[
https://issues.apache.org/jira/browse/KAFKA-14264?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Philip Nee updated KAFKA-14264:
-------------------------------
Description:
To refactor the consumer, we changed how the coordinator is called. However,
there will be a time period where the old and new implementation need to
coexist, so we will need to override some of the methods and create a new
implementation of the coordinator. In particular:
# ensureCoordinatorReady needs to be non-blocking or we could just use the
sendFindCoordinatorRequest.
# joinGroupIfNeeded needs to be broken up into more find grain stages for the
new implementation to work.
We also need to create the coordinator state machine.
was:
To refactor the consumer, we changed how the coordinator is called. However,
there will be a time period where the old and new implementation need to
coexist, so we will need to override some of the methods and create a new
implementation of the coordinator. In particular:
# ensureCoordinatorReady needs to be non-blocking or we could just use the
sendFindCoordinatorRequest.
# joinGroupIfNeeded needs to be broken up into more find grain stages for the
new implementation to work.
Summary: Refactor coordinator code (was: Create a coordinator class
for the new implementation)
> Refactor coordinator code
> -------------------------
>
> Key: KAFKA-14264
> URL: https://issues.apache.org/jira/browse/KAFKA-14264
> Project: Kafka
> Issue Type: Sub-task
> Components: consumer
> Reporter: Philip Nee
> Assignee: Philip Nee
> Priority: Major
>
> To refactor the consumer, we changed how the coordinator is called. However,
> there will be a time period where the old and new implementation need to
> coexist, so we will need to override some of the methods and create a new
> implementation of the coordinator. In particular:
> # ensureCoordinatorReady needs to be non-blocking or we could just use the
> sendFindCoordinatorRequest.
> # joinGroupIfNeeded needs to be broken up into more find grain stages for
> the new implementation to work.
> We also need to create the coordinator state machine.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
