[Cloud Native Theme-38]: K8S - Core Concepts - Storage Abstraction - Configurable Size Directory Mount PV+PVC

(3) It is impossible to control the pod to limit the size of storage space occupied by the shared file directory. It is possible that one pod will fill up the entire shared storage space and lack the security control mechanism.

(4) Strong association between Pod and NFS

Introduction of 1.2 PV+PVC

To solve these problems, K8S adds a PV+PVC layer between NFS and od.

PV (Persistent Volume) is the meaning of persistent volumes and is an abstraction of the underlying shared storage. In general, PV is created and configured by the k8s administrator, it is related to the underlying specific shared storage technology, and it is docked with the shared storage through plug-ins.

PVC (Persistent Volume Claim) is the meaning of the Persistent Volume declaration and is a user's declaration of storage requirements. In other words, PVC is actually a kind of resource demand application that users send to the k8s system.

As shown in the figure below, users persist data through the PVC interface, while PVs bind to specific underlying file systems.

The introduction of PV+PVC has successfully solved the problems encountered in using NFS alone.

For further principles of PV+PVC, please refer to:

https://blog.csdn.net/HiWangWenBing/article/details/122820381https://blog.csdn.net/HiWangWenBing/article/details/122820381

Chapter 2 How to Create a Static Persistent Store Using PV

2.1 Preconditions (Executed on NFS server)

PV is built on top of NFS or other network file systems, so NFS network file systems need to be created first

https://blog.csdn.net/HiWangWenBing/article/details/122828322https://blog.csdn.net/HiWangWenBing/article/details/122828322

[root@k8s-master1 nfs]# mkdir volumes
[root@k8s-master1 nfs]# cd volumes
[root@k8s-master1 volumes]#  mkdir v{1,2,3,4,5}
[root@k8s-master1 volumes]#  ls
v1  v2  v3  v4  v5


[root@localhost volumes]# vi /etc/exports
/nfs/volumes/v1 *(insecure,rw,async,no_root_squash)
/nfs/volumes/v2 *(insecure,rw,async,no_root_squash)
/nfs/volumes/v3 *(insecure,rw,async,no_root_squash)
/nfs/volumes/v4 *(insecure,rw,async,no_root_squash)
/nfs/volumes/v5 *(insecure,rw,async,no_root_squash)


[root@localhost volumes]# exportfs -r

[root@localhost volumes]# showmount -e
Export list for localhost.localdomain:
/nfs/volumes/v5
/nfs/volumes/v4
/nfs/volumes/v3
/nfs/volumes/v2
/nfs/volumes/v1

$ Create an identifying test file in a shared directory for later validation
cd /nfs/volumes/
[root@k8s-master1 volumes]# touch ./v1/v1.txt
[root@k8s-master1 volumes]# touch ./v2/v2.txt
[root@k8s-master1 volumes]# touch ./v3/v3.txt
[root@k8s-master1 volumes]# touch ./v4/v5.txt
[root@k8s-master1 volumes]# touch ./v4/v4.txt
[root@k8s-master1 volumes]# touch ./v5/v5.txt

2.2 PV configuration file (executed on either NFS server or Client)

apiVersion: v1
kind: PersistentVolume
metadata:
  name: pv001
  labels:
    name: pv001
spec:
  nfs:
   path: /nfs/volumes/v1
   server: 172.24.130.172
  accessModes: ["ReadWriteMany","ReadWriteOnce"]
  capacity:
    storage: 1Gi
---
apiVersion: v1
kind: PersistentVolume
metadata:
  name: pv002
  labels:
    name: pv002
spec:
  nfs:
   path: /nfs/volumes/v2
   server: 172.24.130.172
  accessModes: ["ReadWriteMany","ReadWriteOnce"]
  capacity:
    storage: 2Gi
---
apiVersion: v1
kind: PersistentVolume
metadata:
  name: pv003
  labels:
    name: pv003
spec:
  nfs:
   path: /nfs/volumes/v3
   server: 172.24.130.172
  accessModes: ["ReadWriteMany","ReadWriteOnce"]
  capacity:
    storage: 3Gi
---
apiVersion: v1
kind: PersistentVolume
metadata:
  name: pv004
  labels:
    name: pv004
spec:
  nfs:
   path: /nfs/volumes/v4
   server: 172.24.130.172
  accessModes: ["ReadWriteMany","ReadWriteOnce"]
  capacity:
    storage: 4Gi
---
apiVersion: v1
kind: PersistentVolume
metadata:
  name: pv005
  labels:
    name: pv005
spec:
  nfs:
   path: /nfs/volumes/v5
   server: 172.24.130.172
  accessModes: ["ReadWriteMany","ReadWriteOnce"]
  capacity:
    storage: 5Gi

Key parameters:

server: 172.24.130.172: IP address of NFS server

/nfs/volumes/v1:Shared store from NFS server export

This means that there are PV s responsible for mapping and mount ing the underlying NFS

2.3 Creating a persistent space using PV

$ kubectl apply -f pv.yaml

2.4 Check current PV

[root@k8s-node1 ~]# kubectl get pv
NAME    CAPACITY   ACCESS MODES   RECLAIM POLICY   STATUS      CLAIM   STORAGECLASS   REASON   AGE
pv001   1Gi        RWO,RWX        Retain           Available                                   39s
pv002   2Gi        RWO,RWX        Retain           Available                                   39s
pv003   3Gi        RWO,RWX        Retain           Available                                   39s
pv004   4Gi        RWO,RWX        Retain           Available                                   39s
pv005   5Gi        RWO,RWX        Retain           Available                                   39s

Remarks:

Creation of PVS can be performed on either NFS Server or NFS Client, and K8S kubectl masks these differences.

Chapter 3 How to dynamically apply for persistent storage using PVC

3.1 PVC profile (usually with Pod on Client)

apiVersion: v1
kind: PersistentVolumeClaim
metadata:
  name: mypvc
  namespace: default
spec:
  accessModes: ["ReadWriteMany"]
  resources:
    requests:
      storage: 4Gi
---
apiVersion: v1
kind: Pod
metadata:
  name: pod-vol-pvc
  namespace: default
spec:
  containers:
  - name: myapp
    image: ikubernetes/myapp:v1
    volumeMounts:
    - name: html
      mountPath: /usr/share/nginx/html
  volumes:
  - name: html
    persistentVolumeClaim:
      claimName: mypvc

mountPath: /usr/share/nginx/html: Specify the mount path for the pod

storage: 4Gi: The size of disk space requested.

Remarks:

PVC does not need to know the NFS file path shared by NFS server s, which is PV's responsibility. PVC only needs to know the size of disk space it needs to apply for, and PV will return NFS shared storage space that has the closest hard disk space and meets the size requirements to PVC.

3.2 Use PVC to apply for persistence space

$ kubectl apply -f pvc.yaml

3.3 Check current PVC, PV

[root@k8s-node1 ~]# kubectl get pv
NAME    CAPACITY   ACCESS MODES   RECLAIM POLICY   STATUS      CLAIM            STORAGECLASS   REASON   AGE
pv001   1Gi        RWO,RWX        Retain           Bound       default/mypvc1                           12m
pv002   2Gi        RWO,RWX        Retain           Available                                            12m
pv003   3Gi        RWO,RWX        Retain           Available                                            12m
pv004   4Gi        RWO,RWX        Retain           Bound       default/mypvc                            12m
pv005   5Gi        RWO,RWX        Retain           Available                                            12m


[root@k8s-node1 ~]# kubectl get pvc
NAME     STATUS   VOLUME   CAPACITY   ACCESS MODES   STORAGECLASS   AGE
mypvc    Bound    pv004    4Gi        RWO,RWX                       5m7s
mypvc1   Bound    pv001    1Gi        RWO,RWX                       2m9s

Chapter 4 Use of persistent space by pod (same as NFS)

4.1 Enter inside pod

(1) Entry through dashboard

(2) Enter by command line

[root@k8s-node1 ~]# kubectl get pod
NAME                            READY   STATUS    RESTARTS   AGE
my-deploy-8686b49bbd-b8w65      1/1     Running   0          7h9m
my-deploy-8686b49bbd-f2m4c      1/1     Running   0          7h8m
my-deploy-8686b49bbd-jz5cx      1/1     Running   0          7h9m
nginx-pv-demo-db866fc95-6d9nz   1/1     Running   0          63m
nginx-pv-demo-db866fc95-qpx48   1/1     Running   0          63m
pod-vol-pvc                     1/1     Running   0          6m23s


$  kubectl exec -it pod-vol-pvc -- /bin/sh

4.2 Access to shared storage through PVC applications

The shared storage space requested by the pvc has a size limit that is limited by the size of the shared storage allocated by the PV.

cd /usr/share/nginx/html/

/usr/share/nginx/html # ls
v4.txt

4.3 Generating files in shared storage

# Persist files deep in pod
$ touch pod-v4.txt


# Check on master
[root@k8s-master1 v4]# ls
pod-v4.txt  v4.txt
[root@k8s-master1 v4]#

Generate persistence files in the / usr/share/nginx/html/directory of the pod as seen in master/nfs/volumes/v4/.

Files generated in master's/nfs/volumes/v4/can also be seen in pod's/usr/share/nginx/html/directory.

Reference resources:

pv and pvc concepts of k8s - do not understand 123 - all containers within the same pod of blog Park storage volume share storage network users and other spatial pod containers are k8s running on the basis of pause mirror. Gcr. Io/pause 3.1 da86e6ba6ca1 17 mon 74https://www.cnblogs.com/yxh168/p/11031003.html

Author Home Page ( Silicon-based Workshop for Potato Sugar): Blog of Potato Sugar (Wang Wenbing)_ Silicon-based Workshop for Potato Sugar_ CSDN Blog

Website address for this article: https://blog.csdn.net/HiWangWenBing/article/details/122832783

Topics: Kubernetes Cloud Native

Programmer Think