// the find
provectus/kafka-ui
Open-Source Web UI for Apache Kafka Management
A web UI for managing Apache Kafka clusters — browse topics, messages, consumer groups, brokers, and schema registries from a browser instead of CLI tools. Targets teams running Kafka who want visibility without dropping into kafka-topics.sh every time. Last pushed July 2024, and the repo shows signs of slowing activity.
Multi-cluster support in a single install is genuinely useful — you're not running a separate UI per environment. The custom serde plugin system is well thought out: you can plug in Avro, Protobuf, AWS Glue, or write your own without forking the project. RBAC with OAuth2 (GitHub/GitLab/Google) covers the 'we need to let the team browse without giving everyone cluster admin' case. The Docker image with DYNAMIC_CONFIG_ENABLED lets you add clusters at runtime without restarting — that alone saves headaches in staging environments.
Last commit was July 2024 and the project has been quietly transferred to a community fork (ui-for-apache-kafka) — the Provectus-branded repo is effectively unmaintained, which is a meaningful risk before you build operational tooling around it. The reactive Spring WebFlux stack is a reasonable choice but makes the Java code harder to contribute to and debug than plain Spring MVC would be. No native KRaft/KIP-500 monitoring story beyond what the broker APIs expose — if you're running Kafka 3.x in KRaft mode you lose some visibility. Message browsing for high-throughput topics with large payloads can be slow; there's no sampling or offset-based pagination that works reliably across all encoding types.