Public | Automated Build

Last pushed: 2 years ago
Short Description
kafka cluster on rancher
Full Description


rancher-kafka image based in rancher-jvm8

To build

docker build -t <repo>/rancher-kafka:<version> .

To run:

docker run -it <repo>/rancher-kafka:<version> 

How it works

  • The docker has the entrypoint /usr/bin/, that that check rancher-metadata server connectivity, starts confd and monit. It checks, reconfigures and restart the kafka cluster, every $CONFD_INTERVAL seconds.
  • Kafka memory params could be overrided by JVMFLAGS env variable.
  • Scale could be from 1 to n nodes. Recommended to use odd values: 3,5,7,...
  • Default env variables values:

CONFD_BACKEND=${CONFD_BACKEND:-"rancher"} # Default confd backend
CONFD_PREFIX=${CONFD_PREFIX:-"/latest"} # Default prefix to rancher-metadata backend
CONFD_INTERVAL=${CONFD_INTERVAL:-60} # Default check interval
RANCHER_METADATA=${RANCHER_METADATA:-"rancher-metadata.rancher.internal"} # Default rancher-metadata server
KAFKA_HEAP_OPTS=${JVMFLAGS:-"-Xmx1G -Xms1G"} # Default kafka memory value

Docker Pull Command
Source Repository