Analysis of official documentation, with no testing on a paid account.

In this guide
On PlanetHoster, restore only the scope needed to resolve the incident. If only files have been replaced or deleted, start with their directory. If the data recorded by the site is damaged, process the database separately. Restore files and database together only when the incident touches both.
This guide is for a freelancer or small team that needs to restore a site without blindly erasing orders, forms or updates received after the backup point. Official N0C and PlanetHoster help consulted on 19 October 2026.
Choose the scope before clicking Restore
A hosting backup contains at least two families of elements: the site files and its databases. Files include code, extensions, themes and media on the server. The database keeps the contents and settings that the application records there. Restoring one does not automatically imply restoring the other.
| Incident | Initial scope to assess | Risk to be avoided |
|---|---|---|
| A file or directory has been deleted, replaced or incorrectly deployed. | The affected files, from a point prior to the modification. | Needlessly revert to recent database data. |
| Content or settings stored as a base have been deleted or corrupted. | The base concerned, after exporting its current state. | Overwrite data created after the backup without making a copy. |
| The code and data structure have changed together. | Files and database from a consistent point, with an intervention window. | Associate an old code with an incompatible basic structure. |
| The cause remains unknown. | An isolated copy or the notice of support before any restoration. | Turn incomplete diagnosis into additional loss. |
This table is a decision method proposed by Logicelio, not an automatic diagnosis. Before taking any action, record when the incident began, the last known change, the proposed backup point and any new data that could be lost. If the site receives orders or enquiries, pause data writes where possible or plan how to recover them before rolling back.
Keep current, even if defective
The N0C restoration procedure asks to copy the affected files and databases elsewhere before restoring. For a base, it returns to an export PhpMyAdmin. For files, she recommends renaming the current directory with the suffix -old before submitting its saved version.
This copy is not just an abstract precaution. It allows you to find a file created from the backup, compare two configurations or return to the starting state if the selected point does not correct the incident. Store it out of the directory you will restore and check that it is readable before continuing.
PlanetHoster also recalls the 3-2-1 backup rule in its good practices : several copies, on different media, including one at a distance. Also keep a copy independent of the host and make sure the company knows how to retrieve it.
Restore in N0C, then read history
For a Web Project using N0C, the documentation indicates this path:
- Open the panel at the address
mg.n0c.com/fr/, then interface Backup. - For a base, first export its current state, choose the day and the MySQL backup concerned, then launch the restoration.
- For files, choose the day, open Select Files and check only the necessary items. Documentation does not recommend restoring the entire panel, particularly because electronic messages could be lost.
- After ordering, seeBackground at the bottom of the screen. The launch message is not enough: the history must confirm the operation.
The button to restore all files exists, but that does not make it the right default choice. A specific directory limits the scope of the operation and makes checking easier. Creating archives from this interface is listed as temporarily disabled in the N0C documentation updated on 29 September 2026; use the features available in your panel at the time of the incident.
Example: an update breaks the display, but the form continues
Hypothetical example proposed by Logicelio: a plugin is updated at 9 a.m. At 9:15 a.m., the layout becomes unusable. Nevertheless, the form records three requests before 10 a.m. A backup from 8 a.m. is available.
Immediately restoring the entire database to its 8 h state could erase the three enquiries. If examination shows the incident comes only from the plugin's files, the narrowest approach is to save the current directory, then restore that directory from the 8 h backup. The database remains intact. Then check the affected page and the three enquiries.
This decision changes if the update also changed the database structure, if its data is corrupted or if the plugin does not work with its older files. In that case, keep the current export, identify data added since 8 h and prepare a coordinated restore. The article does not claim that restoring files is sufficient for every plugin.
If the Web Project still uses cPanel
Thehelp PlanetHoster for World with cPanel returns from the client space to the Backup button and the R1Soft tool. Restore points allow you to browse the file archive, restore a selection or download it. The bases have a separate route.
The full download offered by R1Soft contains files, not databases. If your recovery plan needs both, also retrieve the database using a documented method, such as PhpMyAdmin or mysqldump depending on your configuration. Do not confuse a successful file archive with a full copy of the site.
Before following a procedure, check the Control panel indicated in the details of the Web Project. PlanetHoster distinguishes N0C and cPanel, and the screens are not interchangeable. The guide details of a Web Project also specifies where to check whether N0C backups are active.
Validate resumption before deleting copy
Do not conclude success because the home page is displayed. First check the symptom that triggered the procedure, then the functions that depend on the restored perimeter:
- Open the page or path initially in error;
- control a public page, administration connection and media;
- if the site collects data, perform a clearly identifiable test submission and confirm its arrival;
- If a database has been restored, look for the content created around the time of the chosen point;
- read the N0C history and record the backup point, scope and result.
Keep the renown directory and current export until the end of these controls. Then set a deletion date based on your storage capacity and conservation rules. A copy kept without owner or date quickly becomes a new risk.
Cases requiring further intervention
A restoration returns data to an earlier state; it does not explain the cause. If you suspect an intrusion, compromised account or malicious software, restore and reopen the site without correcting access or vulnerability can reproduce the incident. Isolate the service, keep the necessary diagnostic information and request appropriate assistance.
For a site that cannot accept a break or a loss of transactions, test the back of a copy and organize an intervention window. The documents consulted describe the tools, not a guaranteed time frame, a universal shelf life or the recovery of each application. If no usable points appear, contact the support with the Web Project, the time of the incident and the items to be preserved.
To prepare for this recovery before the breakdown, see PlanetHoster review, the guide to distinguishing backup and restoration and comparative Hostinger or PlanetHoster.
Partner link
Sources and method of this analysis
N0C: Restore files and databases separately ; PlanetHoster France: The World hosting ; PlanetHoster: good backup practices 3-2-1 ; PlanetHoster: restore World hosting with cPanel ; PlanetHoster : panel, backups and details of a Web Project Functions and history must be checked on the panel used. Sources consulted on 19 October 2026. Methodology · Trade transparency · Report correction.