Logging for Containers
summary
Fluentd and Fluent Bit are presented as complementary pieces: Fluentd handles aggregation, parsing and buffering with 600+ plugins and wide production use (Microsoft OMS, CNCF adoption), while Fluent Bit is a C-based, low-overhead forwarder and library that hits 500,000 messages per second on a single core and is used to push millions of events per second at Treasure Data. Docker adds JSON metadata (container ID, name, timestamp) and Kubernetes runs Fluentd on each node reading log files; Fluent Bit supports library mode and Go plugins via shared libraries.
AI-written summary · about
chapters (9)
- 0:00Start
- 4:38Standard Input and Output Interfaces
- 11:22Distributed Logging
- 13:01Items To Consider
- 13:49Source Aggregation Patterns
- 15:19Destination Aggregation
- 19:15Output Plugins
- 25:19Architecture
- 41:06Gulp Plugins
related talks
22mPetabyte Scale Logging with Fluentd and Fluent Bit: A Use Case From...
34mFluentd: A Complete Logging Ecosystem for Kubernetes
27mUnder the Hood with Fluent Bit Operator: Kubernetes-native Log Processor
33mFluent Bit V2.0: Unifying Open Standards For Logs, Metrics & Traces
09mLightning Talk: Fluentd at Scale; Keys to successful Logging Syndication
25mConnecting with Prometheus World: Logs and Metrics
40mLive, Laugh, Log
31mLogging at Scale, Billion+ messages, 100K msg/sec