Set Up a Database for Your Site

If your site stores data, such as sign-ups, posts or orders, it needs a database. The editor’s Database tab sets one up and lets you work with it.

Hiveku Database

A managed PostgreSQL database on Supabase, created in one click, with sign-in, file storage and automatic backups

Bring your own

A PostgreSQL database you already run, such as Supabase, Neon, PlanetScale or Railway, connected through DATABASE_URL

A site that only shows pages, such as a marketing site, a blog or a portfolio, needs neither. You can add a database at any time.

Create a Hiveku Database

Open the Database tab

In the editor, click Database. A site without a database shows No Database Configured and two choices.

Create it

Under Create Hiveku Database, click Create Database. The button reads Creating… until the database is ready, and then the tab shows Database Connected.

Use it

There is nothing to copy. Hiveku saves the connection details as your site’s environment variables (listed below), and your code reads them.

Creating a database is a paid feature, and your plan sets how many databases your account can have. You need the Database permission (Websites › Database) to create one.

The variables it sets

Variable	What it is for
`DATABASE_URL`	The connection your site uses. It goes through the database’s connection pooler, which suits a site that runs serverless.
`DIRECT_URL`	A direct connection, for migrations and tools that need one long session, such as `prisma migrate`.
`NEXT_PUBLIC_SUPABASE_URL`, `NEXT_PUBLIC_SUPABASE_ANON_KEY`	The Supabase client in the browser: sign-in, storage and realtime.
`SUPABASE_SERVICE_ROLE_KEY`	Full access to the database from server code. Never send it to the browser.

Each one is also saved for Development and Production separately (for example DATABASE_URL_DEV and DATABASE_URL_PROD). They point at the same database: a site has one database, shared by every branch and environment, unless you turn on database branching. See Environment Variables for how environment-specific variables work.

Connect Your Own Database

Copy the connection string

From your provider, copy the PostgreSQL connection string. Use its pooled connection if it offers one (Supabase and Neon both do): your site runs serverless, and every running copy opens its own connections.

postgresql://user:password@host:5432/dbname

Add it as DATABASE_URL

In the Database tab, click Add DATABASE_URL under Bring Your Own Database. Your site’s environment variables open in a new tab. Add DATABASE_URL with the connection string and save. If you run Prisma migrations through a pooled connection, also add your provider’s direct connection as DIRECT_URL.

Check it

Back in the Database tab, click Refresh. The tab shows Database Connected, with the type BYOD (External).

With your own database, you manage its backups, scaling and billing with your provider. The Auth, Storage and Functions tabs read the Supabase project behind a Hiveku Database. On your own database they show what they can read, or say why they can’t.

The Database Tab

Tab	What you can do
Overview	See the connection (provider, region, tables and rows), and turn on database branching
Tables	Browse a table’s rows, search them, insert, edit and delete rows, and see the table’s schema
Auth	See the people who signed up to your site (Supabase Auth), the database’s roles, and its row level security policies
Storage	Create file buckets, public or private, and manage their files
Functions	Deploy, edit, test and delete Edge Functions (TypeScript on Supabase’s Deno runtime), and set their secrets
Settings	Project details and API keys, extensions, Postgres settings, the connection pooler and realtime, and restarting the database
SQL	Run SQL and see the results, with your query history and a button to copy the results

Run SQL

Open the SQL tab, type a query and press ⌘+Enter (Ctrl+Enter on Windows). Try:

SELECT 1;

The result shows under the query. A statement that returns no rows, such as CREATE TABLE, says Query executed successfully (no rows returned). To turn on a PostgreSQL extension, run CREATE EXTENSION "extension_name"; here. The Settings tab lists the extensions.

Let the AI create tables

You can also ask the AI in the editor to create or change tables, for example “Add a posts table with title, slug, content, author_id and published_at”. See Create Tables.

Database Branching

With the Database Pro add-on, a site can have database branches, separate copies for its environments such as staging. You can then try changes away from your live data.

Turn it on: in the Overview tab, click Upgrade to Pro — $29/mo and confirm.
Add a branch: type a name (for example staging) and create it. Delete one from its row.
Turn it off: click Downgrade. Branching is turned off for this site and its branches are deleted. If no other site in your account uses Database Pro, the add-on is cancelled right away.

Use the Database in Code

DATABASE_URL is available to your server code once a database is connected.

Prisma
Drizzle
Plain pg

In prisma/schema.prisma, use the pooled connection for queries and the direct one for migrations:

datasource db {
  provider  = "postgresql"
  url       = env("DATABASE_URL")
  directUrl = env("DIRECT_URL")
}

import { PrismaClient } from "@prisma/client";

// One client per server instance, reused across requests
const globalForPrisma = globalThis as unknown as { prisma?: PrismaClient };
export const prisma = globalForPrisma.prisma ?? new PrismaClient();
if (process.env.NODE_ENV !== "production") globalForPrisma.prisma = prisma;

import { drizzle } from "drizzle-orm/node-postgres";
import { Pool } from "pg";

const pool = new Pool({ connectionString: process.env.DATABASE_URL });
export const db = drizzle(pool);

import { Pool } from "pg";

// Create the pool once and reuse it; a new pool per request runs out of connections
export const pool = new Pool({ connectionString: process.env.DATABASE_URL });

Backups

A Hiveku Database is backed up automatically. To see backups, make one before a risky change, or restore one, see Back Up Your Database.

Troubleshooting

The tab says Connection Error

For a Hiveku Database, click Repair Connection. Hiveku finds the database’s correct address, resets its password and saves fresh connection variables. Your tables, rows and files are kept. Redeploy each environment afterwards: a deployed site keeps using the old password until it is deployed again.For your own database, check DATABASE_URL on your site’s environment variables page. It must match your provider’s connection string exactly, and copy-paste slips are the most common cause.

Creating a database says it is a paid feature

A Hiveku Database needs a paid plan, and each plan allows a set number of databases. Upgrade the plan, or connect your own database instead.

'Too many connections' errors

Each running copy of your site opens its own connections. Use a pooled connection string (a Hiveku Database’s DATABASE_URL already is one), and create one database client per server instance rather than one per request.

Prisma migrations hang or fail

Migrations need a direct connection, not the pooled one. Set directUrl = env("DIRECT_URL") in your Prisma schema, as above.

My provider refuses the connection

Some providers only accept connections from addresses you list. Allow connections from anywhere, using SSL and a strong password.

I ran the wrong DROP TABLE

Restore the most recent backup. See Back Up Your Database.

What’s Next?

Environment Variables

Manage secrets and settings for each environment

Create Tables

Design your schema by hand or with the AI

Get Started

Dashboards

AI Assistant

Code Editor

Design Studio

CMS

Email Marketing

Social Media

Advertising

Marketing Delivery

Commerce

Calendar

Communications

Helpdesk

Sales

Publishing

Native Mobile

Email

Integrations

How-Tos

Account & Settings

Partners

Set Up a Database for Your Site

Hiveku Database

Bring your own

Create a Hiveku Database

The variables it sets

Connect Your Own Database

The Database Tab

Run SQL

Let the AI create tables

Database Branching

Use the Database in Code

Backups

Troubleshooting

What’s Next?

Environment Variables

Create Tables

Hiveku Database

Bring your own

​Create a Hiveku Database

​The variables it sets

​Connect Your Own Database

​The Database Tab

​Run SQL

​Let the AI create tables

​Database Branching

​Use the Database in Code

​Backups

​Troubleshooting

​What’s Next?

Environment Variables

Create Tables

Create a Hiveku Database

The variables it sets

Connect Your Own Database

The Database Tab

Run SQL

Let the AI create tables

Database Branching

Use the Database in Code

Backups

Troubleshooting

What’s Next?