Kafka Monthly Digest – October 2020
See what's going on in the Kafka community this month
In this 33rd edition of the Kafka Monthly Digest, I’ll cover what happened in the Apache Kafka community in October 2020.
For last month’s digest, see Kafka Monthly Digest: September 2020.
2.7.0: This release entered code freeze on October 21. Now only blocking issues or regressions should be addressed. There are currently 2 blockers pending fixes. The release is still expected for November. All the details are in the release plan.
2.6.1: On October 1, I volunteered to run the 2.6.1 bug-fix release. The release plan is available on the wiki. It is expected to release in November.
Last month, the community submitted 5 KIPs (KIP-675 to KIP-679), and these are the ones that caught my eye.
KIP-676: Respect logging hierarchy. Since Kafka 2.4.0, a brokers’ log levels can be altered and retrieved via an API. However, the way the log hierarchy is handled by Kafka is different from log4j. So, for many users, this creates unexpected behaviors. This KIP aims to address this inconsistency and switch to log4j’s logging hierarchy.
KIP-679: Producer will enable the strongest delivery guarantee by default. This KIP proposes changing the Producer default configuration to provide stronger delivery semantics. This change would set
alland enable idempotency to provide “exactly once” semantics by default.
In this section, I will cover releases of some community projects. This only includes projects that are Open Source.
Debezium 1.3. This new version has been updated to use Kafka 2.6. Out of the many changes, there is a new connector (incubating) for Oracle and significant updates for the SQL Server and MongoDB connectors.
Strimzi 0.20. Strimzi is a Kafka Operator for Kubernetes. Strimzi now supports Kafka 2.5.1 and 2.6.0. This new release includes dozens of fixes and improvements. Kafka connects to ZooKeeper via native TLS, so TLS sidecar containers have been removed. This version also adds support for dynamic logging configuration for brokers, Kafka Connect, and MirrorMaker 2. Finally, the
listenersconfiguration has been redesigned to enable more listeners in clusters.
Get started with Kafka
IBM Event Streams for Cloud is Apache Kafka-as-a-Service for IBM Cloud. Get started with IBM Event Streams today.