Jump to: navigation, search

kafka-<cluster-name> Section


Kafka cluster–specific configuration sections enable you to specify connection and topic information that enables Genesys Info Mart to consume data from Apache Kafka, in order to make data from producer applications available in the Info Mart database for downstream reporting applications.

If you want Genesys Info Mart to process data that a particular data source published to a Kafka topic, add the applicable configuration section(s) and options on the Options tab of the Genesys Info Mart Application object. Name the section kafka-<cluster-name>, where <cluster-name> is any alphanumeric string that identifies the Kafka cluster in your deployment, such as kafka-1.

Each cluster-specific configuration section can contain the following options.


Default Value: No default value
Valid Values: Any valid host:port combination that identifies a Kafka server in the cluster
Changes Take Effect: On the next ETL cycle
Dependencies: None

Specifies the location of the Kafka broker(s) in the cluster, in the form of host:port. If there are multiple servers, use a comma-separated list.

bootstrap.servers is a standard Kafka consumer option. The option is mandatory for Genesys Info Mart, as a Kafka consumer, to know where to connect for the initial connection to the Kafka cluster. In high availability (HA) deployments, Genesys recommends that you list all the brokers in the cluster.

For more information about the bootstrap.servers option, see the Apache Kafka documentation (https://kafka.apache.org/documentation#consumerapi).


Default Value: No default value
Valid Values: Any Genesys Info Mart–defined <mapping-id>, as listed in the option description
Changes Take Effect: On the next ETL cycle
Dependencies: None

Specifies a Kafka topic to consume and how messages in this topic will be mapped. The <topic-name> in the option name identifies the topic Genesys Info Mart will look for in Kafka message headers and matches the topic name configured in the producer application. The option value identifies the ID in the CTL_XML_CONFIG table that Genesys Info Mart will use to map that topic into the database schema.

Configure a separate g:topic:<topic-name> option for each topic to which the Kafka producers publish reporting data. A particular kafka-<cluster-name> section might contain multiple g:topic:* options, some of which might have the same <mapping-id> value, depending on the data source.

Genesys Info Mart supports the following <mapping-id> values:

Producer application <mapping-id> Supported since Genesys Info Mart release
Bot Gateway Server (BGS) BGS_K
Genesys Co-browse (GCB) COBROWSE


Default Value: 10 (seconds)
Valid Values: A number (of seconds) or duration in ISO 8601 format
Changes Take Effect: On the next ETL cycle
Dependencies: None

Specifies the idle timeout for polling Kafka records. If polling does not return any records within the timeout, Genesys Info Mart stops polling Kafka until the next ETL cycle.

In releases earlier than, the timeout was hard-coded, with a value of 2 seconds. In high-latency networks, the hard-coded value sometimes caused Genesys Info Mart to skip transformation of Kafka records, even if data was available.

This page was last modified on November 29, 2018, at 13:21.


Comment on this article:

blog comments powered by Disqus