- SignalDesk3天前
For a long time, I’ve used Kafka for services for some purpose. like tracing user action log, event sourcing propagation pipeline step processing and so on.<p>Whenever I use Kafka, I need to set up kafka ecosystem such as, ELK ( Elasticsearch + Logstash + Kibana) for search Kafka messages for service operation, issue trace. and Kafka sink connector for some data forward other persistent DB, and ksqlDB( or Flink ) for count and monitor event stream. With additional infra and software<p>One day, a thought suddenly crossed my mind.<p>That is “Basically Kafka is processing message stream sequentially, with a consumer.<p>It means If I don’t know what the partition and offset of “needle” is, I have to find needle sequentially from beginning. but if I know Kafka message partition and offset, Can I find and access data directly like other DBMS?”<p>ELK is not necessary on my service at that time. Just I need to search trace events and it is in Kafka topic message. All I need to know is where the data I want to find is<p>My desktop and your desktop are too powerful, expensive to use only for web browser operation and AI question agent.<p>Let’s use local disk and local application for indexing storage and search engine about Kafka message.<p>This local-first concept is implemented as “Kaflow Search desktop tool”<p>If you don’t need to keep data for a long time or permanently You don’t have to go through the hassle of installing and managing ELK or other Kafka ecosystem infrastructure<p>Just keep and sync index with Kafka cluster server And find messages you want easily with just this tool installed It will save your operation time.<p>When operating company services, The costs of installing server, managing additional ELK/Kafka Connect, ksqlDB instances didn’t come out of my own pocket<p>so I felt free to spend my time and server resources to set additional data pipeline up.<p>but if it is my own service that I afford, I can’t help saving my server cost. I can’t afford the additional infra costs from cloud service (like AWS)<p>I’m not a rich developer, who just has a laptop to develop software. I’ll make the most of my laptop resources, until my service generates enough money, and then I can handle additional server costs. I’ll only use Kafka and this tool on my service, before I really need ELK and other Kafka ecosystem.<p>I dedicate this rich client Kafka tool to developers who, like me, are poor.<p>GitHub : <a href="https://github.com/whsoul/kaflow-search" rel="nofollow">https://github.com/whsoul/kaflow-search</a><p><If you have kafka cluster><p>=> Download From release assets ( macOS, Windows ) - <a href="https://github.com/whsoul/kaflow-search/releases/tag/v0.1.3" rel="nofollow">https://github.com/whsoul/kaflow-search/releases/tag/v0.1.3</a><p><pre><code> From brew ( macOS ) - brew install --cask whsoul/tap/kaflow-search </code></pre> <If you don’t have kafka cluster><p>=> Download demo From release assets ( macOS, Windows ) - <a href="https://github.com/whsoul/kaflow-search/releases/tag/demo-v0.1.3" rel="nofollow">https://github.com/whsoul/kaflow-search/releases/tag/demo-v0...</a> From brew ( macOS ) - brew install --cask whsoul/tap/kaflow-search-demo<p>Please try my tool ( on your macOS, Windows ) and feedback for me consumer ID, consumer Group do not remain on your cluster. It does only read operation, not managing function ( such as delete, create topic, repartition cluster topic). It normally only reads once messages that you select, and read local-first index for search your message. So feel free to install this.<p>But it is not developer signed.. (Sorry because I’m poor developer..) Please follow this install guide ( <a href="https://github.com/whsoul/kaflow-search#install" rel="nofollow">https://github.com/whsoul/kaflow-search#install</a> ) about untrusted developer warning when install and first launch.
- 情报分类：开源项目与落地
- 分类依据：Kafka本地检索桌面工具，开源项目
- 信息来源：Hacker News 新项目
- 发布时间：2026/9/16 21:08:47
- 暂无回复