You might be knowing that data in the Pod exists till the life time of the Pod. If the Pod dies all your data that belongs to the Pod is also goes away along with Pod. So if you want to persist your data beyond the life cycle of the Pod then you must have some thing called as a Persistent volume in Kubernetes.
So lets study how to How to create HostPath persistent volume which is very easy to experiment. Also to gain knowledge about the fundamentals about the Persistent volume.
There are following types of Persistent volume types available to use within kubernetes by different vendors.
- FC (Fibre Channel)
- RBD (Ceph Block Device)
- Cinder (OpenStack block storage)
- Quobyte Volumes
- HostPath (Single node testing only — local storage is not supported in any way and WILL NOT WORK in a multi-node cluster)
- Portworx Volumes
- ScaleIO Volumes
As you can see for the HostPath it should be used only for the testing purpose. Also it does not support multi-node cluster. In case you want to explore more about the Persistent volumes you may follow this link.
The Basic process for Persistent volumes is as follows:
- K8s admin create the persistence volume in cluster.
- User will claim it using Persistent volume claim once they claimed it status becomes “Bound”.
- Then Pod use that volume for storing out the data which will persist across the life-cycle of Pod.
Enough for the theory Part Lets jump the Technical steps towards it:
- Create the persistent volume
In this step we are using following manifest yaml file to achieve the same.
# cat hostpath-pv.yaml apiVersion: v1 kind: PersistentVolume metadata: name: pv-hostpath labels: type: local spec: storageClassName: manual capacity: storage: 1Gi accessModes: - ReadWriteOnce hostPath: path: "/tmp/kube"
As shown in the above definition file it is for the size 1GB. Path is “/tmp/kube”. Lets create the PV as below:
# kubectl create -f hostpath-pv.yaml persistentvolume/pv-hostpath created
Recheck the PV and persistent volume claim using below command:
# kubectl get pv,pvc -o wide NAME CAPACITY ACCESS MODES RECLAIM POLICY STATUS CLAIM STORAGECLASS REASON AGE VOLUMEMODE persistentvolume/pv-hostpath 1Gi RWO Retain Available manual 6s Filesystem
As you can see the PV is created having status as Available and since we haven’t specified the reclaim policy default is applied which is “Retain” meaning that the even if the pvc (Persistent volume claim) gets deleted the PV and data wont get deleted automatically. We will test it out that also in a bit.
- Create the Persistent volume claim
In order to use the PV we need to create the Persistent volume claim or pvc to use it. Here is the manifest yaml file for the same.
# cat pvc-hostpath.yaml apiVersion: v1 kind: PersistentVolumeClaim metadata: name: pvc-hostpath spec: storageClassName: manual accessModes: - ReadWriteOnce resources: requests: storage: 100Mi
Kindly note in the above definition that the claim is only for the 100mb(>= size of PV) also the Access mode is “ReadWriteOnce” which is same as that of PV. Hence we can able to create the PVC as below:
# kubectl create -f pvc-hostpath.yaml persistentvolumeclaim/pvc-hostpath created
Check the status of pv and pvc.
# kubectl get pv,pvc -o wide NAME CAPACITY ACCESS MODES RECLAIM POLICY STATUS CLAIM STORAGECLASS REASON AGE VOLUMEMODE persistentvolume/pv-hostpath 1Gi RWO Retain Bound default/pvc-hostpath manual 20s Filesystem NAME STATUS VOLUME CAPACITY ACCESS MODES STORAGECLASS AGE VOLUMEMODE persistentvolumeclaim/pvc-hostpath Bound pv-hostpath 1Gi RWO manual 4s Filesystem
You will see that the status of the pv becomes Bound from Available which was earlier.
- Create the Pod to utilize this PV as a mount point inside it.
# cat busybox-pv-hostpath.yaml apiVersion: v1 kind: Pod metadata: name: busybox spec: volumes: - name: host-volume persistentVolumeClaim: claimName: pvc-hostpath containers: - image: busybox name: busybox command: ["/bin/sh"] args: ["-c", "sleep 600"] volumeMounts: - name: host-volume mountPath: /tmp/mydata
As describe in the Pod definition file it will create the mount point /tmp/mydata inside the Pod. Lets create the Pod using above definition file.
# kubectl create -f busybox-pv-hostpath.yaml pod/busybox created
Check the status and inspect the Pod:
# kubectl get all -o wide NAME READY STATUS RESTARTS AGE IP NODE NOMINATED NODE READINESS GATES pod/busybox 1/1 Running 0 2m4s 10.244.1.114 kworker01 <none> <none> NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE SELECTOR service/kubernetes ClusterIP 10.96.0.1 <none> 443/TCP 35d <none> # kubectl describe pod busybox Name: busybox Namespace: default Priority: 0 Node: kworker01/10.253.121.32 Start Time: Mon, 06 Jul 2020 02:43:16 -0400 Labels: <none> Annotations: <none> Status: Running IP: 10.244.1.114 IPs: IP: 10.244.1.114 Containers: busybox: Container ID: docker://6d1cfa9b6440efe2770244d1edc6a78c0dd7649bbf905121e70a013ad3b1dd1e Image: busybox Image ID: docker-pullable://busybox@sha256:9ddee63a712cea977267342e8750ecbc60d3aab25f04ceacfa795e6fce341793 Port: <none> Host Port: <none> Command: /bin/sh Args: -c sleep 600 State: Running Started: Mon, 06 Jul 2020 02:43:25 -0400 Ready: True Restart Count: 0 Environment: <none> Mounts: /tmp/mydata from host-volume (rw) /var/run/secrets/kubernetes.io/serviceaccount from default-token-49xz2 (ro) Conditions: Type Status Initialized True Ready True ContainersReady True PodScheduled True Volumes: host-volume: Type: PersistentVolumeClaim (a reference to a PersistentVolumeClaim in the same namespace) ClaimName: pvc-hostpath ReadOnly: false default-token-49xz2: Type: Secret (a volume populated by a Secret) SecretName: default-token-49xz2 Optional: false QoS Class: BestEffort Node-Selectors: <none> Tolerations: node.kubernetes.io/not-ready:NoExecute for 300s node.kubernetes.io/unreachable:NoExecute for 300s Events: Type Reason Age From Message ---- ------ ---- ---- ------- Normal Scheduled <unknown> default-scheduler Successfully assigned default/busybox to kworker01 Normal Pulling 64s kubelet, kworker01 Pulling image "busybox" Normal Pulled 58s kubelet, kworker01 Successfully pulled image "busybox" Normal Created 58s kubelet, kworker01 Created container busybox Normal Started 57s kubelet, kworker01 Started container busybox
In the describe output you can see that, /tmp/mydata volume got created using host-volume from the claim pvc-hostpath. Also the Pod is scheduled/create on the node “kworker01”.
Lets login inside the Pod to create the sample file. In order to demonstrate the life cycle of the data even if the Pod dies.
# kubectl exec -it busybox -- sh / # hostname busybox / # cd /tmp/ /tmp # ls mydata /tmp # cd mydata/ /tmp/mydata # echo "hello from K8S" > Hello.txt /tmp/mydata # ls -ltr total 4 -rw-r--r-- 1 root root 15 Jul 6 06:46 Hello.txt /tmp/mydata #
In the above demo we have created file “Hello.txt” inside /tmp/mydata. Now lets delete the Pod.
# kubectl delete pod busybox pod "busybox" deleted root@vbhost:~/kubernetes/yamls# kubectl get all NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE service/kubernetes ClusterIP 10.96.0.1 <none> 443/TCP 35d
Pod got deleted successfully lets login to the node “kworker01” where Pod got scheduled earlier to check if the data still persist after the deletion of the Pod.
sh-4.2# hostname kworker01 sh-4.2# cd /tmp sh-4.2# ls kube sh-4.2# cd kube/ sh-4.2# ls Hello.txt sh-4.2# cat Hello.txt hello from K8S sh-4.2# exit
You can see that our file “Hello.txt” still exists on the Node even the Pod dies.
So this is all about “How to create HostPath persistent volume” in Kubernetes.