OpenStack Compute

OpenStack Compute resources represent Nova servers, backed up and restored through the openstack-instance protocol of the OpenStack integration. A backup holds the server’s flavor and networks, its root disk, and every Cinder volume attached to it. A restore recreates the disks as new Cinder volumes, so the original disks are never modified.

Inventory Management

Managed inventories can discover Nova servers automatically once the OpenStack inventory is connected to a project. OpenStack Compute resources are only discovered by the OpenStack inventory and cannot be set up by a self-managed inventory.

Backup and restore

A backup snapshots the root disk and each attached volume, and copies them through Glance without stopping the server. The disks are captured one after another rather than at the same instant, so they can reflect slightly different points in time. Every disk must fit within the Glance image size limit of the cloud. The limit is 1 TiB by default but it can be raised by the operator of the cloud.

Servers that boot from a volume are not supported. Their root disk is a Cinder volume rather than a Nova disk, and the backup refuses them.

A restore onto an existing server creates a new Cinder volume for every backed-up disk, including the root disk, and attaches them to that server. No server is created, and the disks already attached to the server are left as they are.

Backup flow

flowchart TD
  subgraph OpenStack["OpenStack Project"]
    Server["Nova server"]
    Volumes["Attached Cinder volumes"]
    Images["Glance images"]
  end

  subgraph Plakar["Plakar Control Plane"]
    Source["OpenStack Compute
Source app"]
    Backup["Backup process
Encrypt & deduplicate"]
  end

  Store["Kloset Store"]

  Source -->|"uses OpenStack APIs"| Server
  Server --> Volumes
  Server -->|"root disk snapshot"| Images
  Volumes -->|"volume snapshots"| Images
  Images -->|"read disks"| Backup
  Backup --> Store

Restore flow

flowchart TD
  Store["Kloset Store"]

  subgraph Plakar["Plakar Control Plane"]
    Destination["OpenStack Compute
Destination app"]
    Restore["Restore process"]
  end

  subgraph OpenStack["OpenStack Project"]
    Images["Temporary Glance images"]
    Volumes["New Cinder volumes"]
    Server["Existing Nova server"]
  end

  Store --> Restore
  Destination --> Restore
  Restore -->|"upload disks"| Images
  Images -->|"create volume
from each disk"| Volumes
  Volumes -->|"attach"| Server

Shared configuration

The following settings are available when configuring both source and destination apps.

• Openstack Auth Url: Required. The Keystone v3 endpoint, for example https://keystone.example.com:5000/v3.
• Openstack Region: Required. A single region. For a source app, the region the server runs in. For a destination app, the region the server is restored to.
• Openstack Application Credential Id and Openstack Application Credential Secret: The application credential to authenticate with. Both are required together, and they take precedence over a username and password.
• Openstack Username and Openstack Password: The credentials of an OpenStack user, used when no application credential is set.
• Openstack Project Id: The project to authenticate to with a username and password. Takes precedence over the project name.
• Openstack Project Name: The project to authenticate to when no project ID is set.
• Openstack Domain Name: The domain of the user, and of the project when it is given by name. Defaults to Default.

Permissions

Backing up and restoring servers creates and deletes snapshots, images and volumes in the project, so the credentials need the reader and member roles on it. See Managing Users and Application Credentials for how to create a user and an application credential with those roles.

Role names depend on how the OpenStack cloud is configured. The reader and member roles are default Keystone roles, but some providers define their own. Consult the documentation of the provider for the equivalent roles.