: 1Gi # 需要1G的存储资源
3、创建StatefulSet资源
3.1 根据上文的配置文件创建StatefulSet资源
[root@k8s-master ~]# kubectl create -f web.yaml
service/nginx created
statefulset.apps/web created
3.2 查看创建的StatefulSet资源的信息
通过get选项可以看到statefulset的资源已经创建完成
# 查看statefulset的信息
[root@k8s-master ~]# kubectl get sts
NAME READY AGE
web 2/2 58s
3.3 查看pod信息（pod名字是有序的）
这时候pod的名字很有序，不像deploy创建的pod，名字都是一个随机值。
# 查看pod的信息，这时候pod的名字很有序，不像deploy创建的pod，名字都是一个随机值。
[root@k8s-master ~]# kubectl get pod
NAME READY STATUS RESTARTS AGE
web-0 1/1 Running 0 64s
web-1 1/1 Running 0 63s
3.4 查看service信息
查看service的信息，我们刚创建的service名字是nginx，目前service还是无ip的信息，因为我们目前还未做具体的绑定的操作。（绑定service和外部的信息）目前只是映射到容器内而已，我们有一个基于服务名访问就可以了（可以通过k8s集群中创建一个新的pod来访问）。
# 查看service的信息，我们刚创建的service名字是nginx，目前service还是无ip的信息，因为我们目前还未做具体的绑定的操作。（绑定service和外部的信息）目前只是映射到容器内而已，我们有一个基于服务名访问就可以了
[root@k8s-master ~]# kubectl get svc
NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
kubernetes ClusterIP 10.1.0.1 <none> 443/TCP 4d4h
nginx ClusterIP None <none> 80/TCP 70s
3.5 查看StatefulSet资源的描述信息
从如下的信息中可以看到，创建的StatefulSet资源已经创建成功
[root@k8s-master ~]# kubectl describe sts web
Name: web
Namespace: default
CreationTimestamp: Sat, 24 Feb 2024 02:20:14 +0800
Selector: app=nginx
Labels: <none>
Annotations: <none>
Replicas: 2 desired | 2 total
Update Strategy: RollingUpdate
Partition: 0
Pods Status: 2 Running / 0 Waiting / 0 Succeeded / 0 Failed
Pod Template:
Labels: app=nginx
Containers:
nginx:
Image: nginx:1.20
Port: 80/TCP
Host Port: 0/TCP
Environment: <none>
Mounts: <none>
Volumes: <none>
Volume Claims: <none>
Events:
Type Reason Age From Message
---- ------ ---- ---- -------
Normal SuccessfulCreate 3m4s statefulset-controller create Pod web-0 in StatefulSet web successful
Normal SuccessfulCreate 3m3s statefulset-controller create Pod web-1 in StatefulSet web successful
3.6 通过busybox这个镜像来测试pod的dns是否正常
busybox 是linux内的一个工具的镜像，这个工具镜像内会有我们常用的一些工具。
StatefulSet中每个Pod的DNS格式为（.namespace.svc.cluster.local
可以省略）
statefulSetName-{O…N-1).serviceName.namespace.svc.cluster.local
[root@k8s-master ~]# kubectl run -it --image busybox:1.28.4 dns-test /bin/sh
If you don't see a command prompt, try pressing enter.
/ # ping web-0.nginx
PING web-0.nginx (10.2.1.35): 56 data bytes
64 bytes from 10.2.1.35: seq=0 ttl=64 time=0.129 ms
64 bytes from 10.2.1.35: seq=1 ttl=64 time=0.225 ms
64 bytes from 10.2.1.35: seq=2 ttl=64 time=0.134 ms
64 bytes from 10.2.1.35: seq=3 ttl=64 time=0.127 ms
^C
--- web-0.nginx ping statistics ---
4 packets transmitted, 4 packets received, 0% packet loss
round-trip min/avg/max = 0.127/0.153/0.225 ms
/ # ping web-1.nginx
PING web-1.nginx (10.2.2.18): 56 data bytes
64 bytes from 10.2.2.18: seq=0 ttl=62 time=0.736 ms
64 bytes from 10.2.2.18: seq=1 ttl=62 time=1.568 ms
64 bytes from 10.2.2.18: seq=2 ttl=62 time=1.469 ms
64 bytes from 10.2.2.18: seq=3 ttl=62 time=0.795 ms
^C
--- web-1.nginx ping statistics ---
4 packets transmitted, 4 packets received, 0% packet loss
round-trip min/avg/max = 0.736/1.142/1.568 ms
## 以下信息可以查看看pod的dns 域名所绑定的信息
/ # nslookup web-0.nginx
Server: 10.1.0.10
Address 1: 10.1.0.10 kube-dns.kube-system.svc.cluster.local
Name: web-0.nginx
Address 1: 10.2.1.35 web-0.nginx.default.svc.cluster.local
/ # nslookup web-1.nginx
Server: 10.1.0.10
Address 1: 10.1.0.10 kube-dns.kube-system.svc.cluster.local
Name: web-1.nginx
Address 1: 10.2.2.18 web-1.nginx.default.svc.cluster.local
/ # exit
Session ended, resume using 'kubectl attach dns-test -c dns-test -i -t' command when the pod is running
4、扩容和缩容
扩容和缩容两种写法
kubectl scale statefulset web --replicas=3
kubectl patch statefulset web - p ‘{“spec”:{“replicas”:3))’
4.1 扩容
本文3.1小节创建了statefulset资源是2个，现在扩容到5个，如下，扩容也是有序的扩容
[root@k8s-master ~]# kubectl scale statefulset web --replicas=5
statefulset.apps/web scaled
# 查看sts信息
[root@k8s-master ~]# kubectl get sts
NAME READY AGE
web 5/5 39m
# 查看pod信息，有序的5个资源
[root@k8s-master ~]# kubectl get pod
NAME READY STATUS RESTARTS AGE
dns-test 1/1 Running 1 (6m51s ago) 13m
web-0 1/1 Running 0 39m
web-1 1/1 Running 0 39m
web-2 1/1 Running 0 21s
web-3 1/1 Running 0 19s
web-4 1/1 Running 0 18s
# 查看sts的资源描述
[root@k8s-master ~]# kubectl describe sts web
Name: web
Namespace: default
CreationTimestamp: Sat, 24 Feb 2024 02:20:14 +0800
Selector: app=nginx
Labels: <none>
Annotations: <none>
Replicas: 5 desired | 5 total
Update Strategy: RollingUpdate
Partition: 0
Pods Status: 5 Running / 0 Waiting / 0 Succeeded / 0 Failed
Pod Template:
Labels: app=nginx
Containers:
nginx:
Image: nginx:1.20
Port: 80/TCP
Host Port: 0/TCP
Environment: <none>
Mounts: <none>
Volumes: <none>
Volume Claims: <none>
Events:
Type Reason Age From Message
---- ------ ---- ---- -------
Normal SuccessfulCreate 39m statefulset-controller create Pod web-0 in StatefulSet web successful
Normal SuccessfulCreate 39m statefulset-controller create Pod web-1 in StatefulSet web successful
Normal SuccessfulCreate 48s statefulset-controller create Pod web-2 in StatefulSet web successful
Normal SuccessfulCreate 46s statefulset-controller create Pod web-3 in StatefulSet web successful
Normal SuccessfulCreate 45s statefulset-controller create Pod web-4 in StatefulSet web successful
[root@k8s-master ~]#
4.2 缩容
上面小节中创建了5个副本，现在缩容到3个，如下，缩容也是有序的缩容
[root@k8s-master ~]# kubectl scale statefulset web --replicas=3
statefulset.apps/web scaled
[root@k8s-master ~]# kubectl get sts
NAME READY AGE
web 3/3 42m
[root@k8s-master ~]# kubectl get pod
NAME READY STATUS RESTARTS AGE
dns-test 1/1 Running 1 (10m ago) 16m
web-0 1/1 Running 0 42m
web-1 1/1 Running 0 42m
web-2 1/1 Running 0 3m38s
[root@k8s-master ~]# kubectl describe sts web
Name: web
Namespace: default
CreationTimestamp: Sat, 24 Feb 2024 02:20:14 +0800
Selector: app=nginx
Labels: <none>
Annotations: <none>
Replicas: 3 desired | 3 total
Update Strategy: RollingUpdate
Partition: 0
Pods Status: 3 Running / 0 Waiting / 0 Succeeded / 0 Failed
Pod Template:
Labels: app=nginx
Containers:
nginx:
Image: nginx:1.20
Port: 80/TCP
Host Port: 0/TCP
Environment: <none>
Mounts: <none>
Volumes: <none>
Volume Claims: <none>
Events:
Type Reason Age From Message
---- ------ ---- ---- -------
Normal SuccessfulCreate 43m statefulset-controller create Pod web-0 in StatefulSet web successful
Normal SuccessfulCreate 43m statefulset-controller create Pod web-1 in StatefulSet web successful
Normal SuccessfulCreate 4m33s statefulset-controller create Pod web-2 in StatefulSet web successful
Normal SuccessfulCreate 4m31s statefulset-controller create Pod web-3 in StatefulSet web successful
Normal SuccessfulCreate 4m30s statefulset-controller create Pod web-4 in StatefulSet web successful
Normal SuccessfulDelete 64s statefulset-controller delete Pod web-4 in StatefulSet web successful
Normal SuccessfulDelete 62s statefulset-controller delete Pod web-3 in StatefulSet web successful
5、镜像更新
- 镜像更新（目前还不支持直接更新image,需要patch来间接实现)
5.1 更新方式-RollingUpdate
StatefulSet也可以采用滚动更新策略，同样是修改pod，template属性后会触发更新，但是由于od是有序的，在StatefulSet中更新时是基于pod的顺序倒序更新的
5.1.1 更新镜像信息
[root@k8s-master ~]# kubectl patch sts web --type='json' -p='[{"op": "replace","path": "/spec/template/spec/containers/0/image","value": "nginx:1.21"}]'
statefulset.apps/web patched
5.1.2 查看statefulset的状态
[root@k8s-master ~]# kubectl get sts web
NAME READY AGE
web 3/3 54m
5.1.3 查看pod的状态
[root@k8s-master ~]# kubectl get po
NAME READY STATUS RESTARTS AGE
dns-test 1/1 Running 1 (21m ago) 28m
web-0 1/1 Running 0 25s
web-1 1/1 Running 0 27s
web-2 1/1 Running 0 30s
5.1.4 查看statefulset的描述信息
可以从最后的events信息看到，更新pod的时候是从序号大的更新到序号小的。
[root@k8s-master ~]# kubectl describe sts web
Name: web
Namespace: default
CreationTimestamp: Sat, 24 Feb 2024 02:20:14 +0800
Selector: app=nginx
Labels: <none>
Annotations: <none>
Replicas: 3 desired | 3 total
Update Strategy: RollingUpdate
Partition: 0
Pods Status: 3 Running / 0 Waiting / 0 Succeeded / 0 Failed
Pod Template:
Labels: app=nginx
Containers:
nginx:
Image: nginx:1.21
Port: 80/TCP
Host Port: 0/TCP
Environment: <none>
Mounts: <none>
Volumes: <none>
Volume Claims: <none>
Events:
Type Reason Age From Message
---- ------ ---- ---- -------
Normal SuccessfulCreate 15m statefulset-controller create Pod web-3 in StatefulSet web successful
Normal SuccessfulCreate 15m statefulset-controller create Pod web-4 in StatefulSet web successful
Normal SuccessfulDelete 12m statefulset-controller delete Pod web-4 in StatefulSet web successful
Normal SuccessfulDelete 12m statefulset-controller delete Pod web-3 in StatefulSet web successful
Normal SuccessfulDelete 46s statefulset-controller delete Pod web-2 in StatefulSet web successful
Normal SuccessfulCreate 45s (x2 over 15m) statefulset-controller create Pod web-2 in StatefulSet web successful
Normal SuccessfulDelete 43s statefulset-controller delete Pod web-1 in StatefulSet web successful
Normal SuccessfulCreate 42s (x2 over 54m) statefulset-controller create Pod web-1 in StatefulSet web successful
Normal SuccessfulDelete 41s statefulset-controller delete Pod web-0 in StatefulSet web successful
Normal SuccessfulCreate 40s (x2 over 54m) statefulset-controller create Pod web-0 in StatefulSet web successful
5.1.5 查看statefulset的历史更新信息
[root@k8s-master ~]# kubectl rollout history sts web --revision=0
statefulset.apps/web
REVISION CHANGE-CAUSE
1 <none>
2 <none>
[root@k8s-master ~]# kubectl rollout history sts web --revision=2
statefulset.apps/web with revision #2
Pod Template:
Labels: app=nginx
Containers:
nginx:
Image: nginx:1.21
Port: 80/TCP
Host Port: 0/TCP
Environment: <none>
Mounts: <none>
Volumes: <none>
[root@k8s-master ~]#
5.1.6 利用滚动更新的属性可以实现–灰度发布/金丝雀发布
灰度发布（金丝雀发布）是指在测试环境测试正常，然后上线到正式环境，通过在线上部分服务器发布，验证程序的可靠性，最后通过滚动发布，所有的资源都全部更新。
目标：将项目上线后产生的问题，尽量降到最低。
核心：如何控制每次只更新那些版本，就需要利用partition字段。
- 利用滚动更新中的partition属性，可以实现简易的灰度发布的效果。
- 例如我们有5个pod,如果当前partition设置为3，那么此时滚动更新时，只会更新那些序号>=3的pod。
- 利用该机制，我们可以通过控制partition的值，来决定只更新其中一部分pod,确认没有问题后再主键增大更新的pod数量，最终实现全部pod更新。
通过 【kubectl edit sts web】查看statefulset中多了一个滚动更新的属性 partition
5.1.6.1 扩容到5个节点
[root@k8s-master ~]# kubectl scale statefulset web --replicas=5
statefulset.apps/web scaled
[root@k8s-master ~]# kubectl get statefulsets.apps web
NAME READY AGE
web 4/5 10h
[root@k8s-master ~]# kubectl get statefulsets.apps web
NAME READY AGE
web 5/5 10h
5.1.6.2 修改partition的值，默认是0
通过 kubectl edit sts web 修改
5.1.6.3 修改image的版本信息
5.1.6.4 查看statefulset的状态
通过以下已经查看到statefulset已经更新完成了，ready的状态
[root@k8s-master ~]# kubectl get statefulsets.apps web
NAME READY AGE
web 4/5 10h
[root@k8s-master ~]# kubectl get statefulsets.apps web
NAME READY AGE
web 5/5 10h
5.1.6.5 查看pod的信息
刚才我们设置的partition的数字为3，总共的副本是5个，通过以下信息可以查看到，更新了web-4和web-3，web-2、web-1、web-0都没有更新。
[root@k8s-master ~]# kubectl describe pod web-4 |grep Image
Image: nginx:1.20
Image ID: docker-pullable://nginx@sha256:03f3cb0afb7bd5c76e01bfec0ce08803c495348dccce37bcb82c347b4853c00b
[root@k8s-master ~]# kubectl describe pod web-3 |grep Image
Image: nginx:1.20
Image ID: docker-pullable://nginx@sha256:03f3cb0afb7bd5c76e01bfec0ce08803c495348dccce37bcb82c347b4853c00b
[root@k8s-master ~]# kubectl describe pod web-2 |grep Image
Image: nginx:1.21
Image ID: docker-pullable://nginx@sha256:0d17b565c37bcbd895e9d92315a05c1c3c9a29f762b011a10c54a66cd53c9b31
[root@k8s-master ~]# kubectl describe pod web-1 |grep Image
Image: nginx:1.21
Image ID: docker-pullable://nginx@sha256:0d17b565c37bcbd895e9d92315a05c1c3c9a29f762b011a10c54a66cd53c9b31
[root@k8s-master ~]# kubectl describe pod web-0 |grep Image
Image: nginx:1.21
Image ID: docker-pullable://nginx@sha256:0d17b565c37bcbd895e9d92315a05c1c3c9a29f762b011a10c54a66cd53c9b31
5.1.6.5 查看更新的历史版本内容信息
[root@k8s-master ~]# kubectl rollout history sts web --revision=0
statefulset.apps/web
REVISION CHANGE-CAUSE
2 <none>
3 <none>
# resivion 3的更新内容为 Image: nginx:1.20
[root@k8s-master ~]# kubectl rollout history sts web --revision=3
statefulset.apps/web with revision #3
Pod Template:
Labels: app=nginx
Containers:
nginx:
Image: nginx:1.20
Port: 80/TCP
Host Port: 0/TCP
Environment: <none>
Mounts: <none>
Volumes: <none>
5.1.6.6 查看statefulset的描述信息
通过evevts，可以看到了statefulset资源只更新了web-4和web-3
[root@k8s-master ~]# kubectl describe sts web
Name: web
Namespace: default
CreationTimestamp: Sat, 24 Feb 2024 02:20:14 +0800
Selector: app=nginx
Labels: <none>
Annotations: <none>
Replicas: 5 desired | 5 total
Update Strategy: RollingUpdate
Partition: 3
Pods Status: 5 Running / 0 Waiting / 0 Succeeded / 0 Failed
Pod Template:
Labels: app=nginx
Containers:
nginx:
Image: nginx:1.20
Port: 80/TCP
Host Port: 0/TCP
Environment: <none>
Mounts: <none>
Volumes: <none>
Volume Claims: <none>
Events:
Type Reason Age From Message
---- ------ ---- ---- -------
Normal SuccessfulDelete 9m6s (x2 over 9h) statefulset-controller delete Pod web-4 in StatefulSet web successful
Normal SuccessfulCreate 9m4s (x3 over 9h) statefulset-controller create Pod web-4 in StatefulSet web successful
Normal SuccessfulDelete 9m2s (x2 over 9h) statefulset-controller delete Pod web-3 in StatefulSet web successful
Normal SuccessfulCreate 9m1s (x3 over 9h) statefulset-controller create Pod web-3 in StatefulSet web successful
5.1.6.7 滚动更新其他资源
- 刚才我们是更新了2个pod资源，还剩下3个pod资源没有更新，把partition的值设置为0的时候，其他资源就全部更新了。
- 实际生产环境中，灰度发布可以根据实际使用情况使用，可以多次灰度发布。
[root@k8s-master ~]# kubectl describe sts web
Name: web
Namespace: default
CreationTimestamp: Sat, 24 Feb 2024 02:20:14 +0800
Selector: app=nginx
Labels: <none>
Annotations: <none>
Replicas: 5 desired | 5 total
Update Strategy: RollingUpdate
Partition: 0
Pods Status: 5 Running / 0 Waiting / 0 Succeeded / 0 Failed
Pod Template:
Labels: app=nginx
Containers:
nginx:
Image: nginx:1.20
Port: 80/TCP
Host Port: 0/TCP
Environment: <none>
Mounts: <none>
Volumes: <none>
Volume Claims: <none>
Events:
Type Reason Age From Message
---- ------ ---- ---- -------
Normal SuccessfulDelete 12m (x2 over 9h) statefulset-controller delete Pod web-4 in StatefulSet web successful
Normal SuccessfulCreate 12m (x3 over 9h) statefulset-controller create Pod web-4 in StatefulSet web successful
Normal SuccessfulDelete 12m (x2 over 9h) statefulset-controller delete Pod web-3 in StatefulSet web successful
Normal SuccessfulCreate 12m (x3 over 9h) statefulset-controller create Pod web-3 in StatefulSet web successful
Normal SuccessfulCreate 8s (x3 over 9h) statefulset-controller create Pod web-2 in StatefulSet web successful
Normal SuccessfulDelete 8s (x2 over 9h) statefulset-controller delete Pod web-2 in StatefulSet web successful
Normal SuccessfulDelete 6s (x2 over 9h) statefulset-controller delete Pod web-1 in StatefulSet web successful
Normal SuccessfulCreate 4s (x3 over 10h) statefulset-controller create Pod web-1 in StatefulSet web successful
Normal SuccessfulDelete 2s (x2 over 9h) statefulset-controller delete Pod web-0 in StatefulSet web successful
Normal SuccessfulCreate 1s (x3 over 10h) statefulset-controller create Pod web-0 in StatefulSet web successful
5.2 更新方式-OnDelete
OnDelete的更新策略是删除pod后才会更新，修改template模板的时候并不会更新。
5.2.1 修改更新策略
5.2.2 修改镜像版本信息
5.2.3 镜像更新信息
[root@k8s-master ~]# kubectl rollout history sts web --revision=0
statefulset.apps/web
REVISION CHANGE-CAUSE
3 <none>
4 <none>
[root@k8s-master ~]# kubectl rollout history sts web --revision=4
statefulset.apps/web with revision #4
Pod Template:
Labels: app=nginx
Containers:
nginx:
Image: nginx:1.21
Port: 80/TCP
Host Port: 0/TCP
Environment: <none>
Mounts: <none>
Volumes: <none>
5.2.4 查看pod，pod暂时未更新
[root@k8s-master ~]# kubectl describe pod web-4 | grep Image
Image: nginx:1.20
Image ID: docker-pullable://nginx@sha256:03f3cb0afb7bd5c76e01bfec0ce08803c495348dccce37bcb82c347b4853c00b
[root@k8s-master ~]# kubectl describe pod web-3 | grep Image
Image: nginx:1.20
Image ID: docker-pullable://nginx@sha256:03f3cb0afb7bd5c76e01bfec0ce08803c495348dccce37bcb82c347b4853c00b
[root@k8s-master ~]# kubectl describe pod web-2 | grep Image
Image: nginx:1.20
Image ID: docker-pullable://nginx@sha256:03f3cb0afb7bd5c76e01bfec0ce08803c495348dccce37bcb82c347b4853c00b
[root@k8s-master ~]# kubectl describe pod web-1 | grep Image
Image: nginx:1.20
Image ID: docker-pullable://nginx@sha256:03f3cb0afb7bd5c76e01bfec0ce08803c495348dccce37bcb82c347b4853c00b
[root@k8s-master ~]# kubectl describe pod web-0 | grep Image
Image: nginx:1.20
Image ID: docker-pullable://nginx@sha256:03f3cb0afb7bd5c76e01bfec0ce08803c495348dccce37bcb82c347b4853c00b
5.2.5 删除一个pod，可以看到删除的pod信息已经更新
[root@k8s-master ~]# kubectl delete pod web-4
pod "web-4" deleted
[root@k8s-master ~]# kubectl get sts web
NAME READY AGE
web 5/5 10h
[root@k8s-master ~]# kubectl get pod web-4
NAME READY STATUS RESTARTS AGE
web-4 1/1 Running 0 21s
[root@k8s-master ~]# kubectl describe pod web-4 | grep Image
Image: nginx:1.21
Image ID: docker-pullable://nginx@sha256:0d17b565c37bcbd895e9d92315a05c1c3c9a29f762b011a10c54a66cd53c9b31
[root@k8s-master ~]# kubectl describe pod web-3 | grep Image
Image: nginx:1.20
Image ID: docker-pullable://nginx@sha256:03f3cb0afb7bd5c76e01bfec0ce08803c495348dccce37bcb82c347b4853c00b
5.2.6 滚动更新其他资源
[root@k8s-master ~]# kubectl delete pod web-3
pod "web-3" deleted
[root@k8s-master ~]# kubectl delete pod web-2
pod "web-2" deleted
[root@k8s-master ~]# kubectl delete pod web-1
pod "web-1" deleted
[root@k8s-master ~]# kubectl delete pod web-0
pod "web-0" deleted
[root@k8s-master ~]# kubectl describe sts web
Name: web
Namespace: default
CreationTimestamp: Sat, 24 Feb 2024 02:20:14 +0800
Selector: app=nginx
Labels: <none>
Annotations: <none>
Replicas: 5 desired | 5 total
Update Strategy: OnDelete
Pods Status: 5 Running / 0 Waiting / 0 Succeeded / 0 Failed
Pod Template:
Labels: app=nginx
Containers:
nginx:
Image: nginx:1.21
Port: 80/TCP
Host Port: 0/TCP
Environment: <none>
Mounts: <none>
Volumes: <none>
Volume Claims: <none>
Events:
Type Reason Age From Message
---- ------ ---- ---- -------
Normal SuccessfulDelete 27m (x2 over 9h) statefulset-controller delete Pod web-4 in StatefulSet web successful
Normal SuccessfulDelete 27m (x2 over 9h) statefulset-controller delete Pod web-3 in StatefulSet web successful
Normal SuccessfulDelete 14m (x2 over 9h) statefulset-controller delete Pod web-2 in StatefulSet web successful
Normal SuccessfulDelete 14m (x2 over 9h) statefulset-controller delete Pod web-1 in StatefulSet web successful
Normal SuccessfulDelete 14m (x2 over 9h) statefulset-controller delete Pod web-0 in StatefulSet web successful
Normal SuccessfulCreate 2m4s (x4 over 10h) statefulset-controller create Pod web-4 in StatefulSet web successful
## 以下信息可以看到其他的4个副本也同步更新了
Normal SuccessfulCreate 18s (x4 over 10h) statefulset-controller create Pod web-3 in StatefulSet web successful
Normal SuccessfulCreate 15s (x4 over 10h) statefulset-controller create Pod web-2 in StatefulSet web successful
Normal SuccessfulCreate 13s (x4 over 10h) statefulset-controller create Pod web-1 in StatefulSet web successful
Normal SuccessfulCreate 9s (x4 over 10h) statefulset-controller create Pod web-0 in StatefulSet web successful
5.3 有状态的资源删除方式
- 删除方式：删除statefulset和Headlass service
- 级联删除：删除statefulset时会同时制除pods
- kubectl delete statefulset web
- 非级联删除：删除statefulset时不会删除 pod ，删除sts后，pod不会删除
- kubectl deelte sts web --cascade=false
- 删除service
- kubectl delete service nginx
5.3.1 级联删除
5.3.1.1 删除statefulset资源
[root@k8s-master ~]# kubectl delete sts web
statefulset.apps "web" deleted
5.3.1.2 查看statefule资源和pod资源
[root@k8s-master ~]# kubectl get sts web
Error from server (NotFound): statefulsets.apps "web" not found
[root@k8s-master ~]# kubectl get pod
NAME READY STATUS RESTARTS AGE
dns-test 1/1 Running 1 (10h ago) 10h
5.3.1.3 查看service资源
[root@k8s-master ~]# kubectl get service
NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
kubernetes ClusterIP 10.1.0.1 <none> 443/TCP 4d15h
nginx ClusterIP None <none> 80/TCP 10h
5.3.1.4 删除service资源
[root@k8s-master ~]# kubectl delete svc nginx
service "nginx" deleted
[root@k8s-master ~]# kubectl get service
NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
kubernetes ClusterIP 10.1.0.1 <none> 443/TCP 4d15h
5.3.2 非级联删除
5.3.2.1 创建资源
由于刚才的资源我们已经删除，现在重新创建一个资源
[root@k8s-master ~]# kubectl create -f web.yaml
service/nginx created
statefulset.apps/web created
[root@k8s-master ~]# kubectl get sts web
NAME READY AGE
web 2/2 8s
[root@k8s-master ~]# kubectl get pod
NAME READY STATUS RESTARTS AGE
dns-test 1/1 Running 1 (10h ago) 10h
web-0 1/1 Running 0 14s
web-1 1/1 Running 0 12s
[root@k8s-master ~]# kubectl get service
NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
kubernetes ClusterIP 10.1.0.1 <none> 443/TCP 4d15h
nginx ClusterIP None <none> 80/TCP 25s
5.3.2.2 删除statefulset资源（添加选项 --cascade=false）
[root@k8s-master ~]# kubectl delete sts web --cascade=false
warning: --cascade=false is deprecated (boolean value) and can be replaced with --cascade=orphan.
statefulset.apps "web" deleted
5.3.2.3 查看statefule资源和pod资源
[root@k8s-master ~]# kubectl get sts web
Error from server (NotFound): statefulsets.apps "web" not found
[root@k8s-master ~]# kubectl get pod
NAME READY STATUS RESTARTS AGE
dns-test 1/1 Running 1 (10h ago) 10h
web-0 1/1 Running 0 3m15s
web-1 1/1 Running 0 3m13s
5.3.2.4 删除pod资源
[root@k8s-master ~]# kubectl get pod
NAME READY STATUS RESTARTS AGE
dns-test 1/1 Running 1 (10h ago) 10h
web-0 1/1 Running 0 3m15s
web-1 1/1 Running 0 3m13s
[root@k8s-master ~]# kubectl delete po web-0 web-1
pod "web-0" deleted
pod "web-1" deleted
5.3.2.5 删除service资源
[root@k8s-master ~]# kubectl get svc
NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
kubernetes ClusterIP 10.1.0.1 <none> 443/TCP 4d15h
nginx ClusterIP None <none> 80/TCP 4m29s
[root@k8s-master ~]# kubectl delete svc nginx
service "nginx" deleted
[root@k8s-master ~]# kubectl get svc
NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
kubernetes ClusterIP 10.1.0.1 <none> 443/TCP 4d15h
K8S/Kubernetes社区为您提供最前沿的新闻资讯和知识内容
更多推荐
26
0- 0
已为社区贡献18条内容
所有评论(0)