Connections

Connections define how to reach external systems — reusable connection parameters for services, storage, and messaging endpoints.

What are Connections?

A Connection Asset stores the technical parameters needed to connect to an external system. Unlike Sources and Sinks (which define what to read or write), Connections define how to reach the system itself:

• Endpoint configuration — hostnames, ports, regions, and URLs
• Authentication — credentials, tokens, certificates, and OAuth settings
• Protocol specifics — connection timeouts, encryption modes, and protocol versions
• Connection pooling — reuse and lifecycle management settings

Connections are referenced by Sources, Sinks, and Services, allowing you to define connection parameters once and reuse them across multiple Assets.

Connections vs Sources/Sinks

Understanding the separation helps you build maintainable integrations:

Connections	Sources / Sinks
Define how to connect (endpoint, credentials)	Define what to do (read from, write to)
Reusable across multiple Source/Sink	Specific to a particular operation
Configure once, reference many times	Configure location and behavior
Example: Kafka broker URLs and TLS settings	Example: Topic name, consumer group, offset strategy

Benefits of separating Connections

Define a Connection Asset when:

• The same external system is used across multiple Sources or Sinks
• You need to centralize credential management for security
• Connection parameters change infrequently compared to operational settings
• You're connecting to shared infrastructure (Kafka clusters, cloud storage, email servers)

Example workflow:

1. Create one Connection Kafka with your broker URLs and TLS configuration
2. Reference it from multiple Kafka Source and Kafka Sink Assets
3. When brokers change, update the Connection once — all Sources and Sinks automatically use the new settings

Connections Overview

Connection	Purpose	Common Use Case
AWS	Connect to Amazon Web Services	S3, SQS, SNS, Kinesis, EventBridge integration
Email	Connect to mail servers	SMTP, IMAP, POP3, MS 365 email services
FTP	Connect to FTP/SFTP servers	File transfer with FTP, FTPS, or SFTP
Google Cloud	Connect to Google Cloud Platform	GCS buckets and Google Cloud services
Kafka	Connect to Apache Kafka clusters	Stream processing infrastructure
MSGraph	Connect to Microsoft Graph API	OneDrive, SharePoint, Teams integration
NFS	Connect to NFS shares	Network file system access
SMB	Connect to SMB/CIFS shares	Windows file sharing and NAS devices
Virtual File System	Abstracted file system connections	Unified access to multiple storage types
WebDAV	Connect to WebDAV servers	Web-based distributed authoring and versioning

Common Configuration Patterns

Most Connection Assets share these configuration sections:

• Name & Description — Asset identification and documentation
• Required Roles — Restrict usage to specific cluster node roles
• Endpoint Settings — Hosts, ports, regions, and connection URLs
• Authentication — Credentials, API keys, certificates, or OAuth configuration
• Advanced Options — Timeouts, encryption, connection pooling, and protocol-specific settings

See Also

• Sources — Use Connections to define where data comes from
• Sinks — Use Connections to define where data goes
• Services — Use Connections for database, cache, and API service access