search

confluentinc / cp-helm-charts

The Confluent Platform Helm charts enable you to deploy Confluent Platform services on Kubernetes for development, test, and proof of concept environments.
https://cnfl.io/getting-started-kafka-kubernetes
Apache License 2.0
788 stars 846 forks source link

How to disable kafka control-center/ connect logs? #459

Open donton17 opened 4 years ago

donton17 commented 4 years ago

I have started cp-helm-charts with a default persistence volume of 5 GB. After several days it will be getting full. I was at " /opt/kafka/data-0/logs" dir. And I found that "_confluent-monitoring" took 2GB, And "_confluent-controlcenter" took 2.5GB. I think these logs are not needed. And we don't have any property for the cleanup control-center's log.

mr-flannery commented 3 years ago

I used the following workaround: If you take a look at the stateful set that is deployed for the broker pods, there's the following environment variable:

Containers:
  cp-kafka-broker:
    Environment:
      KAFKA_METRIC_REPORTERS: io.confluent.metrics.reporter.ConfluentMetricsReporter

Removing this variable seems to have solved the problem for me (though I haven't yet observed it for the whole period that it took before for the brokers to exhaust the disk). You could also remove it from charts/cp-kafka/templates/statefulset.yaml before installing the chart.

Not sure if this is a good idea or whether it has other unwanted consequences.