Kafka Web UI Tool

Purpose & User Intent: This tool provides browser-based Kafka cluster management for developers, operators, and data engineers. It solves the problem of administering topics, consumer groups, and basic data flows without local tooling. Use cases include validating topic configurations, inspecting consumer lag, and provisioning topics in staging or production environments when you need visibility and control across multiple clusters. The intended users are platform teams, SREs, and data engineers who require operational insight and safe topic management in distributed streaming pipelines.
Core Logic & Features: Required features include cluster discovery via bootstrap servers, metadata retrieval for brokers, topics, and consumer groups, and topic lifecycle operations (create, alter, delete) with basic config management. Optional/advanced features include ACL-aware operations, multi-cluster aggregation, per-partition lag analysis, end-to-end data flow testing, and JSON export of results. The tool uses Kafka AdminClient APIs, fetches end offsets, and computes lag as end_offset - committed_offset for each partition, aggregating by topic. It enforces validation against Kafka naming rules and respects cluster RBAC when credentials are provided.
Inputs & Outputs: Inputs include bootstrap_servers (string), security_protocol (string), sasl_mechanisms (string, optional), sasl_username (string), sasl_password (string, optional), topic_name (string), partitions (integer), replication_factor (integer), operation (string), config_overrides (key/value map). Outputs include topics list, consumer_groups, lag_metrics, topic_configs, and exportable JSON. Outputs are presented as structured data per topic and per group, with numeric fields in standard units (offsets, lag, throughput).
Algorithms & Calculations: Uses AdminClient to gather metadata; computes lag per partition as high_watermark - committed_offset; aggregates to topic level with max lag and average lag; uses end_offsets() per partition; applies small sample producer/consumer to validate flow when requested.
Error & Edge Cases: Handles invalid bootstrap_servers, authentication failures, missing topic names, unsupported operations, insufficient permissions; handles cluster version differences; gracefully reports errors in a structured way without side effects.
Industry/Region & Localization: Units are counts and offsets; times in ms; supports clusters across regions; timestamps use standard formats; no currency handling; localization focuses on number formatting and time zones where supported.