Replica configuration for high availability and fault tolerance

For high availability and fault tolerance to be effective, set the appropriate number of replicas for Apache Zookeeper, for embedded Elasticsearch nodes, and for the administrative service.

Apache Flink Zookeeper

It is imperative to have at least 3 Zookeeper replicas running. The number of replicas must be an odd number.

Apache Flink Zookeeper configuration

Initial IBM® Business Automation Configuration Container configuration: Apache Flink > Apache Zookeeper Setup Details > Zookeeper replicas
Helm chart configuration: flink.zookeeper.replicas
Default value: 1
High-availability value: 3

Embedded Elasticsearch

If you plan to install embedded Elasticsearch, set at least 2 client nodes, 2 data nodes, and 3 master nodes.

Client node configuration

Initial IBM Business Automation Configuration Container configuration: Install Elasticsearch and Kibana > Elasticsearch Settings > Client Node Settings > Client node replicas
Helm chart configuration: ibm-dba-ek.elasticsearch.client.replicas
Default value: 1
High-availability value: 2

Data node

Initial IBM Business Automation Configuration Container configuration: Install Elasticsearch and Kibana > Elasticsearch Settings > Data Node Settings > Data node replicas
Helm chart configuration: ibm-dba-ek.elasticsearch.data.replicas
Default value: 1
High-availability value: 2

Master node configuration

Initial IBM Business Automation Configuration Container configuration: Install Elasticsearch and Kibana > Elasticsearch Settings > Master Node Settings > Master node replicas
Helm chart configuration: ibm-dba-ek.elasticsearch.master.replicas
Default value: 1
High-availability value: 3

Administration service

Initial IBM Business Automation Configuration Container configuration: Apache Service > Administration service replicas
Helm chart configuration: admin.replicas
Default value: 1
High-availability value: 2