Jeroen Gordijn

Results 21 comments of Jeroen Gordijn

Can't access the process state in receiveCommand callback in ProcessStep

Agreed. I'm open for PR ;-)

Functional extension: guide the process by a 'pattern matching'-like construct

Yes, that would be nice. Maybe we can first write a test case for it to see when in particular this will be useful.

Functional extension: guide the process by a 'pattern matching'-like construct

I added you as collaborator @gvandeglind . Can you try again?

Improve the DSL of the process framework

Do you have any particular place in mind where it currently falls short?

Improve isolation by making process steps more functional

This will make it possible to drop all the `Option` fields you'll have in the current way to implement state. The compiler can give better support to tell whether your...

Functional extension: process versioning

That would indeed be nice.

Slow (timeout) shutdown and rebalancing with `withRebalanceSafeCommits `

If I set `maxRebalanceDuration`, won't I fallback to old behaviour and get duplicates. Also, please look at the rebalance time. It is not only the shutdown that seems to fail.

Slow (timeout) shutdown and rebalancing with `withRebalanceSafeCommits `

I start consumer1. Then I start consumer 2, which doesn't start immediately (or after a short delay). Meanwhile the new consumer doesn't show anything in the log, and in consumer1...

Slow (timeout) shutdown and rebalancing with `withRebalanceSafeCommits `

A yes, flag (`withoutPartitionPreFetching`) seems to work. Why is this flag there? The Kafka Client also has prefetching, right? I'm a bit worried about the amount of flags and combination...

Slow (timeout) shutdown and rebalancing with `withRebalanceSafeCommits `

I read about prefetching here: https://www.conduktor.io/kafka/kafka-consumer-important-settings-poll-and-internal-threads-behavior/#Kafka-Consumer-Poll-Behavior-0 . Is that incorrect?