Using Stash with Rook Managed Ceph Storage

This guide will show you how to use Stash to backup and restore volumes of a Kubernetes workload in Rook managed Ceph storage. Here, we are going to backup a volume of a Deployment into AWS S3 compatible Ceph Object Storage. Then, we are going to show how to restore this volume into a PersistentVolumeClaim of Ceph Object Storage. We are going to also re-deploy deployment using this recovered volume.

Before You Begin

At first, you need to have a Kubernetes cluster. If you don’t already have a cluster, you can create one by using kind.
Install Stash in your cluster following the steps here.
You should be familiar with the following Stash concepts:
You will need a Rook deployment with Ceph Object Storage and Ceph Block Storage configured. If you do not already have a Rook Storage Service configured, you can create one by following this quickstart guide.

To keep everything isolated, we are going to use a separate namespace called demo throughout this tutorial.

$ kubectl create ns demo
namespace/demo created

Note: YAML files used in this tutorial are stored in docs/guides/platforms/rook/examples directory of stashed/docs repository.

Choosing StorageClass:

Ceph Block Storage allows mounting Rook storage into pod using a PersistentVolumeClaim. In order to do that, we have to create a PersistentVolumeClaim with rook-ceph-blockStorageClass. Verify the StorageClass exist by the following command:

$ kubectl get storageclass
NAME                 PROVISIONER                AGE
rook-ceph-block      ceph.rook.io/block         89m
standard (default)   k8s.io/minikube-hostpath   104m

Backup the Volume of a Deployment

Here, we are going to deploy a Deployment with a PVC. This Deployment will automatically generate some sample data into the PVC. Then, we are going to backup this sample data using Stash.

Prepare Workload

At first, let’s deploy the workload whose volumes we are going to backup. Here, we are going create a PVC and deploy a Deployment with this PVC.

Create PVC:

Below is the YAML of the sample PVC that we are going to create,

kind: PersistentVolumeClaim
apiVersion: v1
metadata:
  name: source-pvc
  namespace: demo
spec:
  accessModes:
  - ReadWriteOnce
  storageClassName: "rook-ceph-block"
  resources:
    requests:
      storage: 1Gi

Let’s create the PVC we have shown above,

$ kubectl apply -f https://github.com/stashed/docs/raw/v2024.8.27/docs/guides/platforms/rook/examples/pvc.yaml
persistentvolumeclaim/source-pvc created

Deploy Deployment:

Now, we are going to deploy a Deployment that uses the above PVC. This Deployment will automatically generate sample data (data.txt file) in /source/data directory where we have mounted the PVC.

Below is the YAML of the Deployment that we are going to create,

apiVersion: apps/v1
kind: Deployment
metadata:
  labels:
    app: stash-demo
  name: stash-demo
  namespace: demo
spec:
  replicas: 1
  selector:
    matchLabels:
      app: stash-demo
  template:
    metadata:
      labels:
        app: stash-demo
      name: busybox
    spec:
      containers:
      - args: ["echo sample_data > /source/data/data.txt && sleep 3000"]
        command: ["/bin/sh", "-c"]
        image: busybox
        imagePullPolicy: IfNotPresent
        name: busybox
        volumeMounts:
        - mountPath: /source/data
          name: source-data
      restartPolicy: Always
      volumes:
      - name: source-data
        persistentVolumeClaim:
          claimName: source-pvc
  strategy:
    rollingUpdate:
      maxSurge: "0%"
      maxUnavailable: "100%"

Let’s create the Deployment we have shown above.

$ kubectl apply -f https://github.com/stashed/docs/raw/v2024.8.27/docs/guides/platforms/rook/examples/deployment.yaml
deployment.apps/stash-demo created

Now, wait for the pods of the Deployment to go into the Running state.

$ kubectl get pod -n demo
NAME                          READY   STATUS    RESTARTS   AGE
stash-demo-69f9ffbbf7-98lth   1/1     Running   0          13s

To verify that the sample data has been created in /source/data directory, use the following command:

$ kubectl exec -n demo stash-demo-69f9ffbbf7-98lth -- cat /source/data/data.txt
sample_data

Prepare Backend

We are going to store our backed up data into an Ceph Storage Bucket. At first, we need to create a secret with the access credentials to our Ceph storage bucket. Then, we have to create a Repository crd that will hold the information about our backend storage. If you want to use a different backend, please read the respective backend configuration doc from here.

Create Secret:

Let’s create a secret called rook-secret with access credentials to our desired Ceph Storage Bucket,

$ echo -n 'changeit' > RESTIC_PASSWORD
$ echo -n '<your-rook-access-key-here>' > AWS_ACCESS_KEY_ID
$ echo -n '<your-rook-secret-key-here>' > AWS_SECRET_ACCESS_KEY
$ kubectl create secret generic -n demo rook-secret \
    --from-file=./RESTIC_PASSWORD \
    --from-file=./AWS_ACCESS_KEY_ID \
    --from-file=./AWS_SECRET_ACCESS_KEY
secret/rook-secret created

Verify that the secret has been created successfully,

$ kubectl get secret -n demo rook-secret -o yaml

apiVersion: v1
data:
  AWS_ACCESS_KEY_ID: Tk5HTkcwUVowS1lVOEhKMEFQVEQ=
  AWS_SECRET_ACCESS_KEY: alExVHJFNU9oa2QzUEk0QzlQbkYwTjVaV0hvd2Yycm9BS2U1MEVsdA==
  RESTIC_PASSWORD: Y2hhbmdlaXQ=
kind: Secret
metadata:
  creationTimestamp: "2019-07-25T10:34:25Z"
  name: rook-secret
  namespace: demo
  resourceVersion: "43037"
  selfLink: /api/v1/namespaces/demo/secrets/rook-secret
  uid: 49f412a4-f1e7-4eb1-92f5-49c0a96cc18d
type: Opaque

Create Repository:

Now, let’s create a Repository with the information of our desired Ceph storage bucket. Below is the YAML of Repository crd we are going to create,

apiVersion: stash.appscode.com/v1alpha1
kind: Repository
metadata:
  name: rook-repo
  namespace: demo
spec:
  backend:
    s3:
      endpoint: 'http://rook-ceph-rgw-my-store-external.rook-ceph.svc'
      bucket: rook-bucket
      prefix: /source/data
    storageSecretName: rook-secret

Let’s create the Repository we have shown above,

$ kubectl apply -f https://github.com/stashed/docs/raw/v2024.8.27/docs/guides/platforms/rook/examples/repository.yaml
repository.stash.appscode.com/rook-repo created

Now, we are ready to backup our sample data into this backend.

Backup

We have to create a BackupConfiguration crd targeting the stash-demo Deployment that we have deployed earlier. Stash will inject a sidecar container into the target. It will also create a CronJob to take a periodic backup of /source/data directory of the target.

Create BackupConfiguration:

Below is the YAML of the BackupConfiguration crd that we are going to create,

apiVersion: stash.appscode.com/v1beta1
kind: BackupConfiguration
metadata:
  name: deployment-backup
  namespace: demo
spec:
  repository:
    name: rook-repo
  schedule: "*/5 * * * *"
  target:
    ref:
      apiVersion: apps/v1
      kind: Deployment
      name: stash-demo
    volumeMounts:
    - name: source-data
      mountPath: /source/data
    paths:
    - /source/data
  retentionPolicy:
    name: 'keep-last-5'
    keepLast: 5
    prune: true

Here,

spec.repository refers to the Repository object rook-repo that holds backend information.
spec.target.ref refers to the stash-demo Deployment for backup target.
spec.target.volumeMounts specifies a list of volumes and their mountPath that contain the target paths.
spec.target.paths specifies list of file paths to backup.

Let’s create the BackupConfiguration crd we have shown above,

$ kubectl apply -f https://github.com/stashed/docs/raw/v2024.8.27/docs/guides/platforms/rook/examples/backupconfiguration.yaml
backupconfiguration.stash.appscode.com/deployment-backup created

Verify Sidecar:

If everything goes well, Stash will inject a sidecar container into the stash-demo Deployment to take backup of /source/data directory. Let’s check that the sidecar has been injected successfully,

$ kubectl get pod -n demo
NAME                          READY   STATUS    RESTARTS   AGE
stash-demo-76d78d8966-nbkrl   2/2     Running   0          39s

Look at the pod. It now has 2 containers. If you view the resource definition of this pod, you will see that there is a container named stash which is running run-backup command.

$ kubectl get pod -n demo stash-demo-76d78d8966-nbkrl -o yaml
apiVersion: v1
kind: Pod
metadata:
  generateName: stash-demo-76d78d8966-
  labels:
    app: stash-demo
    pod-template-hash: 76d78d8966
  name: stash-demo-76d78d8966-nbkrl
  namespace: demo
  ...
spec:
  containers:
  - args:
    - echo sample_data > /source/data/data.txt && sleep 3000
    command:
    - /bin/sh
    - -c
    image: busybox
    imagePullPolicy: IfNotPresent
    name: busybox
    resources: {}
    terminationMessagePath: /dev/termination-log
    terminationMessagePolicy: File
    volumeMounts:
    - mountPath: /source/data
      name: source-data
    - mountPath: /var/run/secrets/kubernetes.io/serviceaccount
      name: default-token-2njxz
      readOnly: true
  - args:
    - run-backup
    - --backup-configuration=deployment-backup
    - --secret-dir=/etc/stash/repository/secret
    - --enable-cache=true
    - --max-connections=0
    - --metrics-enabled=true
    - --pushgateway-url=http://stash-operator.kube-system.svc:56789
    - --enable-status-subresource=true
    - --use-kubeapiserver-fqdn-for-aks=true
    - --logtostderr=true
    - --alsologtostderr=false
    - --v=3
    - --stderrthreshold=0
    env:
    - name: NODE_NAME
      valueFrom:
        fieldRef:
          apiVersion: v1
          fieldPath: spec.nodeName
    - name: POD_NAME
      valueFrom:
        fieldRef:
          apiVersion: v1
          fieldPath: metadata.name
    image: suaas21/stash:volumeTemp_linux_amd64
    imagePullPolicy: IfNotPresent
    name: stash
    resources: {}
    terminationMessagePath: /dev/termination-log
    terminationMessagePolicy: File
    volumeMounts:
    - mountPath: /etc/stash
      name: stash-podinfo
    - mountPath: /etc/stash/repository/secret
      name: stash-secret-volume
    - mountPath: /tmp
      name: tmp-dir
    - mountPath: /source/data
      name: source-data
    - mountPath: /var/run/secrets/kubernetes.io/serviceaccount
      name: default-token-2njxz
      readOnly: true
  dnsPolicy: ClusterFirst
  enableServiceLinks: true
  nodeName: minikube
  priority: 0
  restartPolicy: Always
  schedulerName: default-scheduler
  securityContext: {}
  serviceAccount: default
  serviceAccountName: default
  terminationGracePeriodSeconds: 30
  tolerations:
  - effect: NoExecute
    key: node.kubernetes.io/not-ready
    operator: Exists
    tolerationSeconds: 300
  - effect: NoExecute
    key: node.kubernetes.io/unreachable
    operator: Exists
    tolerationSeconds: 300
  volumes:
  - name: source-data
    persistentVolumeClaim:
      claimName: source-pvc
  - emptyDir: {}
    name: tmp-dir
  - downwardAPI:
      defaultMode: 420
      items:
      - fieldRef:
          apiVersion: v1
          fieldPath: metadata.labels
        path: labels
    name: stash-podinfo
  - name: stash-secret-volume
    secret:
      defaultMode: 420
      secretName: rook-secret
  - name: default-token-2njxz
    secret:
      defaultMode: 420
      secretName: default-token-2njxz
  ...
...

Verify CronJob:

It will also create a CronJob with the schedule specified in spec.schedule field of BackupConfiguration crd.

Verify that the CronJob has been created using the following command,

$ kubectl get cronjob -n demo
NAME                SCHEDULE      SUSPEND   ACTIVE   LAST SCHEDULE   AGE
deployment-backup   */1 * * * *   False     0        13s             1m50s

Wait for BackupSession:

The deployment-backup CronJob will trigger a backup on each schedule by creating a BackupSession crd. The sidecar container will watch for the BackupSession crd. When it finds one, it will take backup immediately.

Wait for the next schedule for backup. Run the following command to watch BackupSession crd,

$ watch -n 3 kubectl get backupsession -n demo
Every 3.0s: kubectl get backupsession -n demo        suaas-appscode: Mon Jul 22 15:01:21 2019

NAME                           INVOKER-TYPE          INVOKER-NAME        PHASE       AGE
deployment-backup-1663786062   BackupConfiguration   deployment-backup   Succeeded   21s

We can see from the above output that the backup session has succeeded. Now, we are going to verify whether the backed up data has been stored in the backend.

Verify Backup:

Once a backup is complete, Stash will update the respective Repository crd to reflect the backup. Check that the repository rook-repo has been updated by the following command,

$ kubectl get repository -n demo
NAME         INTEGRITY   SIZE   SNAPSHOT-COUNT   LAST-SUCCESSFUL-BACKUP   AGE
rook-repo    true        30 B   2                3m10s                    5m20s

Here, BACKUPCOUNT field indicates the number of backup snapshots has taken in this repository.

Restore the Backed up Data

This section will show you how to restore the backed up data from Ceph Storage Bucket we have taken in the earlier section.

Stop Taking Backup of the Old Deployment:

At first, let’s stop taking any further backup of the old Deployment so that no backup is taken during the restore process. We are going to pause the BackupConfiguration that we created to backup the stash-demo Deployment. Then, Stash will stop taking any further backup for this Deployment. You can learn more how to pause a scheduled backup here

Let’s pause the deployment-backup BackupConfiguration,

$ kubectl patch backupconfiguration -n demo deployment-backup --type="merge" --patch='{"spec": {"paused": true}}'
backupconfiguration.stash.appscode.com/deployment-backup patched

Now, wait for a moment. Stash will pause the BackupConfiguration. Verify that the BackupConfiguration has been paused,

$ kubectl get backupconfiguration -n demo
NAME                TASK   SCHEDULE      PAUSED   AGE
deployment-backup          */1 * * * *   true     26m

Notice the PAUSED column. Value true for this field means that the BackupConfiguration has been paused.

Deploy Deployment:

We are going to create a new Deployment named stash-recovered with a new PVC and restore the backed up data inside it.

Below are the YAMLs of the Deployment and PVC that we are going to create,

apiVersion: v1
kind: PersistentVolumeClaim
metadata:
  name: restore-pvc
  namespace: demo
spec:
  accessModes:
  - ReadWriteOnce
  storageClassName: "rook-ceph-block"
  resources:
    requests:
      storage: 1Gi
---
apiVersion: apps/v1
kind: Deployment
metadata:
  labels:
    app: stash-recovered
  name: stash-recovered
  namespace: demo
spec:
  replicas: 1
  selector:
    matchLabels:
      app: stash-recovered
  template:
    metadata:
      labels:
        app: stash-recovered
      name: busybox
    spec:
      containers:
      - args:
        - sleep
        - "3600"
        image: busybox
        imagePullPolicy: IfNotPresent
        name: busybox
        volumeMounts:
        - mountPath: /restore/data
          name: restore-data
      restartPolicy: Always
      volumes:
      - name: restore-data
        persistentVolumeClaim:
          claimName: restore-pvc
  strategy:
    rollingUpdate:
      maxSurge: "0%"
      maxUnavailable: "100%"

Let’s create the Deployment and PVC we have shown above.

$ kubectl apply -f https://github.com/stashed/docs/raw/v2024.8.27/docs/guides/platforms/rook/examples/recovered_deployment.yaml
persistentvolumeclaim/restore-pvc created
deployment.apps/stash-recovered created

Create RestoreSession:

Now, we need to create a RestoreSession crd targeting the stash-recovered Deployment.

Below is the YAML of the RestoreSesion crd that we are going to create,

apiVersion: stash.appscode.com/v1beta1
kind: RestoreSession
metadata:
  name: deployment-restore
  namespace: demo
spec:
  repository:
    name: rook-repo
  target: # target indicates where the recovered data will be stored
    ref:
      apiVersion: apps/v1
      kind: Deployment
      name: stash-recovered
    volumeMounts:
    - name: restore-data
      mountPath: /source/data
    rules:
    - paths:
      - /source/data/

Here,

spec.repository.name specifies the Repository crd that holds the backend information where our backed up data has been stored.
spec.target.ref refers to the target workload where the recovered data will be stored.
spec.target.volumeMounts specifies a list of volumes and their mountPath where the data will be restored.
- mountPath must be same mountPath as the original volume because Stash stores absolute path of the backed up files. If you use different mountPath for the restored volume the backed up files will not be restored into your desired volume.

Let’s create the RestoreSession crd we have shown above,

$ kubectl apply -f https://github.com/stashed/docs/raw/v2024.8.27/docs/guides/platforms/rook/examples/restoresession.yaml
restoresession.stash.appscode.com/deployment-restore created

Once, you have created the RestoreSession crd, Stash will inject init-container into stash-recovered Deployment. The Deployment will restart and the init-container will restore the desired data on start-up.

Verify Init-Container:

Wait until the init-container has been injected into the stash-recovered Deployment. Let’s describe the Deployment to verify that init-container has been injected successfully.

$ kubectl describe deployment -n demo stash-recovered
Name:                   stash-recovered
Namespace:              demo
CreationTimestamp:      Thu, 25 Jul 2019 19:14:18 +0600
Labels:                 app=stash-recovered
Selector:               app=stash-recovered
Replicas:               1 desired | 1 updated | 1 total | 1 available | 0 unavailable
StrategyType:           RollingUpdate
MinReadySeconds:        0
RollingUpdateStrategy:  100% max unavailable, 0% max surge
...
Pod Template:
  Labels:       app=stash-recovered
  Annotations:  stash.appscode.com/last-applied-restoresession-hash: 10604314699482840900
  Init Containers:
   stash-init:
    Image:      suaas21/stash:volumeTemp_linux_amd64
    Port:       <none>
    Host Port:  <none>
    Args:
      restore
      --restore-session=deployment-restore
      --secret-dir=/etc/stash/repository/secret
      --enable-cache=true
      --max-connections=0
      --metrics-enabled=true
      --pushgateway-url=http://stash-operator.kube-system.svc:56789
      --enable-status-subresource=true
      --use-kubeapiserver-fqdn-for-aks=true
      --logtostderr=true
      --alsologtostderr=false
      --v=3
      --stderrthreshold=0
    Environment:
      NODE_NAME:   (v1:spec.nodeName)
      POD_NAME:    (v1:metadata.name)
    Mounts:
      /etc/stash/repository/secret from stash-secret-volume (rw)
      /source/data from restore-data (rw)
      /tmp from tmp-dir (rw)
  Containers:
   busybox:
    Image:      busybox
    Port:       <none>
    Host Port:  <none>
    Args:
      sleep
      3600
    Environment:  <none>
    Mounts:
      /restore/data from restore-data (rw)
  Volumes:
   restore-data:
    Type:       PersistentVolumeClaim (a reference to a PersistentVolumeClaim in the same namespace)
    ClaimName:  restore-pvc
    ReadOnly:   false
   tmp-dir:
    Type:       EmptyDir (a temporary directory that shares a pod's lifetime)
    Medium:
    SizeLimit:  <unset>
   stash-podinfo:
    Type:  DownwardAPI (a volume populated by information about the pod)
    Items:
      metadata.labels -> labels
   stash-secret-volume:
    Type:        Secret (a volume populated by a Secret)
    SecretName:  rook-secret
    Optional:    false
...

Notice the Init-Containers section. We can see that the init-container stash-init has been injected which is running restore command.

Wait for RestoreSession to Succeeded:

Now, wait for the restore process to complete. You can watch the RestoreSession phase using the following command,

$ watch -n 2 kubectl get restoresession -n demo
Every 3.0s: kubectl get restoresession --all-namespaces                 suaas-appscode: Mon Jul 28 18:17:22 2019

NAMESPACE   NAME                 REPOSITORY-NAME   PHASE       AGE
demo        deployment-restore   rook-repo         Succeeded   100s

So, we can see from the output of the above command that the restore process has succeeded.

Note: If you want to restore the backed up data inside the same Deployment whose volumes were backed up, you have to remove the corrupted data from the Deployment. Then, you have to create a RestoreSession targeting the Deployment.

Verify Restored Data:

In this section, we are going to verify that the desired data has been restored successfully. At first, check if the stash-recovered pod of the Deployment has gone into Running state by the following command,

$ kubectl get pod -n demo
NAME                               READY   STATUS    RESTARTS   AGE
stash-recovered-5c59587895-76tsx   1/1     Running   0          73s

Verify that the sample data has been restored in /restore/data directory of the stash-recovered pod of the Deployment using the following command,

$ kubectl exec -n demo stash-recovered-5c59587895-76tsx  -- cat /restore/data/data.txt
sample_data

Cleaning Up

To clean up the Kubernetes resources created by this tutorial, run:

kubectl delete -n demo deployment stash-demo
kubectl delete -n demo deployment stash-recovered
kubectl delete -n demo backupconfiguration deployment-backup
kubectl delete -n demo restoresession deployment-restore
kubectl delete -n demo repository rook-repo
kubectl delete -n demo secret rook-secret
kubectl delete -n demo pvc --all

Next Steps

See a step by step guide to backup/restore volumes of a StatefulSet here.
See a step by step guide to backup/restore volumes of a DaemonSet here.
See a step by step guide to Backup/restore Stand-alone PVC here

KubeDB

Stash

KubeVault

Voyager

ConfigSyncer

Guard

RESOURCES

Elasticsearch

MariaDB

MongoDB

MySQL

Percona XtraDB

PostgreSQL

Etcd

NATS

Redis

KubeDump