Fluent Bit V2.0: Unifying Open Standards For Logs, Metrics & Traces
summary
Fluent Bit 2.0 claims unified support for logs, metrics and traces and demonstrates it by sending OpenTelemetry spans to Jaeger, Prometheus metrics to Grafana via Prometheus remote_write, and showing a live tap live sampling HTTP endpoint that returns real payload samples for debugging. The release adds threaded inputs/outputs for multi‑core scaling, TLS input support, a storage engine with exposed internal metrics, and WASM and Go plugins for custom filters and inputs; demoed features run successfully in the presented Docker demo.
AI-written summary · about
chapters (18)
- 0:00Fluent Bit V2.0 intro
- 1:02Fluent Bit in CNCF
- 3:12Vendor lock-in risks
- 4:39Why Fluent Bit replaced Fluentd
- 6:48Deployment options
- 8:57Telemetry and TLS in V2.0
- 11:29Unified observability
- 13:35Performance improvements
- 14:39Developer experience
- 16:09Live demo: Architecture
- 21:57Q&A: Migration
- 23:25Q&A: Fluent Bit vs Prometheus
- 25:54Q&A: Plugin support
- 27:28Q&A: Node exporter parity
- 28:54Q&A: OpenTelemetry metrics
- 30:28Q&A: Kafka input
- 31:52Q&A: Tap security
- 32:12Closing remarks
related talks
36mFluent Bit: Logs, OpenMetrics, and OpenTelemetry all-in-one
05mFluentd and Fluent Bit Project Updates
14mFluent Bit: Past, Present, and Future
28mBuilding Observability Pipeline with Fluent-Bit
25mConnecting with Prometheus World: Logs and Metrics
13mFluent Bit
25mOpening Remarks + Project Updates
34mFluentd and Fluent Bit