Connections API

Endpoints

CRUD Operations

• List Connections - Get all connections for a project
• Get Connection - Get connection details
• Create Connection - Create a new connection (general)
• Update Connection - Update a connection (general)
• Delete Connection - Delete a connection

Connection Types

Each connection type has its own documentation page with complete examples:

• Email Connection
• Kafka Connection
• Elasticsearch Connection
• RabbitMQ Connection
• FTP Connection
• Graylog Connection
• Syslog Connection
• Webhook Connection
• Logback Connection
• Apache ActiveMQ Connection
• SNMP Connection
• Linux Script Connection
• Database Connection
• LDAP Connection

Authentication

All endpoints require authentication using a Personal API Access Token.

Permissions

GET Operations (List/Get Connection)

• User must have CONNECTIONS + VIEW permission in the project
• If the asset category does not exist, user must have at least one permission in the project

POST/PUT/DELETE Operations (Create/Update/Delete Connection)

• User must have CONNECTIONS + MANAGE permission in the project

Permission Requirements

Operation	Required Permission
List/Get Connection	CONNECTIONS + VIEW (or any permission if category doesn't exist)
Create/Update/Delete Connection	CONNECTIONS + MANAGE

Secret Fields

• List Operations: Secret fields (passwords, API keys, etc.) are masked and returned as null
• Get Single Connection: Secret fields are never returned either; the response carries them as null. Round-trip a GET response into PUT as-is and the stored secrets are kept (absent = keep, "" = clear, value = replace)
• Kafka properties / Webhook headers: the same absent-keeps/empty-clears/value-replaces rule also covers a Kafka connection's propertiesMap entries and a Webhook connection's headerList headers whose key/name looks sensitive — see Kafka Connection and Webhook Connection
• Security: Never commit connection configurations with secrets to version control

Portable References

Kafka, Syslog, LDAP, Webhook, and Email use portable certificate/key-store names on APIops. Their storage ID fields are internal: APIops rejects any request containing one (including null or blank forms), omits them from List/Get, and hides them from OpenAPI. For PUT and POST upsert, an absent or null name preserves a still-valid stored reference, "" clears an optional reference, and a non-empty name resolves a replacement. A dangling or foreign stored reference must be repaired by name or cleared before the connection can be written.

Related Documentation

• Authentication Guide - How to obtain and use API tokens
• Error Handling - Error response formats