k8s pvc详解-526互联

一、概述
PVC 的全称是：PersistentVolumeClaim（持久化卷声明），PVC 是用户存储的一种声明，PVC 和 Pod 比较类似，Pod 消耗的是节点，PVC 消耗的是 PV 资源，Pod 可以请求 CPU 和内存，而 PVC 可以请求特定的存储空间和访问模式。对于真正使用存储的用户不需要关心底层的存储实现细节，只需要直接使用 PVC 即可。

PVC作为用户对存储资源的需求申请，主要涉及存储空间请求、访问模式、 PV选择条件和存储类别等信息的设置。

二、参数解释
创建pvc，如下：

PVC也可以设置访问模式，用于描述用户应用对存储资源的访问权限，其设置参数和pv一致

resources：资源请求

描述对存储的请求数量，通过 resources.requests.storage字段设置需要的存储空间大小。

volumeMode：存储卷格式，用于描述用户使用的PV存储卷模式，包括文件系统（Filesystem）和块设备 (Block)，与pv一致。

storageClassName：存储类型

在创建pvc时需要指定后端存储类型，以减少对后端存储详细信息的依赖。

如果系统中存在一个DefaultStorageClass，可以在创建pvc的时候不指定storageclassname，系统会自动使用默认的存储类型。

如果将多个StorageClass都定义为default，则由于不唯一，系统将无法创建PVC。

三、pod使用pvc
编辑配置文件，如下：

---
apiVersion: v1
kind: PersistentVolumeClaim
metadata:
name: test-pvc
spec:
accessModes:
- ReadWriteOnce
resources:
requests:
storage: 2Gi
storageClassName: nfs
volumeMode: Filesystem

---
apiVersion: v1
kind: Pod
metadata:
name: mypod
spec:
containers:
- name: myfrontend
image: docker.io/library/centos:7
command: ["init"]
imagePullPolicy: IfNotPresent
volumeMounts:
- mountPath: /data name: test-path
volumes:
- name: test-path
persistentVolumeClaim:
claimName: test-pvc

执行创建命令如下：
kubectl apply -f pod.yaml
[root@node1 yaml]# kubectl get po -o wide
NAME READY STATUS RESTARTS AGE IP NODE NOMINATED NODE READINESS GATES
mypod 1/1 Running 0 4s 10.233.96.10 node2 <none> <none>

进入pod查看挂载的目录，如下：
[root@node1 yaml]# kubectl exec -it mypod bash
kubectl exec [POD] [COMMAND] is DEPRECATED and will be removed in a future version. Use kubectl exec [POD] -- [COMMAND] instead.
[root@mypod /]# ls
anaconda-post.log bin data dev etc home lib lib64 media mnt opt proc root run sbin srv sys tmp usr var
[root@mypod /]# ll
total 12
-rw-r--r-- 1 root root 12114 Nov 13 2020 anaconda-post.log
lrwxrwxrwx 1 root root 7 Nov 13 2020 bin -> usr/bin
drwxrwxrwx 2 root root 6 Jun 5 06:27 data ####目录被自动挂载
drwxr-xr-x 5 root root 360 Jun 5 06:34 dev
drwxr-xr-x 1 root root 41 Jun 5 06:34 etc
drwxr-xr-x 2 root root 6 Apr 11 2018 home

四、pv、pvc状态变化
不同情况下，PV和PVC的状态变化如下：

具体操作如下：

#############################
1:创建pv
---
apiVersion: v1
kind: PersistentVolume
metadata:
name: pv-nfs
spec:
accessModes:
- ReadWriteOnce
capacity:
storage: 2Gi
nfs:
path: /data/
server: 192.168.5.79
persistentVolumeReclaimPolicy: Retain
storageClassName: nfs
volumeMode: Filesystem

[root@node1 yaml]# kubectl apply -f nfs-pv.yaml
persistentvolume/pv-nfs created
[root@node1 yaml]#
[root@node1 yaml]# kubectl get pv | grep pv-nfs
pv-nfs 2Gi RWO Retain Available nfs 23s
[root@node1 yaml]#
此时查看pv的状态为Available

##############################
2：创建pvc,如下：
---
apiVersion: v1
kind: PersistentVolumeClaim
metadata:
name: pvc-nfs
spec:
accessModes:
- ReadWriteOnce
resources:
requests:
storage: 2Gi
storageClassName: nfs
volumeMode: Filesystem

[root@node1 yaml]# kubectl apply -f pvc.yaml
persistentvolumeclaim/pvc-nfs created
[root@node1 yaml]# kubectl get pvc ###正常是状态先为pending，然后是Bound
NAME STATUS VOLUME CAPACITY ACCESS MODES STORAGECLASS AGE
pvc-nfs Bound pv-nfs 2Gi RWO nfs 7s

############################
3：删除pv,查看pvc的状态
[root@node1 yaml]# kubectl delete pv pv-nfs
persistentvolume "pv-nfs" delete ####会一直卡在这里

打开新的终端，查看pv状态已经是Terminating
[root@node1 ~]# kubectl get pv
NAME CAPACITY ACCESS MODES RECLAIM POLICY STATUS CLAIM STORAGECLASS REASON AGE
pv-nfs 2Gi RWO Retain Terminating default/pvc-nfs nfs 11m

出现删除不掉问题的主要原因是pv已经和pvc绑定，pv内部有kubernetes.io/pv-protection这个参数进行保护，编辑pv,删除这个参数，kubectl edit pv pv-nfs，如下：

apiVersion: v1
kind: PersistentVolume
metadata:
annotations:
kubectl.kubernetes.io/last-applied-configuration: |
{"apiVersion":"v1","kind":"PersistentVolume","metadata":{"annotations":{},"name":"pv-nfs"},"spec":{"accessModes":["ReadWriteOnce"],"capacity":{"storage":"2Gi"},"nfs":{"path":"/data/","server":"192.168.5.79"},"persistentVolumeReclaimPolicy":"Retain","storageClassName":"nfs","volumeMode":"Filesystem"}}
pv.kubernetes.io/bound-by-controller: "yes"
creationTimestamp: "2023-06-05T08:12:28Z"
deletionGracePeriodSeconds: 0
deletionTimestamp: "2023-06-05T08:23:27Z"
# finalizers: ####将此参数删除
# - kubernetes.io/pv-protection
name: pv-nfs
resourceVersion: "9420641"
uid: cb3b7c54-b8e9-4132-b273-16e9415ae643
spec:
accessModes:
- ReadWriteOnce
capacity:
storage: 2Gi

再次查看pv已经被删除了，此时pvc状态为Lost，如下：
[root@node1 yaml]# kubectl get pvc
NAME STATUS VOLUME CAPACITY ACCESS MODES STORAGECLASS AGE
pvc-nfs Lost pv-nfs 0 nfs 13m

解决方式，将删除的pv进行重新创建，pvc和pv的状态都会变成Bound
[root@node1 yaml]# kubectl apply -f nfs-pv.yaml
persistentvolume/pv-nfs created
[root@node1 yaml]# kubectl get pv
NAME CAPACITY ACCESS MODES RECLAIM POLICY STATUS CLAIM STORAGECLASS REASON AGE
pv-nfs 2Gi RWO Retain Bound default/pvc-nfs nfs 4s
[root@node1 yaml]# kubectl get pvc
NAME STATUS VOLUME CAPACITY ACCESS MODES STORAGECLASS AGE
pvc-nfs Bound pv-nfs 2Gi RWO nfs 15m

############################
4：删除pvc，查看pv状态变化，如下：
[root@node1 yaml]# kubectl delete pvc pvc-nfs
persistentvolumeclaim "pvc-nfs" deleted
[root@node1 yaml]# kubectl get pv
NAME CAPACITY ACCESS MODES RECLAIM POLICY STATUS CLAIM STORAGECLASS REASON AGE
pv-nfs 2Gi RWO Retain Released default/pvc-nfs nfs 84s

此时的pv状态已经是released，但如果此时在创建新的pvc，能够进行pv和pvc的绑定呢，大答案是不行的，因为pvc只能和Available状态的pv进行绑定。
在pv变成released的时候，查看claimRef属性，其中依然保留着PVC的绑定信息，如下：
[root@node1 yaml]# kubectl get pv pv-nfs -o yaml
apiVersion: v1
kind: PersistentVolume
metadata:
annotations:
kubectl.kubernetes.io/last-applied-configuration: |
{"apiVersion":"v1","kind":"PersistentVolume","metadata":{"annotations":{},"name":"pv-nfs"},"spec":{"accessModes":["ReadWriteOnce"],"capacity":{"storage":"2Gi"},"nfs":{"path":"/data/","server":"192.168.5.79"},"persistentVolumeReclaimPolicy":"Retain","storageClassName":"nfs","volumeMode":"Filesystem"}}
pv.kubernetes.io/bound-by-controller: "yes"
creationTimestamp: "2023-06-05T08:29:29Z"
finalizers:
- kubernetes.io/pv-protection
name: pv-nfs
resourceVersion: "9421489"
uid: 09c62907-7fa3-491f-bb48-656003581c8f
spec:
accessModes:
- ReadWriteOnce
capacity:
storage: 2Gi
claimRef: ####查看此处，保留了pvc的绑定信息
apiVersion: v1
kind: PersistentVolumeClaim
name: pvc-nfs
namespace: default
resourceVersion: "9421099"
uid: 15935386-4b7b-4b97-88b3-2bece1ed6ea9
nfs:
path: /data/
server: 192.168.5.79
persistentVolumeReclaimPolicy: Retain
storageClassName: nfs
volumeMode: Filesystem
status:
phase: Released
[root@node1 yaml]#

解决方式是导出pv的信息进行备份，然后删除claimRef出的内容，如下：
[root@node1 yaml]# kubectl get pv pv-nfs -o yaml > pv.yaml
[root@node1 yaml]# kubectl edit pv pv-nfs
---
apiVersion: v1
kind: PersistentVolume
metadata:
annotations:
kubectl.kubernetes.io/last-applied-configuration: |
{"apiVersion":"v1","kind":"PersistentVolume","metadata":{"annotations":{},"name":"pv-nfs"},"spec":{"accessModes":["ReadWriteOnce"],"capacity":{"storage":"2Gi"},"nfs":{"path":"/data/","server":"192.168.5.79"},"persistentVolumeReclaimPolicy":"Retain","storageClassName":"nfs","volumeMode":"Filesystem"}}
pv.kubernetes.io/bound-by-controller: "yes"
creationTimestamp: "2023-06-05T08:29:29Z"
finalizers:
- kubernetes.io/pv-protection
name: pv-nfs
resourceVersion: "9421489"
uid: 09c62907-7fa3-491f-bb48-656003581c8f
spec:
accessModes:
- ReadWriteOnce
capacity:
storage: 2Gi
nfs:
path: /data/
server: 192.168.5.79
persistentVolumeReclaimPolicy: Retain
storageClassName: nfs
volumeMode: Filesystem
status:
phase: Released
persistentvolume/pv-nfs edited
[root@node1 yaml]# kubectl get pv
NAME CAPACITY ACCESS MODES RECLAIM POLICY STATUS CLAIM STORAGECLASS REASON AGE
pv-nfs 2Gi RWO Retain Available nfs 10m
再次查看pv状态变为Available，此在创建pvc，pv/pvc状态都为Bound,如下：
[root@node1 yaml]# kubectl apply -f pvc.yaml
persistentvolumeclaim/pvc-nfs created
[root@node1 yaml]# kubectl get pv
NAME CAPACITY ACCESS MODES RECLAIM POLICY STATUS CLAIM STORAGECLASS REASON AGE
pv-nfs 2Gi RWO Retain Bound default/pvc-nfs nfs 12m
[root@node1 yaml]# kubectl get pvc
NAME STATUS VOLUME CAPACITY ACCESS MODES STORAGECLASS AGE
pvc-nfs Bound pv-nfs 2Gi RWO nfs

kubernetes k8s pvc nfs

centos7 centos k8s 8s