Kafka dies during docker startup on Windows about flowing-retail HOT 5 CLOSED

Suspicion: Related to the auto-creation of the topic on-the-fly (I found some forum entries that point in this direction: https://stackoverflow.com/questions/48117696/kafka-streams-in-docker-compose-takes-long-time-for-partition-assignment).

Tried:

• Own docker container that creates the topic on startup - no effect: kafka.zip
• Added properties to application.properties - no effect:

spring.cloud.stream.bindings.input.consumer.configuration.[session.timeout]=600000 spring.cloud.stream.kafka.binder.configuration.[metadata.max.age.ms]=1000

from flowing-retail.

Suspicion: Probably a Windows problem?

from flowing-retail.

Suspicion: Probably use another docker container (e.g. wurstmeier or original confluent).

Problem: Could not get it to run immediately because of problems with advertised hosts. Haven't investigated further so far.
docker-compose.zip

from flowing-retail.

I now use the official Confluent Kafka & Zookeper image in the latest version.

While everything now starts up my docker environment still freezes regularly - making it unusable for playing around or giving demos.

Will try on a different fresh Windows 10 machine

from flowing-retail.

I could finally solve this problem. It was simply related to Docker Compose not having enough memory available.
On my current machine (16GB physical RAM) what works for me is to allocate 6GB for Docker Compose:

from flowing-retail.