Skip to content

How backups and restore requests work Print

  • 0

A backup is useful when it can recover the right data to a state the business can trust. The schedule and retention window are only part of that outcome. You also need to know what is protected, how a restore is requested, which changes a restore would overwrite and who will verify the recovered application.

Netcloud24 Managed Windows VPS plans use automated daily backups with 14-day rolling retention. The Windows VPS 2027 range instead includes a daily system disk backup retained for 24 hours, with an optional 7-day Premium backup. Confirm your exact service and selected options before relying on a recovery window. These arrangements should not be described interchangeably.

Understand schedule, retention and recovery points

A daily schedule describes how often backup jobs are intended to run. Retention describes how long qualifying recovery points are kept. Neither means that every minute within that period is recoverable. The points available for a particular request depend on successful backup completion, the protected disks and the application state captured at the time.

Rolling retention also means older points expire as the window moves forward. If a file was deleted several days ago, the time to report it matters. Do not wait until the end of the window because the business can still open other files. Provide the last time the data was known to be correct as soon as you identify a problem.

Two planning terms help turn this into business decisions. The recovery point objective describes how much recent work the business can afford to lose. The recovery time objective describes how long it can tolerate the service being unavailable. A daily backup may be insufficient for a database that cannot lose a working day’s transactions, even when retention is generous.

Check what the backup actually covers

Inventory the system disk, additional data volumes, application databases, shared folders and externally stored information. A system disk backup should not be assumed to cover data on another disk or in an external service. When storage is added or an application’s data location changes, review the protection scope as part of that change.

Application consistency needs separate attention. A database that is accepting writes may require an application-aware backup, a supported freeze mechanism or its own database backup and log process. Microsoft’s Volume Shadow Copy Service overview describes the coordination involved in application-consistent capture. The presence of VSS on Windows does not prove that a particular backup product successfully protected every application.

Ask the application vendor for the supported recovery procedure and retain any required encryption keys, certificates or installation information securely. A database file without the keys or software needed to read it may not provide a usable recovery. Do not store the only copy of recovery instructions inside the server they are intended to recover.

Choose the smallest recovery that solves the problem

Individual-file recovery can be appropriate when a document was deleted or damaged while the rest of the server is healthy. Recovering to an alternate folder first lets the owner compare versions and avoids overwriting newer edits. Files with the same name may represent different business states, so the path and recovery date both matter.

Application-level recovery follows the software vendor’s procedure. A database may need to be restored through its engine rather than copied over a live data file. Dependencies, users, scheduled processing and transaction logs may be part of the recovery. The application owner should decide when the result is fit for production use.

Whole-server recovery can be necessary after a serious system failure, but it rolls back more than one person’s files. Accounts, application changes, integrations and work completed after the chosen point may also be affected. Establish which current information must be preserved before approving that scope.

For illustration, recovering yesterday’s deleted spreadsheet into a review folder is a different request from returning an accounting server to yesterday morning. The latter can remove today’s transactions for every user. Clearly describing the business problem helps support propose a proportionate recovery method.

Prepare an actionable restore request

  1. Identify the service. Include the server name and service identifier from the customer portal. Avoid relying on a nickname used only inside your organization.
  2. Describe the affected information. Provide the exact folder path, file names, application or database. State whether the issue is deletion, corruption, an unwanted change or a system outage.
  3. Give the recovery timeline. State when the data was last correct, when the problem was first noticed and the time zone. Approximate times are useful if clearly labelled.
  4. Explain current activity. Say whether users are still working, whether integrations are running and whether new data must be preserved.
  5. Name the approver. Identify the authorized contact who can accept downtime and data rollback. Include an application owner who can validate the restored result.
  6. State the preferred destination. Ask whether an alternate location or isolated recovery is possible before replacing live data.

Do not attach passwords, full customer datasets or confidential records merely to demonstrate the problem. Supply enough context to identify the affected object, and use an approved secure transfer method if support needs additional material.

Handle suspected compromise before restoring

If ransomware, unauthorized access or unexplained account changes are involved, say so in the first message. Restoring data while compromised access remains active can lead to another incident. The recovery point itself may also contain malicious changes or credentials that need to be replaced.

Preserve relevant logs and the timeline. Avoid running broad cleanup utilities or deleting suspicious files before the incident owner has assessed the evidence. Follow the agreed containment process and coordinate any isolation with support. A rushed restart or restore can destroy useful context without resolving the original access path.

Do not reconnect a recovered server to external systems automatically. Outgoing email, payment interfaces, synchronization tools and scheduled exports can repeat actions if their state has been rolled back. Review these integrations before allowing the restored application to resume normal processing.

Verify a recovery before calling it complete

First confirm that the intended recovery point and destination were used. Check file names, timestamps, permissions and representative content. For a business application, open the correct dataset, inspect recent records and run a meaningful report. A successful login or a backup console’s green status is insufficient evidence that the required business information is correct.

Have the application owner check consistency against independent records where practical. Record known missing transactions and the plan for re-entering or reconciling them. Avoid letting multiple people independently reconstruct the same work without coordination, which can introduce duplicates and conflicting results.

After acceptance, verify that normal backups resume for the recovered environment and that any temporary recovery copies are handled under your data retention policy. Update the incident record with the chosen recovery point, validation result, downtime and lessons for prevention.

Test and improve the recovery plan

Schedule periodic tests using representative files or an isolated application instance. Select a scenario that matters, such as recovering an older report, restoring a database with its permissions or rebuilding access after a server failure. Record who performed the business validation and how long the process took.

Keep independent protected copies or business exports where the impact of loss requires them. Independence concerns access and failure exposure as well as location; a second copy that the same compromised account can erase may not provide the protection you expect. Confirm the format remains readable and the required keys are recoverable.

Use the support request guide when opening a restore ticket and review security responsibilities when the cause is uncertain. Recovery planning is strongest when scope, ownership and acceptance criteria are settled before data is missing.


Was this answer helpful?

« Back