Copies of important data
Keep the working copy and two backup copies of data that matters to your operation.
BACKUP & RECOVERY / PREPARE FOR THE UNEXPECTED
Identify the data you need to recover, then choose the copy schedule, retention and restore process. Build the backup plan around your website or application.
Illustrative recovery checkpoints
Review the latest available version.
Compare a copy before a recent change.
Recover within the agreed storage policy.
Example only. No real backups are stored or restored in this website preview.
SEPARATE THE COPIES. TEST THE RECOVERY.
The backup design should consider failure of the original system, the backup copy and the access path to both.
Keep the working copy and two backup copies of data that matters to your operation.
Use different media types to reduce dependence on a single storage failure mode.
Keep a copy away from the primary location, with appropriate separation and access controls.
Planning reference: CISA / US-CERT Data Backup Options. A particular package’s storage design is confirmed in its service scope.
EXPLORE A RECOVERY WORKFLOW
Choose a proposed backup scope and retention window, then inspect an example restore point. This is a planning preview, not a connected backup console.
Proposed retention window
Schedules and retention are planning preferences. Storage capacity, supported workloads, restore procedures and fees require confirmation.
MAKE THE SERVICE BOUNDARIES CLEAR
Identify the files, databases and configuration that must be copied. Confirm storage location, access controls and separation from production.
Agree the copy frequency, retention window, storage allowance and behaviour when limits are reached. Old versions may expire under the retention policy.
Define who can request a restore, where the data is restored and how the restored service is checked before production use.
DEFINE YOUR RECOVERY PRIORITIES
The right design depends on how much recent data you can afford to lose and how quickly the service needs to return.
| Planning decision | What to confirm |
|---|---|
| Recovery point | The acceptable gap between your last usable copy and a data-loss event. |
| Recovery time | The target time to restore a defined workload, with dependencies and testing included. |
| Restore destination | Whether data is restored to an isolated test environment or to the production service. |
| Responsibility | Who requests, approves, performs and validates each recovery step. |
Any recovery objectives, supported restore types or time commitments must be expressly agreed. This page does not establish an RPO, RTO or service-level guarantee.
BACKUP QUESTIONS
Do not assume that a service includes an independent backup. Check the product specification or quotation for the covered data, frequency, retention, storage separation and restore terms. Optional daily backups shown in the existing hosting checkout are a separate example add-on.
The terms describe different mechanisms and can vary by platform. A snapshot may depend on the same underlying system as the original data. Confirm whether a copy remains recoverable when the primary service or its storage is unavailable.
It can, depending on the destination and restore method. Agree the affected files or databases, preserve the current state where appropriate and test the recovery before replacing production data.
Older copies may be removed under the agreed retention policy. Long-term archives, legal holds or custom retention need separate confirmation; keeping copies indefinitely is not assumed.
No. The controls demonstrate a planning and review workflow. They do not connect to storage, inspect a server or modify any files.
BUILD A CONNECTED RECOVERY PLAN
Share the workload, data size and rate of change. Include your recovery priorities so the copy and restore requirements can be reviewed together.