This 67th edition of the Kafka Monthly Digest covers what happened in the Apache Kafka community in August 2023.
For last month’s digest, see Kafka Monthly Digest: July 2023.
Releases
There is one new release in progress: 3.6.0.
3.6.0
The release process for 3.6.0 continued. Feature freeze happened on August 16 and the 3.6 branch was cut on August 21. Then code freeze was on August 31. We are now in the stabilization period before the first release candidate. You can find all the details in the release plan.
Kafka Improvement Proposals
Last month, the community submitted 12 KIPs (KIP-963 to KIP-974). I'll highlight a few of them:
- KIP-965: Support disaster recovery between clusters by MirrorMaker: When mirroring ACLs between clusters, MirrorMaker downgrades allow
ALL
ACLs to allowREAD
ACLs. The reason is that remote topics should only be written to by MirrorMaker itself. However this creates issues in disaster recovery environments where we want both clusters to be identical. This KIP's goal is to remove this limitation and allow MirrorMaker to exactly mirror ACLs between clusters. - KIP-966: Eligible Leader Replicas: This KIP aims at solving a data loss scenario that can happen when there's only a single in-sync replica left and it experiences an unclean shutdown that causes it to lose some committed data. In this case this replica will be re-elected the leader if it restarts and incorrectly cause followers to truncate their logs thus causing data loss.
- KIP-968: Support single-key_multi-timestamp interactive queries (IQv2) for versioned state stores and KIP-969: Support range interactive queries (IQv2) for versioned state stores: Last month, we covered KIP-960 aiming to add support for interactive queries for versioned state stores. This month two additional KIPs propose further improvements to support more complex queries with key and timestamp ranges.
Community Releases
- Sarama 1.41: Sarama is a pure Golang Kafka client. This version corrects how Sarama uses Kafka protocol versions based on the given version in the client configuration. For that reason it may cause breaking changes with old (< 2.1) Kafka clusters.
Blogs
I selected some interesting blog articles that were published last month:
To learn more about Kafka, visit Red Hat Developer's Apache Kafka topic page.