Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.mantrixflow.com/llms.txt

Use this file to discover all available pages before exploring further.

This quick start uses a realistic scenario instead of placeholder demo data. Scenario: Lighthouse Commerce runs its application on Neon PostgreSQL and wants an analytics copy of the orders table in a separate PostgreSQL destination. The team wants one manual validation run first and an hourly schedule after that.

​
Before you begin

Have these values ready:
  • Source host, database, username, password, and schema for the operational PostgreSQL database
  • Destination host, database, username, password, and schema for the reporting database
  • A source user with read access and a destination user with write access
  • SSL enabled for both connections if they are cloud-hosted

​
Step by step

  1. Sign in and switch to the correct organization.
  2. Open Connections and create a Source connection with PostgreSQL.
  3. Fill the form with real values such as:
    • Connection Name: Neon Production Orders
    • Host: ep-lighthouse-prod-123456.us-east-2.aws.neon.tech
    • Port: 5432
    • Database: app
    • Username: mf_readonly
    • Schema: public
    • SSL Mode: require
  4. Click Test Connection and save the source after the test succeeds.
  5. Create a second Destination connection with values such as:
    • Connection Name: Analytics Postgres
    • Host: analytics-db.internal.lighthouse.io
    • Database: warehouse
    • Username: mf_writer
    • Schema: analytics
    • SSL Mode: require
  6. Go to Data Pipelines → + New Pipeline, name it Orders to Analytics, choose Neon Production Orders as the source, and click Create & open canvas.
  7. On the canvas, click ⚙️ on the Source node:
    • Click Discover schema to load available tables.
    • Tick Include next to public.orders.
    • Click Preview to verify rows.
  8. Optionally add a SQL transform. Use {{ source }} as the table reference:
SELECT
  id,
  customer_id,
  LOWER(order_status)          AS order_status,
  total_amount,
  created_at,
  updated_at
FROM {{ source }}
WHERE order_status IS NOT NULL
  1. Click ⚙️ on the Destination node. In the Config tab:
    • Connection: Analytics Postgres
    • Final delivery schema: analytics
    • Sync mode: FULL_TABLE
    • Write mode: Upsert
    • Click Validate config, then open the Preview tab to confirm the target schema.
  2. Click ▷ on the Destination node to start the first manual run.
  3. Click the history icon in the top bar and confirm analytics.orders_live contains the expected rows.
  4. Once the manual run succeeds, open the Scheduling tab in the Destination panel and set an hourly schedule.

​
What success looks like

You now have a working pipeline that uses the same flow most teams follow in production:
  • reusable source and destination connections
  • a named pipeline shell
  • a selected source table
  • an optional SQL transform (for PostgreSQL-to-PostgreSQL pipelines)
  • a configured destination branch
  • saved configuration, run history, and scheduling
From here, you can expand to more tables, more branches, or another supported SQL destination.