The Apache Software Foundation has released Apache Kafka 4.2!

Apache Kafka® 4.2 brings production-ready Kafka Queues, Kafka Streams server-side rebalancing now generally available(GA), dead letter queue support for Kafka Streams, and important standardisation work across CLI tools and metrics. The release contains 38 Kafka Improvement Proposals (KIPs) from 155 community contributors, representing a significant step forward for Kafka operations, Streams, and Connect.

While there's a lot to unpack in Apache Kafka 4.2, here's a summary of some of the key highlights.

Highlights

• KIP-932 GA: Queues for Kafka
  • Kafka Queues are now production-ready, bringing queue-like consumption patterns to Kafka through share groups. This release adds the RENEW acknowledgement type for extended processing (KIP-1222), adaptive batching for coordinators (KIP-1224), and comprehensive lag metrics (KIP-1226) for monitoring consumption progress. Queuing capabilities decouple consumer scaling from topic partitions, enabling cooperative message consumption with individual acknowledgments and retries.
• KIP-1071 GA: Streams Server-Side Rebalance Protocol
  • The server-side group management protocol for Kafka Streams reaches GA. Built on the innovation from KIP-848, this broker-driven rebalancing system is specifically designed for Kafka Streams applications, providing streams-optimised coordination separate from the traditional consumer group protocol.
• KIP-1034: Dead Letter Queue Support for Streams**
  • Kafka Streams now includes dead letter queue support in exception handlers, allowing applications to handle processing errors more gracefully and route problematic records to separate topics for analysis and debugging.
• KIP-848: Continued Improvements**
  • Enhanced reliability through client-side performance improvements and adaptive coordinator linger times, continuing the journey toward making KIP-848 the default rebalance protocol.
• Standardization & Improvements
  • Enhanced usability through CLI tool standardization with consistent arguments such as --bootstrap-server (KIP-1147), metric naming corrections to follow the kafka.COMPONENT convention (KIP-1100), and anchored wall-clock punctuation (KIP-1146) for callbacks at fixed, deterministic times.
• Security
  • KIP-1188: New allowlist connector client configuration override policy addresses vulnerabilities by letting administrators specify which configurations connectors can override.

Client-Side KIPs Available Now in Confluent Cloud
The following client-side KIPs are immediately available for use with Confluent Cloud:

• KIP-1120: Add Client-ID to AppInfo Metrics
• KIP-1136: Make ConsumerGroupMetadata an Interface
• KIP-1161: Unifying LIST-Type Configuration Validation and Default Values
• KIP-1175: Fix Typo in Producer Config
• KIP-1205: Improve RecordHeader to be Thread-Safe

Please read the release announcement from the Apache Software Foundation and watch the release video to learn about all of the innovation included in Apache Kafka 4.2!

What’s New in Confluent Cloud: All the Announcements from Q1 2026

                                                                          Q1 2026Confluent Cloud Launch

Confluent Cloud’s Q1 2026 launch is now live! Check out all the highlights below to see how Confluent is making data streaming easier for any workload, at any scale:

Confluent Intelligence is Confluent’s vision to bring real-time data directly to production AI systems through the power of Apache Kafka and Apache Flink®. Q1 2026 introduces new features including:

• Agent2Agent (A2A) Integration for Streaming Agents (Open Preview)
• Multivariate Anomaly Detection for Built-in ML Functions (Early Access)
• Vector Search for Azure Cosmos DB and Amazon S3 Vectors
• AWS and Azure Private Link for Model Inference, External Tables, and Search
• Confluent Support for Open Source MCP Server

Queues for Kafka KIP-932, (available on Confluent Cloud for Enterprise and Dedicated clusters introduces “share groups” for dynamic scaling and unlocks new job and work queue use cases for Kafka.

Check out the blogs and demo videos to learn more:

• Q1 ‘26 Confluent Cloud launch: Blog
• Q1 ‘26 Confluent Intelligence updates: Blog, Multivariate Anomaly Detection and A2A Demo
• Kafka Copy Paste (KCP): Blog, Demo, GitHub
• Queues for Kafka: Blog, Demo, GitHub
• New scalability and cost savings improvements for serverless clusters: Video

Data Streaming Resources

• Queues for Kafka is now generally available. Dive deep into the world of share groups and share consumers and understand how these work cooperatively, with this informative blog by Andrew Schofield, Principal Engineer, Confluent.
• Read this blog by David Araujo, Director of Product Management for Stream Governance at Confluent on how introducing Schema IDs in Apache Kafka® headers makes it easy to govern, trust, and reuse Kafka data.

Links From Around the Web:

• Stack Overflow published the 2025 Developer Survey. As always, this annual developer survey provides a fascinating glimpse into the needs of the global developer community.
• Learn about this interesting experiment in conducting a vector similarity search on a Parquet file using Apache DataFusion. This small experiment, GitHub repo, leverages the finer details of the Parquet specification to implement an IVF(Inverted File Index) based vector search and demonstrates the power of Rust!

In-Person Meetups

• Mar 26 - Milan with AWS
• Mar 28 - Ahmedabad, Gujarat
• Mar 28 - Bangalore
• Mar 31 - Kuala Lumpur
• Apr 7 - Silicon Valley with Elastic
• Apr 8 - Jakarta
• Apr 9 - SF Bay Area Java Group
• Apr 9 - Melbourne
• Apr 9 - SF Bay Area
• Apr 10 - Singapore
• Apr 16 - Amsterdam Hema
• May 7 - Belgium with Cymo
• May 13 -Chicago with Elastic
• May 14 - Erlangen/Nurnberg with Siemens Healthineers

By the way…

We hope you enjoyed our curated assortment of resources! If you’d like to provide feedback, suggest ideas for content you’d like to see, or you want to submit your own resource for consideration, send us an email at devx_newsletter@confluent.io!

If you’d like to view previous editions of the newsletter, visit our archive.

If you’re viewing this newsletter online, know that we appreciate your readership and that you can get this newsletter delivered directly to your inbox by filling out the signup form on the left-hand side.

P.S. If you want to learn more about Kafka, Flink, or Confluent Cloud, visit our developer site at Confluent Developer.