Analytics and deployment status data collection

Apigee hybrid makes analytics and deployment status data available to you. This data is collected by a data collection pod which sends it to your Apigee account management plane so that you can view and analyze it and set up monitoring and alerts.

About analytics and deployment status data collection

All Message Processor services in hybrid stream analytics and deployment status data via TCP to a data collection pod in the cluster. The data collection pod stores the streamed data on the pod's file system via a fluentd service.

The UDCA (Universal Data Collection Agent) periodically extracts the stored data and sends it to the UAP (Unified Analytics Platform) service in the management plane. The UAP processes the incoming analytics and deployment status data and makes it available to you via the Edge UI or management APIs.

Apigee hybrid implements the data collection pod as a ReplicaSet with a minimum of 2 replicas.

The following image shows the analytics and deployment status data collection process:

alt_text

Note that the analytics and deployment status data is not stored in the same location or accessed in the same way as the logging and metrics data:

  • Logging and metrics data are stored on your GCP Project and accessed via a tool such as Stackdriver or whatever you choose to use.
  • Analytics and deployment status data, on the other hand, is stored in the hybrid management plane and you access it via Apigee services (ie, your Apigee account) such as the Edge UI or the management API.

The following table summarizes the data collected by the data collection pod:

Data Dataset Name Description Update Frequency
Analytics api API usage data including transactions per second, cache usage, errors, latencies, request/response sizes, and traffic counts.

For more information, see Analytics metrics, dimensions, and filters reference in the Edge documentation.

Up to 30 second delay
Deployment Status event The current deployment status of the API proxy.

For more information about how you can see this information, see View deployment status.

Immediately

View analytics & deployment status data in the Edge UI

This section describes how to view the analytics and deployment status data in the Edge UI.

Analytics

Analytics data for hybrid services is accessible in the same way as Edge analytics data. For more information, see Using the analytics dashboards in the Edge documentation.

Proxy deployment status

For information on viewing deployment status, see View deployment status.

Configure analytics and deployment status data collection

To set how and where analytics and deployment status data is collected on the data collection pod, you configure the UDCA service via its configuration properties. The UDCA properties include properties that are general to the UDCA as well as properties that are specific to each dataset.

To configure the UDCA:

  1. Open the overrides.yaml file for editing on your Kubernetes administration machine, as described in Customize the runtime plane services.
  2. Set the values of the UDCA configuration settings. For the UDCA, you can set custom values for properties such as:
    • Polling interval
    • Number of replicas (min and max)
    • Target CPU percentage (that triggers additional replicas)

    For a complete list of UDCA properties that you can customize, see udca.

  3. Save your changes to the overrides.yaml file.
  4. Apply your changes to your cluster by executing the apigeectl apply command, as the following example shows:
    apigeectl apply -v beta