How to backup ParadeDB in PostgreSQL

How do I back up ParadeDB in PostgreSQL?

backup ParadeDB creates a portable dump of all tables, indexes, and extension objects created by ParadeDB so you can restore them on another PostgreSQL server.

Welcome to the Galaxy, Guardian!

Oops! Something went wrong while submitting the form.

Description

Why back up ParadeDB with pg_dump?

ParadeDB stores its data inside ordinary PostgreSQL tables. Using the native pg_dump utility guarantees a consistent snapshot, includes extension metadata, and lets you restore on any compatible server.

What is the safest backup command?

Run pg_dump --format=custom --create --file=paradedb.backup your_db. The custom format supports parallel restore, compression, and integrity checks—ideal for large vector or search indexes built by ParadeDB.

How do I back up only ParadeDB objects?

Filter by schema or extension: pg_dump --schema=paradedb --file=paradedb_objects.sql your_db. This keeps the dump small while preserving all ParadeDB-specific functions, tables, and indexes.

Can I schedule automatic backups?

Yes. Create a cron job or CI pipeline that calls pg_dump daily, uploads the file to S3, and uses --clean --if-exists flags to simplify automated restores.

How do I restore the backup?

Use pg_restore --dbname=target_db --create paradedb.backup. The --create flag rebuilds the database, reinstalls the ParadeDB extension, and replays all data.

Best practice: test restores regularly

Automate a nightly restore into a staging server. Verify that vector search queries still return expected results from Products or Orders.

Best practice: keep schema and data together

Always back up extension metadata (--create) along with table data. Omitting this causes CREATE FUNCTION errors on restore.

Where are the dumps stored?

Save dumps outside the database host. Popular options are AWS S3, GCS, or an on-prem object store. Encrypt files at rest using gpg or S3 server-side encryption.

Why How to backup ParadeDB in PostgreSQL is important

How to backup ParadeDB in PostgreSQL Example Usage


-- Back up only ParadeDB search indexes on the Products table
pg_dump \
  --schema=paradedb \
  --table=products_vect_index \
  --file=products_paradedb.sql \
  galaxy_store

How to backup ParadeDB in PostgreSQL Syntax


pg_dump [OPTIONS] database_name

Main options:
  -Fc, --format=custom          # compressed, parallel-ready dump
  -f,  --file=<path>            # output file
  -c,  --clean                  # include DROP statements
  -C,  --create                 # include CREATE DATABASE
  -n,  --schema=<schema>        # dump only this schema (e.g., paradedb)
  -T,  --table=<pattern>        # dump specific tables
  -j,  --jobs=<n>               # parallel dump
  -Z,  --compress=<0-9>         # compression level

Example (e-commerce database):
pg_dump -Fc -f paradedb.backup -C -Z9 galaxy_store

This dumps the entire <code>galaxy_store</code> database that contains:
  • Customers(id, name, email, created_at)
  • Orders(id, customer_id, order_date, total_amount)
  • Products(id, name, price, stock)
  • OrderItems(id, order_id, product_id, quantity)
plus any ParadeDB indexes built on <code>Products</code>.

Common Mistakes

Skipping the --create flag: Restores fail because the ParadeDB extension is not recreated automatically. Always use --create or manually run CREATE EXTENSION parade_db before pg_restore.
Relying on filesystem snapshots: They capture in-flight transactions and can corrupt ParadeDB indexes. Use pg_dump running in single-transaction mode for consistency.