# linkedin/databus

Source-agnostic distributed change data capture system

Repository: https://github.com/linkedin/databus
Canonical: https://ross.abutalabs.com/products/databus
Language: Java
License: Apache-2.0
License Family: permissive
Last push: 2023-09-28T17:53:01+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 5007, "days_push": 1070, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 3679, forks 737 (observed 2026-08-28T04:08:13.724433+00:00)

## What it is
Databus is LinkedIn's source-agnostic distributed change data capture (CDC) system that reliably captures and streams primary data store changes to derived stores, caches, and indexes. It provides low-millisecond latency transport with thousands of events per second per server, infinite look-back, and rich subscription functionality.

## Use cases
- capture database changes and stream them to search indexes
- invalidate cache entries when primary data changes
- replicate OLTP database changes to derived data stores
- replace dual-write patterns with a reliable change pipeline
- build a scalable event pipeline from transaction logs
- feed downstream data processing systems with primary store mutations

## When to choose
- you need reliable change data capture from primary databases into derived stores
- you want to avoid consistency problems of application-driven dual writes
- you need low-latency event transport with infinite look-back and subscriptions
- you run a JVM-based data infrastructure at large scale

## When to avoid
- you need a modern actively developed CDC tool with broad database connector support
- your stack is not JVM-based
- you want a simpler single-node change capture solution
- you need turnkey support for databases beyond Oracle and MySQL

## Facets
- artifact type: library
- maturity: maintenance
- function: streaming, etl, message-queue
- domain: databases, microservices, big-data
- platform: jvm, cross-platform
- tags: change-data-capture, cdc, data-pipeline, event-streaming, linkedin, log-mining, replication, data-engineering, linux

## Member repositories
- linkedin/databus (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:13.724433+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T18:31:14.067218+00:00, confidence not recorded.
  - readme: https://github.com/linkedin/databus (fetched 2026-08-28T04:08:13.724433+00:00, sha 2ca928ec457d)
- Data as of 2026-08-30T08:39:29.467469+00:00.