PV static creation and dynamic creation

NFS（Network File System）

In the use of Kubernetes In the process of , We often use storage . Maximum function of storage , That is to make the data in the container persistent , Prevent the phenomenon of deleting databases and running away . To realize this function , Cannot leave network file system .kubernetes adopt NFS Network file system , Synchronize the attached data of each node , Then it's guaranteed pod Failover, etc , The stored data can still be read .

One 、 install NFS

stay kubernetes Installation in cluster NFS.

1. All nodes execute

yum install -y nfs-utils

2. nfs Master node execution

echo "/nfs/data/ *(insecure,rw,sync,no_root_squash)" > /etc/exports  #  Exposed the directory /nfs/data/,`*` It means that all nodes can access .
mkdir -p /nfs/data
systemctl enable rpcbind --now
systemctl enable nfs-server --now
#  Configuration takes effect 
exportfs -r

#  Check and verify 
[[email protected] ~]# exportfs
/nfs/data      <world>
[[email protected] ~]#

2. nfs Execute... From node

#  Exhibition 172.31.0.2 Which directories can be mounted 
showmount -e 172.31.0.2  # ip Change to your own master node ip

mkdir -p /nfs/data
#  Mount the local directory and remote directory 
mount -t nfs 172.31.0.2:/nfs/data /nfs/data

Two 、 verification

#  Write a test file 
echo "hello nfs server" > /nfs/data/test.txt

Through these steps , We can see NFS The file system has been successfully installed .172.31.0.2 As the main node of the system , Exposed /nfs/data, Other slave nodes /nfs/data And the master node /nfs/data Mounted . stay kubernetes Within cluster , You can choose any server as server, Other servers act as client.

PV&PVC

PV： Persistent volume （Persistent Volume）, Save the data that the application needs to persist to the specified location .

PVC： Persistent volume declaration （Persistent Volume Claim）, State the persistent volume specifications to be used .

Pod The data in needs to be persisted , Where to save it ? Of course, save to PV in ,PV Where it is , That's what we said NFS file system . So how to use PV Well ？ It is through PVC. like PV Is stored ,PVC Just tell PV, I'm going to start using you .

PV There are generally two kinds of creation , Static creation and dynamic creation . Static creation is to create a lot in advance PV, To form a PV pool , according to PVC According to the specification requirements, choose the right one to supply . Dynamic creation does not need to be created in advance , But according to PVC Specification requirements for , You can create whatever specifications are required . From the perspective of resource utilization , Dynamic creation is better .

One 、 Static creation PV

1. Create data storage directory .

# nfs Master node execution 
[[email protected] data]# pwd
/nfs/data
[[email protected] data]# mkdir -p /nfs/data/01
[[email protected] data]# mkdir -p /nfs/data/02
[[email protected] data]# mkdir -p /nfs/data/03
[[email protected] data]# ls
01  02  03
[[email protected] data]# 

2. establish PV,pv.yml

apiVersion: v1
kind: PersistentVolume
metadata:
  name: pv01-10m
spec:
  capacity:
    storage: 10M
  accessModes:
    - ReadWriteMany
  storageClassName: nfs
  nfs:
    path: /nfs/data/01
    server: 172.31.0.2
---
apiVersion: v1
kind: PersistentVolume
metadata:
  name: pv02-1gi
spec:
  capacity:
    storage: 1Gi
  accessModes:
    - ReadWriteMany
  storageClassName: nfs
  nfs:
    path: /nfs/data/02
    server: 172.31.0.2
---
apiVersion: v1
kind: PersistentVolume
metadata:
  name: pv03-3gi
spec:
  capacity:
    storage: 3Gi
  accessModes:
    - ReadWriteMany
  storageClassName: nfs
  nfs:
    path: /nfs/data/03
    server: 172.31.0.2

[[email protected] ~]# kubectl apply -f pv.yml 
persistentvolume/pv01-10m created
persistentvolume/pv02-1gi created
persistentvolume/pv03-3gi created
[[email protected] ~]# kubectl get persistentvolume
NAME       CAPACITY   ACCESS MODES   RECLAIM POLICY   STATUS      CLAIM   STORAGECLASS   REASON   AGE
pv01-10m   10M        RWX            Retain           Available           nfs                     45s
pv02-1gi   1Gi        RWX            Retain           Available           nfs                     45s
pv03-3gi   3Gi        RWX            Retain           Available           nfs                     45s
[[email protected] ~]# 

Three folders correspond to three PV, The sizes are 10M、1Gi、3Gi, These three PV To form a PV pool .

3. establish PVC,pvc.yml

kind: PersistentVolumeClaim
apiVersion: v1
metadata:
  name: nginx-pvc
spec:
  accessModes:
    - ReadWriteMany
  resources:
    requests:
      storage: 200Mi
  storageClassName: nfs

[[email protected] ~]# kubectl apply -f pvc.yml 
persistentvolumeclaim/nginx-pvc created
[[email protected] ~]# kubectl get pvc,pv
NAME                              STATUS   VOLUME     CAPACITY   ACCESS MODES   STORAGECLASS   AGE
persistentvolumeclaim/nginx-pvc   Bound    pv02-1gi   1Gi        RWX            nfs            14m

NAME                        CAPACITY   ACCESS MODES   RECLAIM POLICY   STATUS      CLAIM               STORAGECLASS   REASON   AGE
persistentvolume/pv01-10m   10M        RWX            Retain           Available                       nfs                     17m
persistentvolume/pv02-1gi   1Gi        RWX            Retain           Bound       default/nginx-pvc   nfs                     17m
persistentvolume/pv03-3gi   3Gi        RWX            Retain           Available                       nfs                     17m
[[email protected] ~]# 

Can be found using 200Mi Of PVC, Will be in PV The best binding in the pool 1Gi The size of pv02-1gi Of PV To use the , Status as Bound. Create Pod or Deployment When , Use PVC You can persist the data to PV in , and NFS Any node of can synchronize .

Two 、 Dynamically create PV

1. Configure the default storage class of dynamic provisioning ,sc.yml

#  Created a storage class 
apiVersion: storage.k8s.io/v1
kind: StorageClass
metadata:
  name: nfs-storage
  annotations:
    storageclass.kubernetes.io/is-default-class: "true"
provisioner: k8s-sigs.io/nfs-subdir-external-provisioner
parameters:
  archiveOnDelete: "true"  ##  Delete pv When ,pv Do you want to back up your content 

---
apiVersion: apps/v1
kind: Deployment
metadata:
  name: nfs-client-provisioner
  labels:
    app: nfs-client-provisioner
  # replace with namespace where provisioner is deployed
  namespace: default
spec:
  replicas: 1
  strategy:
    type: Recreate
  selector:
    matchLabels:
      app: nfs-client-provisioner
  template:
    metadata:
      labels:
        app: nfs-client-provisioner
    spec:
      serviceAccountName: nfs-client-provisioner
      containers:
        - name: nfs-client-provisioner
          image: registry.cn-hangzhou.aliyuncs.com/lfy_k8s_images/nfs-subdir-external-provisioner:v4.0.2
          # resources:
          #    limits:
          #      cpu: 10m
          #    requests:
          #      cpu: 10m
          volumeMounts:
            - name: nfs-client-root
              mountPath: /persistentvolumes
          env:
            - name: PROVISIONER_NAME
              value: k8s-sigs.io/nfs-subdir-external-provisioner
            - name: NFS_SERVER
              value: 172.31.0.2 ##  Specify yourself nfs Server address 
            - name: NFS_PATH  
              value: /nfs/data  ## nfs Server shared directory 
      volumes:
        - name: nfs-client-root
          nfs:
            server: 172.31.0.2
            path: /nfs/data
---
apiVersion: v1
kind: ServiceAccount
metadata:
  name: nfs-client-provisioner
  # replace with namespace where provisioner is deployed
  namespace: default
---
kind: ClusterRole
apiVersion: rbac.authorization.k8s.io/v1
metadata:
  name: nfs-client-provisioner-runner
rules:
  - apiGroups: [""]
    resources: ["nodes"]
    verbs: ["get", "list", "watch"]
  - apiGroups: [""]
    resources: ["persistentvolumes"]
    verbs: ["get", "list", "watch", "create", "delete"]
  - apiGroups: [""]
    resources: ["persistentvolumeclaims"]
    verbs: ["get", "list", "watch", "update"]
  - apiGroups: ["storage.k8s.io"]
    resources: ["storageclasses"]
    verbs: ["get", "list", "watch"]
  - apiGroups: [""]
    resources: ["events"]
    verbs: ["create", "update", "patch"]
---
kind: ClusterRoleBinding
apiVersion: rbac.authorization.k8s.io/v1
metadata:
  name: run-nfs-client-provisioner
subjects:
  - kind: ServiceAccount
    name: nfs-client-provisioner
    # replace with namespace where provisioner is deployed
    namespace: default
roleRef:
  kind: ClusterRole
  name: nfs-client-provisioner-runner
  apiGroup: rbac.authorization.k8s.io
---
kind: Role
apiVersion: rbac.authorization.k8s.io/v1
metadata:
  name: leader-locking-nfs-client-provisioner
  # replace with namespace where provisioner is deployed
  namespace: default
rules:
  - apiGroups: [""]
    resources: ["endpoints"]
    verbs: ["get", "list", "watch", "create", "update", "patch"]
---
kind: RoleBinding
apiVersion: rbac.authorization.k8s.io/v1
metadata:
  name: leader-locking-nfs-client-provisioner
  # replace with namespace where provisioner is deployed
  namespace: default
subjects:
  - kind: ServiceAccount
    name: nfs-client-provisioner
    # replace with namespace where provisioner is deployed
    namespace: default
roleRef:
  kind: Role
  name: leader-locking-nfs-client-provisioner
  apiGroup: rbac.authorization.k8s.io

Note here ip Change it to your own , And the address of the image warehouse has been changed to Alibaba cloud , Prevent the image from downloading .

[[email protected] ~]# kubectl apply -f sc.yml
storageclass.storage.k8s.io/nfs-storage created
deployment.apps/nfs-client-provisioner created
serviceaccount/nfs-client-provisioner created
clusterrole.rbac.authorization.k8s.io/nfs-client-provisioner-runner created
clusterrolebinding.rbac.authorization.k8s.io/run-nfs-client-provisioner created
role.rbac.authorization.k8s.io/leader-locking-nfs-client-provisioner created
rolebinding.rbac.authorization.k8s.io/leader-locking-nfs-client-provisioner created
[[email protected] ~]# 

2. Confirm that the configuration is in effect

[[email protected] ~]# kubectl get sc
NAME                    PROVISIONER                                   RECLAIMPOLICY   VOLUMEBINDINGMODE   ALLOWVOLUMEEXPANSION   AGE
nfs-storage (default)   k8s-sigs.io/nfs-subdir-external-provisioner   Delete          Immediate           false                  6s
[[email protected] ~]#

3. Dynamic supply testing ,pvc.yml

kind: PersistentVolumeClaim
apiVersion: v1
metadata:
  name: nginx-pvc
spec:
  accessModes:
    - ReadWriteMany
  resources:
    requests:
      storage: 200Mi

[[email protected] ~]# kubectl apply -f pvc.yml 
persistentvolumeclaim/nginx-pvc created
[[email protected] ~]# kubectl get pvc
NAME        STATUS   VOLUME                                     CAPACITY   ACCESS MODES   STORAGECLASS   AGE
nginx-pvc   Bound    pvc-7b01bc33-826d-41d0-a990-8c1e7c997e6f   200Mi      RWX            nfs-storage    9s
[[email protected] ~]# kubectl get pv
NAME                                       CAPACITY   ACCESS MODES   RECLAIM POLICY   STATUS   CLAIM               STORAGECLASS   REASON   AGE
pvc-7b01bc33-826d-41d0-a990-8c1e7c997e6f   200Mi      RWX            Delete           Bound    default/nginx-pvc   nfs-storage             16s
[[email protected] ~]# 

pvc The statement needs 200Mi Space , Then it is created 200Mi Of pv, Status as Bound, Test success . Create Pod or Deployment When , Use PVC You can persist the data to PV in , and NFS Any node of can synchronize .

Summary

Then you can understand ,NFS、PV and PVC by kubernetes Clusters provide data storage support , The application is deployed by any node , Then the previous data can still be read .

This paper is written by mdnice Multi platform Publishing