If you liked Spring Fu, read this proposal 😉

Wasn't spring-fu about full control over the configuration and bean registration process (including when the whole initialization code is executed)? With this new proposal is doesn't seem to be the case (the developer have no way to decide when the

initialize

method is executed). Am I correct?

That's true, and I have refined my comment accordingly. This proposal only focus on the part that is not incompatible with Spring Boot core principles like auto-configuration. Spring Fu was IMO too far away from Spring Boot, we have to find a better tradeoff between explicit configuration and auto-configuration. But this improvement could allow later a more explicit configuration model for Spring Boot if the Boot team and community want to support that.

My KotlinConf talk « Next level Kotlin support in Spring Boot 4 » has been accepted. If you want to help to spread the word: • Linkedin • Bluesky • Mastodon See you at Copenhagen in May!

How does Spring MVC work with kotlin coroutine? Do I always have to use kotlin coroutine for my server code? Any examples would be appreciated.

Anyone here knows what is wrong with

spring-dependency-management

plugin and Kotlin/JS compiler? Not long ago I've reported it breaks incremental compilation of K/JS (https://youtrack.jetbrains.com/issue/KT-74378). Now I have problems with serialization (see: https://kotlinlang.slack.com/archives/C7A1U5PTM/p1740943473638099). I've always thought this plugin just reads some BOMs and auto-applies correct version of various dependencies. But how it is possible that it breaks compilation?

I have just merged into Spring Framework main branch a first-class support for programmatic bean registration for Java and Kotlin. Will be part of the upcoming Spring Framework 7.0.0-M3. https://docs.spring.io/spring-framework/reference/7.0-SNAPSHOT/core/beans/java/programmatic-bean-registration.html • Bluesky • Mastodon • LinkedIn

I'm trying to extend the RabbitMQ

AbstractRoutingConnectionFactory

and override

getTargetConnectionFactory

. My function signature is just

override fun getTargetConnectionFactory(lookupKey: Any?): ConnectionFactory?

but the compiler is rejecting that, and only accepts

override fun getTargetConnectionFactory(lookupKey: Any): ConnectionFactory

. Looking at how

getTargetConnectionFactory

is used, it can clearly accept a null

lookupKey

and return a null

ConnectionFactory

. Am I missing something obvious here? I don't see any nullability annotations on

getTargetConnectionFactory

AbstractRoutingConnectionFactory

or any

package-info.java

that specifies package-wide null behavior. Spring Boot 3.4.3, spring-rabbit 3.2.3, Kotlin 1.9.23.

According to the dependency documentation, the upcoming Spring Boot 3.5 will still be on Kotlin 1.9.x. Is there any information about why they do not upgrade to 2.x?

Hello, I have something that I cannot wrap my head around when it comes to observability and coroutines (in a spring boot application). In particular, we want to restore propagate tracing information through SQS, so we are saving trace information on the producer side within the message, and on the consumer we restore the trace before processing the message. Pseudocode (no worries, more code in thread):

// producer thread, saves its trace information within the message
producer.produce(message.withTraceDetails())

// consumer thread, actually even another application instance
// up to this consumer.read, its an independent trace
val message = consumer.read()
processAfterRestoring(message.traceDetails()) {
  // here the trace is restored from the message
  process(message)
}

fun processAfterRestoring(sourceTrace: Trace, supplier: () -> Any)

The code we have works in normal scenarios, but not when

process(message)

uses coroutines, eg

// here we have TraceONE
val message = consumer.read()
processAfterRestoring(message.traceDetails()) {
  // here the trace is restored from the message
  process(message)
}

fun process(message: Message) {
  // here the trace is restored from the message
  launchNewCoroutine {
    // but here the trace is again TraceONE
    doSomethingOn(message)
  }
}

fun processAfterRestoring(sourceTrace: Trace, supplier: () -> Any)

I reproduce some more code (in thread, hopefully I captured all the relevant pieces). I guess the 3 questions that pop up to me are: • are we restoring traces incorrectly (not kotlin)? • is there some issue with our Spring context config (also not kotlin)? • what magic am I not seeing in

observationRegistry.asContextElement()

that causes the issue (since when we don't use coroutines, restoring works I am kinda discounting the first 2 questions and focusing more on this last one)?

Does Kotlin support the annotations present in

org.springframework.lang

? The online Javadoc for the website does indicate that it does https://docs.spring.io/spring-framework/docs/current/javadoc-api/org/springframework/lang/package-summary.html however, on Kotlin's own source file, it is not mentioned anywhere here: https://github.com/JetBrains/kotlin/blob/master/core/compiler.common.jvm/src/org/jetbrains/kotlin/load/java/JvmAnnotationNames.kt. I am confused, does Kotlin support it or not? And if so, how does it do so despite not being listed?

My talk on spring boot + functional kotlin at devnexus is up on youtube: I hope i did it justice!

guys anyone encountered same problem in kotlin spring boot https://stackoverflow.com/questions/79570300/om-mongodb-mongocommandexception-command-failed-with-error-8000-atlaserror

Say I have a

createdAt

field in my

@Entity

. and it gets its value as the Entity is created. Can this info be utilized, so that "all saved entity has non-null createdAt" at kotlin level, or do I have to check against null every time? Or I have no choice but declare it as

var timestamp: Instant? = null,

Are there any out-of-the-box solutions for propagating traces across Kotlin's coroutine

Channel

?

Hi folks, can somebody please help me figure out why my app is being booted in 20+ seconds instead of 5-6 like it was just this morning? I didn't do any particular changes to gradle, I didn't update any version dependencies, it just suddenly started to be incredibly slow. I switched to Linux from Windows for the performance and because Gradle kept hanging up in dying on Windows and the only way to continue work was to reboot, everything was blazingly fast but now for no specific reason everything's just slow af

I know this might not be the correct place to ask, but does anyone know why spring decided to stop publishing their

spring-boot-parent pom

?

spring-boot-parent
The spring-boot-parent module is no longer published. It provides dependency management for internal dependencies used, for example, in Spring Boot's own tests. If you were using spring-boot-parent, replace it with dependency management of your own that meets your application's needs.

The source code of Spring Petklinik, the new reference Kotlin fullstack application I have built for my KotlinConf and Spring IO talks, is now available at https://github.com/sdeleuze/spring-petklinik. I will continue to refine it. If you want to spread the word: - LinkedIn - BlueSky - Mastodon

So I'm upgrading from Spring 3.2 to 3.4 In Spring 3.2 this way of registering an initializer to kick off registering beans for the kotlin beans DSL is deprecated:

context.initializer.classes: mypackage.BeansInitializer

So instead I've used META-INF/spring.factories:

org.springframework.context.ApplicationContextInitializer=\
 mypackage.BeansInitializer

and my application breaks. The issue is that in my BeansInitializer class I'm using ref() to inject a bean that is created via an @Component annotation in a jar file. The trouble is that now the initalizer seems to run before it's dependent bean is created, so I get a message like this: A component required a bean of type 'Whatever' that could not be found. How do correctly use the kotlin beans DSL in Spring 3.4.5 ? NB I can't modify the main function to add the initializer that way as that only works for the main application, not tests, and I don't want to add the initializer to loads of integration tests either.

Hello. I have a very basic setup for running Springboot tests using a postgresql test container. The first test runs fine, however subsequent tests all fail and yell the following at me:

2025-06-21T10:42:16.910+02:00  WARN 74817 --- [app] [    Test worker] com.zaxxer.hikari.pool.PoolBase          : HikariPool-1 - Failed to validate connection org.postgresql.jdbc.PgConnection@5a0c0680 (This connection has been closed.). Possibly consider using a shorter maxLifetime value.

My base test class looks like this:

@Testcontainers
abstract class SpringBootTestBase {

    companion object {

        @Container
        @JvmStatic
        val postgres = PostgreSQLContainer<Nothing>("postgres:15").apply {
            withDatabaseName("testdb")
            withUsername("test")
            withPassword("test")
        }

        @JvmStatic
        @DynamicPropertySource
        fun registerPgProperties(registry: DynamicPropertyRegistry) {
            registry.add("spring.datasource.url", postgres::getJdbcUrl)
            registry.add("spring.datasource.username", postgres::getUsername)
            registry.add("spring.datasource.password", postgres::getPassword)
        }
    }
}

I have overridden properties for hikari. I have no special test properties either. Any idea?

Hello everyone I'm creating an app in Kotlin with Spring. I configure everything to use Kotlin Serialization instead of Jackson and that works. However, I would like to generate the OpenApi doc. But I can't find a plugin (gradle) or dependencies that support the Kotlin Serialization annotation for my Spring controller's DTO. So two questions for you: • Is there a plugin or dependency that support Kx-serializable • (If not for the first question) Do you recommend me to go back on Jackson due to the lack of support of Kotlin Serializable with Spring Have a good day

Hello! I'm using Spring and I would like to use Ktor as client instead of classic Webflux Everything works well, except to integrate span with Datadog. With Webflux, the integration was native, but in Ktor it seems not to be. Is someone of you know how I can have call traces in datadog with Ktor and Spring please?

This message was deleted.

Hey all 👋 I'm on the hunt for a solid course or training resource that covers Kotlin + Spring Boot + Gradle all together, ideally something that goes beyond the basics. Most of the Spring Boot material out there is very Java + Maven-centric, Kotlin tutorials tend to focus on Android, and Gradle… well, Gradle is its own beast 😅. Anything that does combine all three usually stops at the level of a Spring Initializr-generated project and doesn’t go much deeper. As someone new to all three technologies, I’m finding it tricky to learn them properly without having to jump through multiple layers of abstraction and piece things together from different ecosystems. Is there really no comprehensive resource that teaches how to use all of them in one cohesive way? I’ve searched through the kotlin slack archives and found a few older recommendations (mostly from a year or more ago), but they seem to have the same limitations. Hoping there might be some newer materials or hidden gems I’ve missed 🤞 I'm especially looking for anything that helps set up a project for long-term success; things like structure, best practices, conventions, and how to avoid common pitfalls early on. Whether it's a course, blog post, video, GitHub repo, or even just a well-documented example project, I'd really appreciate any pointers. Thanks in advance 🙏

Hey all, for those using Spring Webflux HATEOAS what you normally do when building a link for a post endpoint that receives a non-nullable entity? Do you create a dummy entity just to call

WebFluxLinkBuilder.linkTo

? Wondering if there is a better approach than that 😛

Hey! Just noticed this recording was finally posted: 🎉

Has anyone seen this error with kotlin reflection in spring boot 4? 🧵

Has anyone had production issues because controllers, DTOs, and everything else for endpoints are generated directly from the OpenAPI spec instead of being created manually (or through an LLM)? Automatic deterministic generation sounds great https://github.com/OpenAPITools/openapi-generator (basically generating source code as a derivative of the OpenAPI .yml file — similar to how JOOQ generates source code as a derivative of the database). But maybe there were some maintenance problems (for example, duplicate DTOs in the generated sources). Some people prefer one approach, others prefer the other.

Are there any articles/guides on best practices when validating input objects in Spring Boot? Let's say we have a data class like this:

data class MyInput(val foo: Foo?, val bar: Bar?)

How do we validate that one of these are non null? What is the best practice? Is it:

requireNotNull(input.foo) { "field foo cannot be null" }

Or is it something like:

val foo = input.foo ?: throw SomeException("...")

What is the most idiomatic way?

I'm testing Spring Boot 4.0.0-RC1 (webflux) and I have problems with basic

ServerRequest.awaitBody<T>()

. It does return an object with correct type but empty (no request data is deserialized). What am I missing?