How long are hosting backups kept? Check your recovery window

Quick answer

Backup frequency and retention settings are not a list of recoverable dates. Identify your accepted service terms, then inspect the actual completed points for the affected service. Choose a usable point before the incident and plan how to preserve newer legitimate changes.

There is no single retention period for every Novasean service

Check your actual service and accepted backup terms. Shared - 1 is unmanaged, so the customer is responsible for website backups and recovery. VM Start + Managed OS has an included VM-backup arrangement only where its schedule forms part of the accepted order.

Backup frequency is how often a job is scheduled. Retention settings decide which completed copies should remain. Neither proves that a particular recovery point exists or that it contains a working application. A failed job, missing component or unsuccessful restore can change what is usable.

Identify who made and holds the backup

For a DirectAdmin user-level archive, check who created it, where the original and any downloaded copy are held, and which versions you preserved. DirectAdmin documents a User-level Site Backup interface but no scheduling option in that interface. Administrator or reseller backup jobs can use different schedules, paths and overwrite rules; do not infer their history from a user-level screen.

For VM Start + Managed OS under the published current schedule, Novasean operates a daily VM backup to Proxmox Backup Server in the same Netherlands datacentre. The covered source is the supplied VM disks within the service allocation. Separately mounted remote storage, external databases or services, local devices and changes after a completed backup are outside that coverage. The VM copy does not establish application-consistent database recovery. A customer-owned application export or independent copy may have its own dates and retention.

Read retention tiers as settings, not a calendar promise

The current VM Start + Managed OS schedule specifies three most recent, seven daily, four weekly and six monthly backup retention settings. Treat these as retention settings, not an inventory: they do not promise 20 available snapshots or a copy for every date in the last six months. PBS applies retention options in order, with later options considering older backups rather than counting already-covered snapshots again. Missed or failed jobs can leave fewer recovery points. Proxmox Backup Server describes last, daily, weekly and monthly pruning as ways to select which existing snapshots to keep; it cannot create a missing historical backup.

This schedule applies only where the accepted order incorporates it. An older or customised agreement may use a different backup method, retention rule or responsibility split. The generic Novasean backup and recovery terms require the applicable schedule to state frequency, retention and exclusions; the generic page itself gives no period. If you cannot identify the effective schedule, ask support to confirm it before relying on a particular date.

Check the actual available points before choosing one

Record the affected account or VM, the first known bad time, the last known good time and their time zones. Keep the precise identifier in the authorised support route rather than a public message.
Locate your own archived copies and their creation times. For operator-managed backups, ask the authorised operator which completed snapshots currently exist for this specific service and whether any jobs failed or coverage changed.
Choose a candidate point older than the damaging change but recent enough to limit legitimate data loss. Check whether files, database state and other required components can be recovered consistently; use the related backup-content guide for that inventory.
Ask whether the candidate point can be tested on an isolated target and who can validate the application. The related isolated-restore guide covers a safe rehearsal.

A backup filename, retention tier or green job status is weaker evidence than an inspected available point and a successful application test. Do not postpone an urgent request while an older point may be nearing expiry.

Example: daily scheduling does not prove yesterday is recoverable

A daily job may have failed, or a retained snapshot may not contain a working application. Ask for the specific service’s completed snapshot list and errors. Choose by timestamp, covered components and a safe validation route—not by adding retention settings together or assuming a filename proves success.

If no usable point predates the damaging change, check separately held application exports and copies with their owners. Do not promise reconstruction from a retention setting or delay an urgent request while an older point may expire.

Request recovery without discarding newer data

Give support the service identifier, incident time and time zone, candidate point, affected components, business impact, changes made since that point and the person authorised to approve data loss. Ask for the available points, exact restore target, likely overwrite, expected downtime, scope of included work and the application validation owner. For VM Start under its accepted schedule, ordinary VM-level restoration from an available included backup is part of the service; additional application recovery requires separate agreement before work starts. Do not infer a response or restoration deadline.

Before any destructive restore, preserve the current state where feasible and agree how later orders, uploads, mail or other legitimate changes will be reconciled. Use the related restore-scope guide to choose between one item, matched application data, account components and a whole VM. Do not send passwords, private keys, full backups or raw customer data in an ordinary ticket unless a secure transfer route has been agreed.

Common questions about retention and recovery

Does a daily backup mean I can restore yesterday? No. Ask whether yesterday’s job completed, the snapshot remains available and it contains what your application needs.

Do six monthly retention settings guarantee a six-month-old copy? No. They select among snapshots that actually exist. The daily, weekly and monthly calendar periods can overlap, but PBS applies the options sequentially without counting already-covered snapshots again. The accepted schedule and current snapshot list determine the real options.

Can I see VM Start backup snapshots in DirectAdmin? Do not assume so. DirectAdmin account archives and operator-managed Proxmox VM backups are different systems. Ask the authorised operator for the VM points.

Will a VM restore recover an external database or make my application consistent? The current VM Start schedule excludes external services and does not promise application-consistent database recovery. Check separately held data and have the application owner validate the result.

Sources and technical references

Prepared by Team Novasean from the references above. Menus and available features depend on your installed version and service permissions.

Related guides

Was this guide helpful?

If you accepted optional analytics, your Yes/No answer and this guide’s page address are sent to Google Analytics. No written feedback is collected. Read our privacy statement or change your choice in .

Need help with your service?

Open your client area or contact Novasean. Describe the symptom and checks you have performed. Keep passwords, recovery codes and private keys out of your message.