​
Overview

The OneSignal + Starburst Enterprise integration enables syncing of custom events from your Starburst Enterprise cluster to OneSignal to trigger automated messaging campaigns and Journeys based on user behavior. Starburst Enterprise is a commercial distribution of Trino designed for enterprise analytics and data lake querying across multiple sources.

​
Requirements

​
Starburst Enterprise

  • Starburst Enterprise cluster with network access
  • User credentials with appropriate permissions
  • TLS connection support (required by OneSignal)
  • Event data accessible through Starburst catalogs

​
Setup

1

Get JDBC connection details

Follow Starburst’s documentation to get your JDBC URL for your desired cluster.Example JDBC URL:
jdbc:trino://census-example-cluster.trino.galaxy.starburst.io:[email protected]/accountadmin
Extract hostname for OneSignal:
census-example-cluster.trino.galaxy.starburst.io
OneSignal uses JDBC to connect to Starburst Enterprise. You only need the hostname portion of the JDBC URL.
2

Configure Starburst Enterprise connection

In OneSignal, go to Data > Integrations and click Add Integration.Select Starburst Enterprise and provide the following connection details:
  • Host: Your Starburst cluster hostname (from Step 1)
  • Username: Your Starburst username
  • Password: Your Starburst password
  • Port: 443 (default) or your custom port
3

Configure Advanced Sync Engine (Optional)

For enhanced performance, set up a dedicated CENSUS catalog:
  1. Create a catalog named CENSUS containing a schema named CENSUS
  2. Ensure your connector supports:
    • CREATE TABLE and DROP TABLE operations
    • Table writes (INSERT, DELETE, UPDATE)
    • CREATE OR REPLACE TABLE statement
  3. Grant full permissions on the CENSUS.CENSUS schema to your OneSignal user
Tested configurations include MySQL, PostgreSQL, Snowflake, Iceberg, and Starburst Delta Lake connectors.

​
Event data mapping

Map your to OneSignal’s custom events format:
OneSignal FieldDescriptionRequired
nameevent_nameEvent identifierYes
external_iduser_idUser identifierYes
timestampevent_timestampWhen event occurredNo
propertiesevent_dataNo

​
Example Event Query

-- Example: Recent high-value events across catalogs
SELECT
    event_name,
    user_id,
    event_timestamp,
    CAST(event_properties AS JSON) as event_properties
FROM catalog.schema.user_events
WHERE event_timestamp >= current_timestamp - INTERVAL '7' DAY
    AND JSON_EXTRACT_SCALAR(event_properties, '$.value') > '100'
ORDER BY event_timestamp DESC;

​
Enterprise Data Lake Queries

-- Example: Federated query across enterprise data sources
SELECT
    'enterprise_activity' as event_name,
    u.user_id,
    current_timestamp as event_timestamp,
    JSON_FORMAT(JSON_OBJECT(
        'crm_interactions', c.interaction_count,
        'warehouse_orders', w.order_count,
        'lake_analytics', l.score_value
    )) as event_properties
FROM salesforce_catalog.users.accounts u
LEFT JOIN crm_catalog.interactions.summary c ON u.user_id = c.user_id
LEFT JOIN warehouse_catalog.orders.summary w ON u.user_id = w.user_id
LEFT JOIN datalake_catalog.analytics.scores l ON u.user_id = l.user_id
WHERE u.created_date >= current_date - INTERVAL '30' DAY;

​
Sync Engine Options

​
Basic Sync Engine

  • Works with any Starburst catalog and connector
  • State tracking managed by OneSignal infrastructure
  • Simpler setup with no additional requirements

​
Advanced Sync Engine

  • Enhanced performance with local state tracking
  • Requires dedicated CENSUS.CENSUS catalog and schema
  • Supports connectors with table write operations
  • Recommended for high-volume enterprise event processing

​
Supported Connectors

OneSignal’s Advanced Sync Engine has been tested with:
  • MySQL connector (read-write mode)
  • PostgreSQL connector (read-write mode)
  • Snowflake connector (read-write mode)
  • Iceberg connector (with S3 and AWS Glue)
  • Starburst Delta Lake connector (with AWS Glue catalogs)

​
Enterprise Features

​
Multi-Source Federation

  • Query across enterprise data sources in a single sync
  • Combine CRM, warehouse, and data lake event data
  • Unified customer event profiles from disparate systems

​
Security & Compliance

  • Enterprise-grade authentication and authorization
  • Row-level security and column masking support
  • Audit logging for data access tracking

​
Limitations

  • TLS connection required (OneSignal security requirement)
  • Advanced Sync Engine requires CREATE OR REPLACE TABLE support
  • Warehouse Writeback not yet supported (coming soon)
  • Cannot provide custom table options in WITH clause

​
FAQ

​
How do I get my Starburst Enterprise hostname?

Follow Starburst’s documentation to get your JDBC URL, then extract just the hostname portion (without jdbc:trino:// prefix) for use in OneSignal.

​
Can I query multiple enterprise data sources?

Yes! Starburst Enterprise’s federated query capabilities allow you to combine event data from multiple enterprise sources (Salesforce, SAP, Oracle, etc.) in a single query.

​
Which Starburst release supports the Advanced Sync Engine?

Check your Starburst Enterprise release notes for CREATE OR REPLACE TABLE support, which is required for Advanced Sync Engine functionality.