liveBook · Manning

Knative Eventing

As explained in Google Anthos in Action

Knative Eventing is the Knative component that obtains events from sources, routes them through brokers and triggers, and delivers them to targets.

Knative Eventing orchestrates events from inside or outside a Kubernetes cluster, routes them through event-mesh resources, and delivers them to Knative or Kubernetes Services. It supports predefined integrations and custom event sources while scaling event-driven containers from a few events to sustained streams. section 9.3.3

Decompose an Eventing pipeline into these responsibilities:

  • Event sources send events to a Broker and act as the control-plane integration for external systems. section 9.3.3
  • A Broker provides the event mesh that receives or pulls events and makes them available for routing. section 9.3.3
  • Triggers filter Broker events by CloudEvents attributes and route matching events to subscribers. section 9.3.3
  • Controllers reconcile these resources from Kubernetes custom resource definitions, while the messaging layer can use memory or persistent storage. section 9.3.3

Route a Cloud Pub/Sub event

Deploy a CloudPubSubSource for the testing topic and point its sink at an event-display Service. A Trigger can then select events by type and source from the default Broker and deliver them to a named Service. This separates source integration, filtering, and application delivery instead of embedding event handling in the service.

Install Eventing on a Kubernetes cluster by applying its CRDs, core components, default channel layer, Broker layer, and optional source extensions. YAML files or an operator support installations across managed and on-premises Kubernetes environments. Choose a cluster deployment when portability, direct infrastructure access, custom hardware, or service-mesh integration outweighs the operational simplicity of fully managed Cloud Run. section 9.3.5

How Google Anthos in Action covers it

How deep it goes
Explains sources, Brokers, Triggers, CRDs, CloudEvents, storage choices, and installation, while deferring complex pipelines for basic usage.
Examples
Uses Broker and Trigger manifests, a CloudPubSubSource, and event-delivery Services to demonstrate routing.
What it assumes
Assumes a Kubernetes cluster and introduces Eventing installation through CRDs, core components, channels, Brokers, and sources.
What you can do after
Build event pipelines from source integrations, Brokers, filtered Triggers, and Kubernetes or Knative Service targets.
How it connects
Connects Kubernetes, CloudEvents, event sources, Brokers, Triggers, Knative Serving, Kubernetes Services, Cloud Run, and Anthos.

Also explained in

Compare how 2 books explain Knative Eventing →

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Google Anthos in Action. The section links point to the parts of the book it draws on.