Hostinger AI Builder: How to troubleshoot data synchronization in agentic mode

Fix missing database fields between test and live data by republishing your Hostinger AI Builder project in agentic mode.

Updated 1 week ago

Fix missing database properties or structural fields between test and live data in Hostinger AI Builder agentic mode by republishing your project. The integrated backend separates your testing environment from your live deployment into two independent environments, test data and live data, to protect your production environment. Changes made to your data tables during drafting may not immediately sync to your live site.

Understand the Two-Bucket data model

To correctly troubleshoot synchronization issues, it is essential to understand that Hostinger AI Builder isolates test schemas from live data rows to keep your live application safe:

  • The Separate Bucket Concept: Think of the test and live environments as two identical iron buckets. The test bucket holds apples, while the live bucket holds bananas. Their structural shape is identical, but their contents (the actual rows of data) will always differ.
  • Data Entries (Rows): Creating a mockup user or adding content rows inside your test database view is purely for preview purposes. These rows will not automatically transfer or synchronize into the public live database bucket.

Push structural schema updates live via publishing

While content rows do not transfer between environments, data structures (such as a new role field or a custom table column) must be identical across both buckets. If a field exists in your test data view but is missing from your live production dashboard, your live structure needs an update.

  1. Open your Hostinger AI Builder editor and navigate to your project’s database table view.
  2. Confirm that the new structural properties (such as custom columns or fields) are correctly configured in your test data view.
  3. Click the Publish button in the top-right corner of the editor to republish your project. This triggers a full project republication — not just a database update — which syncs your live data structure to match your current test environment.
  4. Once publishing completes, your live database schema will mirror your test data structure. Existing rows in each environment remain isolated and are not affected.

Data entries remain safely isolated within their respective buckets, but structural updates require a full project republication to sync successfully. If a new data column is missing on your live Hostinger AI Builder site, publishing your latest changes will bring the live schema up to date.

NOTES

  • Unpublished Drafts: If you add new parameters or properties to a data table inside Hostinger AI Builder but exit without publishing, the live site’s database structure will lag behind, resulting in missing fields on the live application.
  • Hostinger AI Builder Product Isolation: Hostinger AI Builder operates on an entirely automated, non-file-based architecture. Standard hosting dashboard tools in hPanel (like File Manager backups or manual database migration wizards) do not apply to Hostinger AI Builder data tables.