This document provides detailed instructions for setting up and configuring the Apigee API hub connector for Private Cloud and API hub, which facilitates the transfer of API metadata and runtime analytics from your Apigee Edge for Private Cloud installation to Google Cloud's API Hub.
The Apigee Edge for Private Cloud and API hub integration is designed to bridge your on-premise Apigee environments with Google Cloud's API Hub. This integration enables the seamless flow of critical API metadata (like API proxy definitions, policies, etc.) and runtime analytics data from your Apigee for Private Cloud installations directly into API Hub, providing a centralized and comprehensive view of your API landscape.
Overview
At its core, this integration allows Apigee for Private Cloud users to onboard their API information into API Hub. This means you can leverage API Hub's capabilities for discovery, governance, and centralized management of APIs originating from your on-premise Apigee deployments.
Key components
- API hub (Google Cloud): The central cloud-based platform where API metadata and runtime data are consolidated and managed.
- Apigee hub connector: A new component installed alongside your Apigee Edge for Private Cloud environment. This connector is the primary agent responsible for extracting data from Apigee Edge for Private Cloud and securely publishing it to Google Cloud.
- Message processors (MPs): Existing Apigee components that are reconfigured to "dual-write" runtime analytics data to a shared Network File System (NFS), making it available for the Apigee API hub connector.
- Network file system (NFS): A shared storage volume used to temporarily store analytics data from MPs before the Apigee API hub connector processes and uploads it.
- Google cloud pub/sub: A messaging service used by the Apigee API Hub for Private Cloud connector to securely stream API metadata and runtime data to API hub.
- Google Cloud service accounts: Used for secure authentication and authorization of the Apigee API Hub for Private Cloud connector to interact with Pub/Sub and API hub services.
The integration flow
The integration follows a structured process to ensure data transfer and accuracy:
- Initial setup & provisioning:
- Begin by setting up a Google Cloud project and provisioning API Hub within it.
- Create Google Cloud Service Accounts and register your Apigee for Private Cloud organizations with API Hub by creating "Plugin Instances." This step provides unique Pub/Sub topics and Plugin Instance IDs essential for configuration.
- Apigee API hub for Private Cloud connector installation:
- The Apigee API hub for Private Cloud connector, an RPM package, is installed on a dedicated VM alongside your Apigee for Private Cloud environment. This requires a compatible Apigee for Private Cloud version.
- Apigee API hub for Private Cloudconnector configuration:
- The connector is configured with details obtained during the API Hub plugin creation (Pub/Sub topics, Plugin Instance IDs, Service Account details).
- Authentication to both Google Cloud services (via service account keys) and the Apigee Management Server (via base64-encoded credentials) is established.
- The path to the shared NFS is also configured here.
- Message processor (MP) configuration:
- Your Apigee message processors are updated to enable a "dual-write" mechanism for analytics data. This means analytics records are written to both the traditional analytics system and the designated NFS share.
- Data flow:
- API metadata: The Apigee API hub for Private Cloud connector pulls API metadata from the Apigee management server and publishes it to the designated metadata Pub/Sub topic in Google Cloud, which then feeds into API hub.
- Runtime data: Message processors write analytics data to the shared NFS. The Apigee hub connector monitors this NFS, processes the runtime data, and publishes it to the designated runtime data Pub/Sub topic in Google Cloud for ingestion into API Hub.
- Monitoring & troubleshooting:
- A local status endpoint on the Apigee API hub for Private Cloud connector allows you to monitor the progress of metadata and runtime data uploads.
- Detailed logs are available for troubleshooting any issues related to startup, data transfer, or connectivity.
Interdependencies:
- The Apigee API Hub for Private Cloud connector relies on the NFS being correctly mounted and accessible by both itself and all Message Processors.
- Message Processors must be configured for dual-write to the NFS to enable runtime data collection.
- Correct Google Cloud Service Account roles and API Hub Plugin Instance details are vital for the connector's authentication and data publishing capabilities.
This integration provides a robust mechanism for bringing your on-premise Apigee APIs into the centralized management and visibility provided by API Hub, empowering better API governance and insights.