Operator Architecture

This document describes the architecture of Doris Operator and how it works.

Architecture#

DorisCluster, DorisInitializer, DorisMonitor, DorisClusterAutoscalerare custom resources defined by CRD.

• DorisClusterdescribes the desired state of the Doris cluster;
• DorisMonitordescribes the monitoring components of the Doris cluster;
• DorisInitializer describes the desired initialization Job of the Doris cluster;
• DorisClusterAutoscaler describes the automatic scaling of the Doris cluster;

The following components are responsible for the orchestration and scheduling logic in a Doris cluster:

• doris-controller-manager is a set of custom controllers in Kubernetes. These controllers constantly compare the desired state recorded in the ``DorisCluster` object with the actual state of the Doris cluster. They adjust the resources in Kubernetes to drive the Doris cluster to meet the desired state and complete the corresponding control logic according to other CRs;
• doris-admission-webhook is a dynamic admission controller in Kubernetes, which completes the modification, verification, operation, and maintenance of Pod, StatefulSet, and other related resources.

Control flow#

The diagram below illustrates the control flow analysis of the Doris Operator:

The overall control flow is described as follows:

1. The user creates a DorisCluster object and other CR objects through kubectl, such as DorisMonitor;
2. Doris Operator watches DorisCluster and other related objects, and constantly adjust the StatefulSet, Deployment, Service, and other objects of FE, BE, Broker, Monitor or other components based on the actual state of the cluster;
3. Kubernetes’ native controllers create, update, or delete the corresponding Pod based on objects such as StatefulSet, Deployment, and Job;

Based on the above declarative control flow, Doris Operator automatically performs health check and fault recovery for the cluster nodes. You can easily modify the DorisCluster object declaration to perform operations such as deployment, upgrade, and scaling.