Rafael Acevedo
Rafael Acevedo
I managed to get it working with Micronaut 3, I'll try to come up with a PR
I dug a bit deeper into the rxjava2 dependency and it seems to work fine as long as the final user includes the `micronaut-rxjava2` dependency (it includes basically the `RxJavaConverterRegistrar`...
I'm having similar issues. It would be great to know if there's anything wrong with this approach.
Hi @xmlking! Thanks for raising this issue. The current code does load the schema multiple times in a concurrent miss scenario, but the performance does not suffer that much, as...
Hi guys! I guess this occurs for every batch consumer. As micronaut sends the whole batch to the user implementation, it has no way of knowing which record caused the...
cc @yawkat this is the issue I mentioned on [gitter](https://gitter.im/micronautfw/questions?at=6321b3bc6837563d1cd2bd1f)
PR with suggested fix: https://github.com/micronaut-projects/micronaut-kafka/pull/602
I can confirm that for some of our workloads, upgrading from 4.5 to 4.6 makes the cache hit ratio decrease significantly, causing performance problems. We'll try to provide more details...
> thanks @racevedoo for the update we looked for OSS users to confirm this - do you have screenshots showing this - just so we will have some user workload...
> I think in easy first step is a kill switch. That will be a regression compared to 4.5 since 4.5 shared index reads for the two bounds of a...