New to KubeStash? Please start here.

Addon

What is Addon

An Addon is a Kubernetes CustomResourceDefinition(CRD) which specifies the backup and restore capabilities for a specific type of target. For example, MySQL addon specifies the backup and restore capabilities of MySQL database where Postgres addon specifies backup and restore capabilities for PostgreSQL database. An Addon CR defines these capabilities as tasks for backup and restore.

When you install KubeStash, some Addons will be pre-installed for supported targets like workloads, etc. However, you can create your own Addon to customize or extend the backup/restore process.

Addon CRD Specification

Like any official Kubernetes resource, an Addon has TypeMeta, ObjectMeta and Spec sections. However, unlike other Kubernetes resources, it does not have a Status section.

A sample Addon object for a PostgreSQL database is shown below:

apiVersion: addons.kubestash.com/v1alpha1
kind: Addon
metadata:
  labels:
    app.kubernetes.io/instance: kubedb
    app.kubernetes.io/managed-by: Helm
    app.kubernetes.io/name: kubedb-kubestash-catalog
    app.kubernetes.io/version: v2024.2.14
    helm.sh/chart: kubedb-kubestash-catalog-v2024.2.14
  name: postgres-addon
spec:
  backupTasks:
    - driver: Restic
      executor: Job
      function: postgres-backup
      name: logical-backup
      parameters:
        - name: args
          required: false
          usage: Arguments to be passed to the dump command.
        - default: pg_dumpall
          name: backupCmd
          required: false
          usage: Backup command to take a database dump (can only be pg_dumpall or pg_dump)
        - default: postgres
          name: user
          required: false
          usage: Specifies database user (not applicable for basic authentication)
        - default: "true"
          name: enableCache
          required: false
          usage: Enable or disable caching. Disabling caching may impact backup performance.
        - default: /kubestash-tmp
          name: scratchDir
          required: false
          usage: Directory for holding temporary files and restic cache.
      singleton: true
      volumeMounts:
        - mountPath: /kubestash-tmp
          name: kubestash-tmp-volume
      volumeTemplate:
        - name: kubestash-tmp-volume
          source:
            emptyDir: {}
          usage: Holds temporary files and restic cache.
    - driver: VolumeSnapshotter
      executor: Job
      function: postgres-csi-snapshotter
      name: volume-snapshot
      parameters:
        - name: volumeSnapshotClassName
          required: false
          usage: The VolumeSnapshotClassName to be used by volumeSnapshot
      singleton: true
    - driver: Restic
      executor: Job
      function: kubedbmanifest-backup
      name: manifest-backup
      parameters:
        - default: "true"
          name: enableCache
          required: false
          usage: Enable or disable caching. Disabling caching may impact backup performance.
        - default: /kubestash-tmp
          name: scratchDir
          required: false
          usage: Directory for holding temporary files and restic cache.
      singleton: true
      volumeMounts:
        - mountPath: /kubestash-tmp
          name: kubestash-tmp-volume
      volumeTemplate:
        - name: kubestash-tmp-volume
          source:
            emptyDir: {}
          usage: Holds temporary files and restic cache.
  restoreTasks:
    - driver: Restic
      executor: Job
      function: postgres-restore
      name: logical-backup-restore
      parameters:
        - name: args
          required: false
          usage: Arguments to be passed to the dump command.
        - default: postgres
          name: user
          required: false
          usage: Specifies database user (not applicable for basic authentication)
        - default: "true"
          name: enableCache
          required: false
          usage: Enable or disable caching. Disabling caching may impact backup performance.
        - default: /kubestash-tmp
          name: scratchDir
          required: false
          usage: Directory for holding temporary files and restic cache.
      singleton: true
      volumeMounts:
        - mountPath: /kubestash-tmp
          name: kubestash-tmp-volume
      volumeTemplate:
        - name: kubestash-tmp-volume
          source:
            emptyDir: {}
          usage: Holds temporary files and restic cache.
    - driver: Restic
      executor: Job
      function: kubedbmanifest-restore
      name: manifest-restore
      parameters:
        - default: "true"
          name: enableCache
          required: false
          usage: Enable or disable caching. Disabling caching may impact backup performance.
        - default: /kubestash-tmp
          name: scratchDir
          required: false
          usage: Directory for holding temporary files and restic cache.
      singleton: true
      volumeMounts:
        - mountPath: /kubestash-tmp
          name: kubestash-tmp-volume
      volumeTemplate:
        - name: kubestash-tmp-volume
          source:
            emptyDir: {}
          usage: Holds temporary files and restic cache.

This Addon has three backup tasks and two restore tasks for a PostgreSQL database. Each task refers to a Function.

Here, we are going to describe the various sections of a Addon crd.

Addon Spec

A Addon object has the following fields in the spec section:

• backupTasks : specifies a list of backup tasks that can be performed by the addon. See task specification to learn more about backup tasks.
• restoreTasks : specifies a list of restore tasks that can be performed by the addon. See task specification to learn more about restore tasks.

Task Specification

Each task contains the following fields:

• name : specifies the name of the task. The name of a task should indicate what this task does. For example, a name logical-backup indicate that this task performs a logical backup of a database.
• function : specifies the name of a Function CR that defines a container definition which will execute the backup/restore logic for a particular application.
• driver : specifies the underlying tool that will be used to upload the data to the backend storage. Valid values are:
  • Restic : uses restic to upload backup data.
  • VolumeSnapshotter : leverages Kubernetes VolumeSnapshot crd and CSI driver to snapshot the PVCs.
• executor : specifies the type of entity that will execute the task. Valid values are:
  • Job : KubeStash will create Job(s) to execute the backup/restore task.
• singleton : specifies whether this task will be executed on a single job or across multiple jobs.
• parameters : defines a list of parameters that is used by the task to execute its logic. Parameter definition defines the parameter names, their usage, their requirements etc.
  • name : specifies the name of the parameter.
  • usage : specifies the usage of this parameter.
  • required : specify whether this parameter is required or not.
  • default : specifies a default value for the parameter
• volumeTemplate : specifies a list of volume templates that is used by the respective backup/restore Job to execute its logic. You can overwrite these volume templates using addonVolumes field of BackupConfiguration. Each volumeTemplate contains:
  • name : specifies the name of the volume.
  • usage : specifies the usage of the volume.
  • source : specifies the source of this volume or a template for volume to use.
• volumeMounts : specifies the mount path of the volumes specified in the VolumeTemplate section. These volumes will be mounted directly on the Job/Container created/injected by KubeStash operator. If the volume type is VolumeClaimTemplate, then KubeStash operator is responsible for creating the volume.

Why Function and Addon?

You might be wondering why we have introduced Function and Addon crd. We have designed Function-Addon model for the following reasons:

• Customizable : Function and Addon enables you to customize backup/recovery process. For example, currently we use mysqldump in mysql-backup task to backup MySQL database. You can build a custom Function using Percona’s xtrabackup tool instead of mysqldump. Then you can write an Addon or add a task in existing MySQL Addon with this custom Function and use it to backup your target MySQL database.
• Extensibility : You can easily backup the databases that are not officially supported by KubeStash. You just need to create Functions and an Addon for your desired database.
• Re-usability : Functions are self-sufficient and independent of KubeStash. So, you can reuse them in any application that uses Function-Addon model.

Available Addons

The following addons are available for KubeStash:

If you are using KubeDB with KubeStash enabled. Then you will find database specific addons. The following KubeDB managed database addons are available for KubeStash:

Next Steps

• Learn how KubeStash backup stand-alone PVC using Function-Addon model from here.