Configure Usage Telemetry for Confluent Manager for Apache Flink
Starting with version 2.5.0, Confluent Manager for Apache Flink® (CMF) sends anonymous usage telemetry to Confluent once a day. Confluent uses this information to understand how Confluent Platform for Apache Flink is used and to guide its future development. Telemetry is on by default, and you can opt out at any time.
What is collected
Telemetry describes the size and shape of a CMF deployment, not what runs on it. It includes:
Counts of the CMF resources that you manage, such as environments, compute pools, applications, and statements.
The versions of the software that your deployment uses, such as Flink and Kubernetes.
The data is anonymous. It is associated only with a randomly generated identifier for your CMF instance, which is stored in the CMF database. Telemetry does not include the names, configuration, or contents of your resources, or the data that your jobs process.
CMF sends telemetry to https://api.segment.io/. Segment is a
third-party analytics service that Confluent uses.
Opt out of telemetry
Telemetry is on by default and is controlled by a Helm value.
Helm value |
Default |
Description |
|---|---|---|
|
|
Turns usage telemetry on or off. Set to |
Set the cmf.telemetry.enabled value to false in your Helm values file to opt out:
cmf:
telemetry:
enabled: false
Then pass the values file when you install or upgrade CMF:
helm upgrade --install cmf confluentinc/confluent-manager-for-apache-flink \
--values values.yaml
Keep this setting in the values file that you use for later upgrades. If you
upgrade without it, telemetry returns to its default value of true.
For general installation steps, see Install Confluent Manager for Apache Flink with Helm.