Pharlux documentation
This is the documentation for Pharlux V1.0.0 — a single statically-linked Rust binary delivering unified OpenTelemetry-native observability for small teams. Mirrored from the pharlux/docs/ directory in the main repo. The repo is the source of truth; this site is regenerated when a new version ships.
Operator docs
Day-to-day operating reference for running Pharlux in production.
- Getting started — install, configure, ingest, query
- Sizing guide — memory + disk accounting, the 10 GB/day FTS threshold, tuning knobs
- OTLP configuration reference — gRPC :4317 + HTTP/protobuf :4318
- SQL query reference — DataFusion dialect, cross-signal JOINs, examples
- Logs query performance — LIKE thresholds, the V1.1 tantivy plan
- Dashboards — the embedded React+ECharts surface, layout JSON, CRUD endpoints
- Alerts — SQL-rule lifecycle, state machine, circuit-breaker semantics, webhook + Slack payloads
- Auth — JWT + Argon2id, user / API-key management, secret rotation
- Reverse proxy and TLS — Caddy and nginx setups in front of Pharlux
- Backup and restore — the
pharlux backupflow and crash-safe restore - Upgrade — V1.0.x patches and the V1.0 → V1.1 path
- Troubleshooting — symptom-driven triage with pointers into the operator runbook
Developer reference
For contributors working on the Pharlux source tree itself.
- Architecture — runtime layout, data flow, crate boundaries
- Crate map — the workspace at a glance: what each crate does and where its tests live
- Testing — the four required commands, crash-recovery tier, proptest invariants
Beyond this site
- Operator runbook — numbered procedures referenced by the troubleshooting page
- SECURITY.md — threat model, trust boundaries, coordinated-disclosure process
- CHANGELOG.md — release notes by version
- GitHub repository — source, issues, releases, discussions