Tutorials overview
These tutorials are end-to-end backup recipes. Each one walks through producing a consistent dump of a specific database, verifying it, optionally restoring it, and uploading it to Lighthouse. Use the scheduling guide after the one-off flow works.
How the tutorials fit together
Every database backup follows the same two-stage shape:
┌─────────────────────────┐ ┌──────────────────────────────┐
│ Stage 1 — Create a dump │ ──▶ │ Stage 2 — Upload snapshot │
│ (per-database guide) │ │ (per-database guide) │
└─────────────────────────┘ └──────────────────────────────┘
app.dump / app.sql ./db-dumps ──▶ snapshot
- Stage 1 is database-specific. Pick your database below.
- Stage 2 uses the same minimal Go SDK
Backup()call in every database tutorial. - Scheduling is optional. When you are ready, use Automated backup with scheduling to wrap the dump and upload in cron or a systemd timer.
All dumps land in a single ./db-dumps directory using a stable filename that you overwrite each run. The SDK then uploads that directory as an incremental snapshot.
Pick your database
| Database | Dump tool | Guide |
|---|---|---|
| 🐘 PostgreSQL | pg_dump | PostgreSQL |
| 🐬 MySQL / MariaDB | mysqldump | MySQL |
| 🪶 SQLite | sqlite3 .backup | SQLite |
| 🍃 MongoDB | mongodump | MongoDB |
| ⚡ Amazon DynamoDB | aws dynamodb | Amazon DynamoDB |
| 🪣 Amazon S3 | aws s3 sync | Amazon S3 |
Then automate it
Once you can produce a dump by hand, wire it into the scheduled job:
➡️ Automated backup with scheduling — a dump script + SDK uploader + cron/systemd timer for a nightly backup.
Before you start
You'll need (once):
-
A Lighthouse account and workspace — Web Portal & Free Workspace.
-
An API key scoped
backup:write,backup:read,snapshots:read— API Keys. -
The BaaS Go SDK installed (for the upload step):
go get github.com/lighthouse-web3/baas-go-sdk@latest
Export your credentials so every tutorial can reuse them:
export LH_API_KEY="lh_xxxxxxxxxxxxxxxxxxxxxxxx"
export LH_WORKSPACE_ID="550e8400-e29b-41d4-a716-446655440000"