A powerful streaming log template miner based on the Drain algorithm in golang
-
Updated
Oct 23, 2024 - Go
A powerful streaming log template miner based on the Drain algorithm in golang
go-drain3 is a golang port of drain3, which is log parsing library written in python.
🚀 An intelligent, LLM-enhanced log parser pipeline that converts multi-format raw logs into structured JSON, learns from missed patterns, and evolves using Drain3 & open-source LLMs.
AI-driven log anomaly detector: Drain3 + Isolation Forest / autoencoder + ELK stack, with one-command demo.
ASDIP — AI Secure Data Intelligence Platform. Miniature SIEM with a 7-stage detection pipeline, Drain3 log clustering, stateful brute-force correlation, secret masking and live SSE risk scoring.
Security log triage CLI — deterministic detection + LLM explanations in Brazilian Portuguese. Scales to multi-GB logs, LLM used only for the explanation layer.
Original AIOps prototype for streaming log parsing, immutable model versioning, hot rollback, and anomaly detection.
Find the most likely root cause in a Linux log, with the exact log lines as proof. Drain3 + Chroma + local LLM + plain-code evidence check.
Hybrid log parser (Drain3 + LLM) for production log analysis
Universal log pre-processing framework: ECS normalisation, ML enrichment, Merkle integrity, SIEM forwarding
A production-ready TypeScript/Node.js streaming log template miner — 1:1 port of the official Python Drain3 algorithm. Extracts structured templates from unstructured logs in real time using fixed-depth prefix tree clustering with zero runtime dependencies.
One triage pipeline for logs, GitHub issues, CI failures and status incidents: deterministic clustering, one LLM verdict per cluster, a rule table decides.
Reduce GB-scale logs to a representative sample - streaming Python library and CLI with pattern mining and anomaly detection
To associate your repository with the drain3 topic, visit your repo's landing page and select "manage topics."