Who Should Use TrilioVault

TrilioVault for OpenStack serves three distinct groups of users, each with a different scope of responsibility and a different set of interactions with the system. Understanding which role applies to you determines which tasks you own, which interfaces you use, and which permissions you need. This page describes OpenStack administrators, backup administrators, and tenant users — and explains how their responsibilities fit together to deliver end-to-end data protection for your cloud.

OpenStack Administrators

If you are responsible for deploying and operating the OpenStack infrastructure itself, TrilioVault asks you to take on the foundational work that makes data protection possible for everyone else. Your responsibilities fall into three broad areas.

Deployment and cluster management. You deploy TrilioVault nodes as QCOW2 images on standalone KVM hosts that sit outside the OpenStack cloud being protected — TrilioVault nodes cannot run as VMs under the same cloud they back up. You provision the cluster IP (VIP) required by Pacemaker/HAProxy, register it as the Keystone service endpoint, and decide whether to run a single-node setup or the recommended three-node cluster for fault tolerance and load balancing. Ansible playbooks, Salt states, and Juju charms in the triliovault-cfg-scripts component streamline this process across diverse environments.

Infrastructure prerequisites. Before tenant users can protect a single workload, you must ensure that OpenStack Keystone, Nova, and Cinder are in place and that a backup target — an NFS share, an S3-compatible endpoint, or OpenStack Swift — is reachable from the TrilioVault nodes. You also ensure that the POSIX UID and GID of the nova user are identical across all TrilioVault cluster nodes and compute nodes, because all read and write operations on the backup target run as nova:nova. Finally, you verify that Cinder quotas are sized to accommodate two Cinder snapshots and one temporary volume per disk being backed up at any given time.

Role and permission management. You assign the TrilioVault Trustee RBAC role to users who need to create workloads, take snapshots, or perform restores. This role is mandatory and cannot be substituted by the OpenStack admin role. You also configure the File Recovery Manager Glance image with the property tvault_recovery_manager=yes so it appears as a valid mount target in the Horizon UI for file-level restores.

Backup Administrators

Backup administrators work at the policy level. You are typically an OpenStack operator or a dedicated data-protection engineer who governs how and when protection happens across the cloud, without necessarily managing the underlying infrastructure.

Snapshot policy management. You define the schedules and retention rules — collectively called snapshot policies — that tell wlm-cron how often to fire automatic snapshots for each workload and how many previous snapshots to retain before older ones are deleted. Getting these policies right is a balance between recovery-point objectives and the storage capacity of your backup target.

Backup target oversight. You monitor the health and capacity of the configured backup targets (NFS, S3, or Swift) and respond when storage thresholds are approached. Because incremental snapshots capture only changed blocks and full snapshots skip zeroed or empty blocks, understanding your data-change rate helps you forecast storage consumption accurately.

Notifications and compliance. You configure the SMTP server used for email notifications and work with tenant project owners to ensure that user accounts have valid email addresses recorded in Keystone — both are required before per-tenant notifications can be delivered.

Horizon plugin administration. If your organization uses the horizon-tvault-plugin, you install and maintain it so that tenant users can access the Workloads panel in the OpenStack Horizon dashboard. You also handle TrilioVault version upgrades using the configuration management scripts in triliovault-cfg-scripts, which are idempotent and safe to re-run.

Tenant Users

Tenant users are the application owners, developers, and cloud consumers who need their workloads protected. You interact with TrilioVault primarily through self-service interfaces — either the Horizon dashboard plugin or the workloadmgr CLI provided by the workloadmanager-client component — and your scope is limited to resources within your own OpenStack project.

Defining workloads. You create workloads by grouping the VMs that make up an application along with their attached Cinder volumes. Because TrilioVault protects at the application level rather than per VM, you can capture a multi-VM application as a consistent unit in a single snapshot. Keep in mind that each VM may belong to only one workload at a time, so plan your groupings around application boundaries.

Taking and scheduling snapshots. You can trigger ad-hoc snapshots on demand or attach a snapshot policy to a workload to automate protection. When a snapshot policy is in place, wlm-cron fires jobs at the configured intervals without any further action on your part.

Restoring workloads and files. When recovery is needed, you choose between a full workload restore — which recreates all VMs and volumes from a chosen snapshot — and a file-level restore, which retrieves individual files from inside a snapshot without spinning up entire VMs. For network topology restores, be aware that the target tenant's networking environment must be completely empty of existing networking components before you begin; any conflict will cause the restore to fail. After the restore completes, you must manually reconnect the restored private networks to the public network, because TrilioVault does not perform that final step automatically.

How the Roles Interact

These three roles form a layered model. OpenStack administrators establish the platform and permissions; backup administrators set the protection policies and maintain the infrastructure those policies depend on; tenant users act within the boundaries that administrators have defined to protect and recover their own applications. In smaller organizations a single person may wear all three hats, but understanding the responsibilities each role carries helps you audit who owns each decision and where to look when something needs attention.

The following examples illustrate the kinds of tasks each role typically performs. They are representative rather than exhaustive.

Tenant user: create a workload and take an on-demand snapshot using the CLI

This shows a tenant user grouping two VMs into a workload named webapp-prod and immediately triggering a snapshot. The user must hold the TrilioVault Trustee RBAC role.

# Create the workload
workloadmgr workload-create \
  --name webapp-prod \
  --instance instance_id=<vm-1-uuid> \
  --instance instance_id=<vm-2-uuid>

# Confirm the workload was created
workloadmgr workload-list

# Trigger an on-demand snapshot
workloadmgr snapshot-create <workload-uuid> --name webapp-prod-snap-01

Expected output after workload-list:

+--------------------------------------+-------------+---------+
| ID                                   | Name        | Status  |
+--------------------------------------+-------------+---------+
| 3f2a1c00-xxxx-xxxx-xxxx-xxxxxxxxxxxx | webapp-prod | available |
+--------------------------------------+-------------+---------+

Expected output after snapshot-create:

+------------+--------------------------------------+
| Property   | Value                                |
+------------+--------------------------------------+
| id         | 7b8e9d11-xxxx-xxxx-xxxx-xxxxxxxxxxxx |
| name       | webapp-prod-snap-01                  |
| status     | executing                            |
+------------+--------------------------------------+

Backup administrator: attach a snapshot policy to an existing workload

This shows a backup administrator updating a workload to apply a daily snapshot schedule with a retention of seven snapshots.

workloadmgr snapshot-policy-create \
  --workload-id <workload-uuid> \
  --fullbackup-interval 1 \
  --retention-policy-type Number_of_Snapshots_to_Keep \
  --retention-policy-value 7

Expected output:

Snapshot policy applied to workload <workload-uuid>.
Next scheduled snapshot: <calculated datetime>

OpenStack administrator: verify the nova user UID/GID consistency across nodes

Before backup jobs can write to the backup target, the nova user's POSIX UID and GID must match on every TrilioVault node and every compute node. Run this on each host and compare results.

id nova

Expected output (must be identical on all nodes):

uid=162(nova) gid=162(nova) groups=162(nova)

If the values differ across nodes, correct them before starting backup operations to avoid permission failures on the backup target.

• Workload — Understand what a workload is, how VMs and volumes are grouped, and why the workload-centric model matters for consistent application recovery.
• Snapshot — Learn how point-in-time snapshots are captured, what data is included (including automatic network topology capture), and how incremental and full snapshots differ.
• Snapshot Policy — Explore how to define schedules and retention rules so that workloads are protected automatically without manual intervention.
• Restore — Understand the full restore workflow, including in-place and cross-project restores, network topology restore prerequisites, and the manual step required after network topology restore.
• File-level Restore — Learn when and how to recover individual files from a snapshot without recreating entire VMs.
• Backup Target — Review the supported backup target types (NFS, S3, Swift) and the configuration requirements for each.
• TrilioVault Trustee RBAC Role — Understand why this role is required and how to assign it through Keystone.
• Workload Manager (WLM) — Get an architectural overview of the wlm-api, wlm-scheduler, wlm-workloads, and wlm-cron services that power TrilioVault behind the scenes.