34 Commits

Author SHA1 Message Date
jkunz 4af9d3de69 test(ui): cover degraded overview health semantics
CI / Type Check & Lint (push) Successful in 6s
CI / Build Test (Current Platform) (push) Successful in 6s
CI / Build All Platforms (push) Successful in 38s
2026-04-21 13:36:53 +00:00
jkunz 405fff91af refactor(health): share health snapshot computation 2026-04-21 13:36:07 +00:00
jkunz 9022c8dbf3 feat(health): include degraded reasons in responses 2026-04-21 13:34:58 +00:00
jkunz 703cceb512 test(config): cover config init and show output 2026-04-21 13:33:38 +00:00
jkunz 9d925f9401 feat(api): assign and echo request ids 2026-04-21 13:32:50 +00:00
jkunz fe4fdb32d7 fix(api): map upstream timeouts to 504 responses 2026-04-21 13:32:12 +00:00
jkunz d6b4c0def1 refactor(api): inject server dependencies in tests 2026-04-21 13:30:52 +00:00
jkunz 9608540792 feat(api): enforce per-minute request rate limits 2026-04-21 13:30:20 +00:00
jkunz 3762fc661e feat(metrics): count api requests auth failures and 5xxs
CI / Type Check & Lint (push) Successful in 6s
CI / Build Test (Current Platform) (push) Successful in 6s
CI / Build All Platforms (push) Successful in 39s
2026-04-21 13:15:34 +00:00
jkunz 6541b2db1c docs(testing): document regression and seam test commands 2026-04-21 13:13:51 +00:00
jkunz da7375c889 refactor(timeout): reuse shared api request timeout 2026-04-21 13:13:13 +00:00
jkunz 44eb9b9173 test(cluster): cover persistence initialization and pruning 2026-04-21 13:12:01 +00:00
jkunz 1f24df0d80 fix(config): warn on ignored legacy config keys 2026-04-21 13:10:55 +00:00
jkunz c95961d596 refactor(api): allow router handler injection in tests 2026-04-21 13:10:00 +00:00
jkunz 0921dfbe5e test(api): cover health metrics and auth flows 2026-04-21 13:09:29 +00:00
jkunz 5172002ec0 fix(api): reject oversized request bodies with 413 2026-04-21 13:08:47 +00:00
jkunz 58eabba84d refactor(runtime): let daemon startup failures bubble up
CI / Type Check & Lint (push) Successful in 6s
CI / Build Test (Current Platform) (push) Successful in 6s
CI / Build All Platforms (push) Successful in 40s
2026-04-21 12:46:24 +00:00
jkunz 5e8ce6690d docs(readme): remove stale runtime references 2026-04-21 12:45:52 +00:00
jkunz 0ea98caed6 refactor(config): extract config loading from modelgrid 2026-04-21 12:45:16 +00:00
jkunz 871afedbb7 test(behavior): cover config routing and registry seams 2026-04-21 12:44:11 +00:00
jkunz 1f6cf51794 build(package): wire scripts to the real deno tasks 2026-04-21 12:42:07 +00:00
jkunz 054875abb5 refactor(config): remove legacy model config aliases 2026-04-21 12:41:49 +00:00
jkunz 3e341bbfda docs(config): sync hints with current config schema 2026-04-21 12:40:45 +00:00
jkunz 9f7308498c fix(cluster): skip persistence scheduling until initialize has run
schedulePersist and scheduleControlPersist can fire from configure() and
the public scheduling paths before initialize() has completed. Without a
guard, those queued microtasks call persistState/persistControlState,
which try to mkdir PATHS.DATA_DIR and write state files from tests and
short-lived scripts that never meant to touch the data directory. That
produced async-leak warnings in the Cluster manager unit tests and
left orphan directories on hosts that only constructed a ClusterManager
to inspect it.

Add an `initialized` flag set at the end of initialize() and early-return
from both schedulers when it is false. Real runtime paths always call
initialize() during Daemon startup, so this changes no production
behavior.
2026-04-21 12:39:50 +00:00
jkunz 952bf394d3 fix(cluster): avoid hostname lookup during construction 2026-04-21 12:37:12 +00:00
jkunz 3b2a16b151 feat(ui): add browser console served by the daemon
CI / Type Check & Lint (push) Successful in 8s
CI / Build Test (Current Platform) (push) Successful in 9s
CI / Build All Platforms (push) Successful in 39s
Introduce a minimal operations console reachable on a dedicated UI port
(default 8081), kept separate from the OpenAI-compatible API port.

- ts_web/ holds the SPA shell (index.html, app.css, vanilla app.js) with
  sidebar navigation for all views from readme.ui.md and a working
  Overview page backed by a new /_ui/overview JSON endpoint.
- scripts/bundle-ui.ts walks ts_web/ and emits ts_bundled/bundle.ts, a
  single generated module exporting every asset as base64. Mirrors the
  @stack.gallery/registry pattern so deno compile binaries embed the
  entire UI with no external filesystem dependency at runtime.
- ts/ui/server.ts (UiServer) serves assets from either the bundled map
  (default, prod) or directly from ts_web/ on disk (dev). The source is
  chosen per-config and can be overridden by UI_ASSET_SOURCE=disk|bundle.
  SPA fallback routes unknown extensionless paths to index.html.
- IModelGridConfig.ui block with enabled/port/host/assetSource defaults;
  config init writes the block, the normalizer fills in defaults on
  load, and the daemon starts/stops the UI server alongside the API.
- deno.json gains a bundle:ui task; compile:all now depends on it so
  released binaries always contain an up-to-date bundle. dev task sets
  UI_ASSET_SOURCE=disk for hot edits.
- ts_bundled/ is gitignored (generated on build).
- test/ui-server.smoke.ts exercises bundle and disk modes end to end
  (index, app.js, SPA fallback, /_ui/overview, 404).
2026-04-21 10:01:44 +00:00
jkunz 9c9c0c90ae docs(readme): ship UI via typedserver + bundled ts module
CI / Type Check & Lint (push) Successful in 6s
CI / Build Test (Current Platform) (push) Successful in 6s
CI / Build All Platforms (push) Successful in 41s
Rework the implementation-notes section of readme.ui.md so the UI
delivery story matches ModelGrid's deno-compile single-binary shape.
Adopt the @stack.gallery/registry pattern: a build step bundles ts_web/
into a generated ts_bundled/bundle.ts exporting a path->bytes map that
typedserver serves at runtime. Add a dev-vs-prod asset-source switch so
UI edits stay hot-reloadable during development while release builds
embed the whole console in the binary.
2026-04-21 09:38:24 +00:00
jkunz 24bb6b3058 docs(readme): add UI concept document
CI / Type Check & Lint (push) Successful in 6s
CI / Build Test (Current Platform) (push) Successful in 6s
CI / Build All Platforms (push) Successful in 40s
Sketch a browser-based operations console for ModelGrid, served by the
daemon on a dedicated UI port. Lays out top-level IA (overview, cluster,
gpus, deployments, models, access, logs, metrics, settings), per-view
content, key user journeys, realtime/auth expectations, and open
questions. Structural idioms adapted from the dcrouter Ops dashboard and
grounded in ModelGrid's own cluster/catalog/vLLM domain.
2026-04-21 09:29:30 +00:00
jkunz cec102e54e docs(readme): fix vLLM config example fence to jsonc
CI / Type Check & Lint (push) Successful in 6s
CI / Build Test (Current Platform) (push) Successful in 6s
CI / Build All Platforms (push) Successful in 39s
Switch the vLLM config example from a bash code fence to jsonc and
convert its inline `#` comments to `//` so the snippet is valid JSONC.
2026-04-21 08:23:10 +00:00
jkunz 02bb3d2d8d v1.1.0
CI / Type Check & Lint (push) Successful in 6s
CI / Build Test (Current Platform) (push) Successful in 7s
Publish to npm / npm-publish (push) Failing after 23s
CI / Build All Platforms (push) Successful in 44s
Release / build-and-release (push) Successful in 56s
v1.1.0
2026-04-20 23:00:50 +00:00
jkunz 4f2266e1b7 feat(cluster,api,models,cli): add cluster-aware model catalog deployments and request routing 2026-04-20 23:00:50 +00:00
jkunz 83cacd0cf1 v1.0.1
CI / Type Check & Lint (push) Failing after 5s
CI / Build Test (Current Platform) (push) Failing after 5s
Publish to npm / npm-publish (push) Failing after 17s
Release / build-and-release (push) Successful in 48s
CI / Build All Platforms (push) Successful in 55s
v1.0.1
2026-01-30 03:18:21 +00:00
jkunz f8a679e028 fix(docs): revamp README with updated branding, expanded features, installation and uninstall instructions, resources, and legal/company information 2026-01-30 03:18:21 +00:00
jkunz daaf6559e3 initial
CI / Type Check & Lint (push) Failing after 5s
CI / Build Test (Current Platform) (push) Failing after 5s
CI / Build All Platforms (push) Successful in 49s
2026-01-30 03:16:57 +00:00