When creating stream dynamically
1) it starts to consume in a reasonable time
2) it starts to consume from the earliest offset (consumer_auto_offset_reset='earliest')
Actual behavior
When creating the stream dynamically on runtime there are occurring two unexpected behaviours
1) After the creation of a new stream, it takes approx 45 seconds until it starts to consume from the stream.
2) It ignores configuration
consumer_auto_offset_reset='earliest'
and starts to consume only from that time coming messages
Question:
I wonder if this is the right way to create the topic dynamically or if there is a better approach.
Hello!
Checklist
master
branch of Faust.Steps to reproduce
App worker app.py
Data producer worker data_producer.py
Expected behavior
When creating stream dynamically 1) it starts to consume in a reasonable time 2) it starts to consume from the earliest offset (consumer_auto_offset_reset='earliest')
Actual behavior
When creating the stream dynamically on runtime there are occurring two unexpected behaviours 1) After the creation of a new stream, it takes approx 45 seconds until it starts to consume from the stream. 2) It ignores configuration
and starts to consume only from that time coming messages
Question: I wonder if this is the right way to create the topic dynamically or if there is a better approach.
Full traceback
app.py
Versions
Python version Python 3.7.3
Faust version 1.9
Operating system NAME=Fedora VERSION="28 (Twenty Eight)" ID=fedora VERSION_ID=28 VERSION_CODENAME="" PLATFORM_ID="platform:f28" PRETTY_NAME="Fedora 28 (Twenty Eight)"
Kafka version docker image confluentinc/cp-enterprise-kafka:5.3.1 confluentinc/cp-schema-registry:5.3.1 confluentinc/cp-zookeeper:5.3.1
Many thanks for any insides!