A powerful logs management solution we all have and use, but we underestimate
summary
Claims that systemd journal is an underrated, production-ready log database: it indexes all fields and values with append-only binary files, offers forward-secure sealing, and recovers corrupted chains; libsystemd lets queries open journal files directly so no central server is required. Netdata improves journalctl performance by ~14×, built log2journal runs anywhere to structure and ingest nginx/HAProxy logs, and provides remote upload/remote replication. Demo scales to ~15 million entries, samples up to 1,000,000 raw rows, and uses only a few megabytes of RAM.
AI-written summary · about
chapters (8)
- 0:00Netdata and systemd journal
- 6:43systemd journal underrated
- 13:27systemd journal as database
- 17:39Netdata for journal improvement
- 24:04Logs management pipeline
- 25:50Key journal fields
- 26:55Visualizing logs at scale
- 28:38Q&A retention and replication
related talks
20mA Powerful Logs Management Solution We All Have and Use but...
33mNetdata: Open Source, Distributed Observability Pipeline – Journey and Challenges
40mLarge-Scale Logging Made Easy
28mFrom Fragmented Logs to Unified Insights: Scaling O11y
38mLogging in the age of Microservices and the Cloud
29mCrucial Log Management Skills for DevSecOps
39mFast log management for your infrastructure
40mTaming Spiky Log Volumes: Maintaining Real-Time Log Accessibility with Kaldb