Closed bbaugher closed 9 years ago
You are correct that the environment variables config is the best way to set java options. The next is what environment variables to set. This is provided by kafka and its run/start server scripts. You can see an example here where its using $KAFKA_JVM_PERFORMANCE_OPTS
and provides the value -server -XX:+UseG1GC -XX:MaxGCPauseMillis=20 -XX:InitiatingHeapOccupancyPercent=35 -XX:+ExplicitGCInvokesConcurrent -Djava.awt.headless=true
by default. There are others for heap ($KAFKA_HEAP_OPTS
).
I'd assume these environment variables are the same as the ones in confluent's kafka package but you may need to verify in its package/distribution.
I'll see if I can add a small bit in the README about this
Perfect thank you for pointing that out, appreciate it.
I didn’t know if it was global java options or per argument ones on Kafka, great to have clarity.
👍
Can you provide an example in the readme about how to set the java arguments for one of the services? In other cookbooks I have used the arguments are provided as a string.
When I set the arguments in this book they are not applied to the running tasks.
The tasks seem to have their default memory settings.