Fluentd Deep Dive
summary
Fluent Bit gains pod-level parsing by honoring the fluentbit.io/parser annotation so a DaemonSet reads /var/log/containers, queries the API for pod annotations, and applies a named parser from parsers.conf regex backend (currently Apache example). Parsers live in config, not per-pod regexes, to avoid compile costs and cluster-wide perf hits; Fluentd will get the feature later. Use filesystem buffering, the DB file to resume positions after restarts, JEMalloc to reduce fragmentation, and note ~700 Fluentd vs ~45 Fluent Bit plugins.
AI-written summary · about
chapters (20)
- 0:00Introduction
- 3:10Fluentd Configuration
- 4:40Fluentd Core
- 5:20Pod Apache
- 6:40Regular Expressions
- 10:45Fluentd
- 12:10Flexibility
- 13:05Elastic Search
- 15:40Documentation
- 17:15Own Killer
- 18:15Wrong approach
- 19:30Memory usage
- 21:45Members limit
- 24:25Reports
- 25:35Limit Data Traffic
- 27:05Matrix in Fluentd
- 28:55Why Fluentd
- 30:45Fluentbed
- 35:35SSH Parser
- 37:15SSH Regex
