Releases: AstorisTheBrave/argus
Releases · AstorisTheBrave/argus
v0.4.2
0.4.2 (2026-06-21)
Features
- config: eager .env bootstrap and host-port auto-detect (#54) (b6c57cf)
- core: fail-open metrics server, scrape isolation, and self-health (#43) (cb421b6)
- dashboard: harden the in-process bot server to match the fleet plane (#44) (b96a838)
- dashboard: optional bearer auth for the /metrics endpoint (#56) (c1ad15c)
- fleet: accept multiple tokens for zero-downtime rotation (#40) (eb99c92)
- fleet: per-identity lease secrets (un-shareable, rotatable identity) (#41) (92b4863)
- fleet: read-side rate limit and optional audit log (#42) (db2f5d1)
- history: drop counter, resilient sink worker, idempotent Argus(bot) (#45) (a2efd36)
- history: sink circuit breaker drives live argus_subsystem_up{sink} (#51) (7d5b2d2)
- logging: opt-in JSON logging for the bot side (#52) (692d903)
- observability: optional Prometheus Pushgateway push (no inbound port) (#58) (4819152)
- observability: ship Prometheus alert/recording rules and a health dashboard (#46) (9b9367c)
- observability: SLO burn-rate alerts and a golden-signals dashboard row (#53) (b9ec60c)
Bug Fixes
- fleet: default member identity to cluster_id; fix copy-paste quickstart (#38) (791c587)
- fleet: write state and identity files owner-only (0600) (#39) (89f9fea)
Documentation
- clarify cluster_id is always distinct, ports only differ when co-located (#36) (2d5d040)
- hosting: recipes for Pterodactyl, PebbleHost, Ori, and Railway (#55) (b90503d)
- lead with push (no inbound port) for Docker bot hosts (#57) (af21611)
- readme: add a How Argus works section with an architecture diagram (#49) (36c3dbf)
v0.4.1
v0.4.0
0.4.0 (2026-06-21)
Features
- core: time prefix (text) command duration (#28) (3c0a4e0)
- fleet: cluster drill-down trend history with sparklines (#25) (23ae1a5)
- fleet: flap detection, source-failure resilience, fuzz/load tests + tutorials (#21) (c680e4d)
- fleet: graceful shutdown, JSON logging, image hardening, k8s example (#26) (153ad01)
- fleet: hardening slice 1 - secure defaults, efficiency, reliability (#16) (6f72a06)
- fleet: hardening slice 2 - setup wizard, doctor, DX (#17) (45bfb6c)
- fleet: hardening slice 4 - rate limiting, retention, single-writer lock (#19) (33d2ad7)
- fleet: hardening slice 5 - Prometheus http_sd auto-discovery (#20) (312ce8b)
- fleet: split ingest and viewer tokens (#18) (b3c726e)
- fleet: standalone Argus Fleet control plane (#14) (79f4c2a)
- fleet: trusted-proxy client IP, heartbeat jitter, view-build latency metric (#24) (523dc7d)
Bug Fixes
v0.3.1
v0.3.0
v0.2.1
v0.2.0
0.2.0 (2026-06-17)
Features
- adapters: add Prometheus adapter with hybrid registry (invariant 4) (efea3f0)
- add ArgusCog and the one-line Argus(bot) integration (43cb69a)
- config: add ArgusConfig with kwargs>env>defaults precedence (f71f721)
- config: add dashboard/history settings (8966a8f)
- core: add hooks + fail-open instrumentation (invariants 3, 5) (5e47799)
- core: add neutral MetricRegistry and metric model (b66fac2)
- core: define the v1 metric catalogue (spec sec.8) (eb34c4c)
- dashboard: add bearer-token auth middleware (f9ddbdd)
- dashboard: analytics API (fail-closed) + ClickHouse wiring (c72a3c8)
- dashboard: JSON metric snapshot (913c840)
- dashboard: serve SPA shell + /api/config (87128d3)
- dashboard: SSE metric stream at /api/stream (ca72b9f)
- dashboard: wire dashboard into ArgusCog (8daa7b6)
- exposition: aiohttp /metrics + /healthz on the bot loop (7dc8e98)
- frontend: metrics data layer (5968f28)
- frontend: scaffold Vite+React+TS SPA with Nimble assets (43e8dac)
- frontend: sidebar hub with overview, interactions, gateway, grafana, analytics (a3968d2)
- history: ClickHouse sink + analytics queries (3ab9aca)
- history: EventSink ABC, NullSink and non-blocking BatchingSink (a3b6e99)
- history: per-guild event capture behind enable_per_guild (0f56009)
- stack: add Grafana dashboards, provisioning, compose, scrape config (3be5320)
Bug Fixes
- clear CodeQL quality alerts (e5fbe4e)