Console: update a worker from its menu and see when a newer version is available
Los mantenedores suelen responder en 1 día
Nadie ha tomado este issue todavía.
Evaluación
- Dificultad
- 4/5
- Tiempo estimado
- 3-5 días
- Aptitud para principiantes
- 35/100
- Tipo de issue
- Nueva funcionalidad
- Claridad
- Bien especificado
- Estado de actividad
- Activo
- Stack tecnológico
- react, typescript
- Área
- developer-experience, frontend
Línea de trabajo
Start with the console fleet screen tests in ui/src/screens/fleet/*.test.tsx and the action model and runner in ui/src/model/actions.ts and ui/src/state/actions.ts; the issue says these depend on #769 and #838. Check the worker view fields and routes from #838 before tracing the row and worker page entry points. Done means the specified update, progress, cancellation, failure, and available notice cases pass in the console tests, and docs/CONSOLE.md documents the action.
Escrito por el modelo de indexación a partir del texto del issue.
Descripción
Part of #825.
Parked (2026-10-21). The console layout gets decided with a clickable prototype before this task is finished: where "available" shows on the fleet table, how update progress and a failure look on a row whose health word is the worst finding, and the dialog's Force control and wording. Known gaps the prototype must close: the action runner sends no body and ignores the answer, and ConfirmSpec has no checkbox (PR #803). The fleet table has no version column (#766). The tests go in the console lane (e2e/console/). There is no agent session in the console. It stays task:draft until then.
Scope
From the console, an operator updates a worker through its ⋯ menu (added by #769), follows the progress on the worker's row, sees a failure, and sees an "available" notice where a newer version exists. Shared values: worker.update, the worker view fields, HTTP routes, error codes, events.
In short
On the fleet table and a worker's page, a worker with a newer version shows 0.4.0 available beside its version. Its ⋯ menu has Update…, which opens a confirm dialog: Update mac-studio-2 from 0.3.0 to 0.4.0? It takes no new leases and waits for its 2 leases to end. The dialog has a Force checkbox ("don't wait for leases") and Update / Cancel buttons. While it runs, the row's state reads updating: waiting for 2 leases, then installing and restarting, with a Cancel update item in the menu while waiting. A failure shows as a red Update failed (<stage>): <message> line on the worker page.
flowchart LR
M[⋯ menu: Update…] --> D[confirm dialog] -->|Update| P[POST /v1/workers/id/update]
P --> S[live worker list: update stage on the row]
S -->|failure| F[failure line on the worker page]
If this task goes wrong, a user would notice an Update item that does nothing, progress that never clears, or a failure the console doesn't show.
Technical spec
Facts
| # | Fact | Source |
|---|---|---|
| F1 | The console reads the worker list once, through useLiveResource<WorkerList>("/v1/workers"), and follows the event stream |
ui/src/state/workers.ts:60; ADR 0013 |
| F2 | The worker's version is shown in the inspector header only | ui/src/screens/workers/index.tsx:147-151; ui/src/model/worker-page.ts:526 |
| F3 | #769 (PR #803) adds the action model, the ⋯ menu, the confirm dialog and the action runner (ui/src/model/actions.ts, ui/src/ds/menu.tsx, ui/src/state/actions.ts); #766 (PR #787) moves the worker screens to ui/src/screens/fleet/ |
PR #803 and #787 file lists |
| F4 | Drained shows as an amber status word; lost or failed is red | ui/src/screens/workers/index.tsx:347; docs/internal/DESIGN.md:94-100 |
| F5 | worker.update routes, worker view fields and error codes are #838's |
#825 Shared values |
Modules touched
Searches: rg "drained" ui/src (display only today), rg "version" ui/src/model, and, on PR #803's branch, rg "drain" ui/src/model/actions.ts.
ui/src/model/actions.ts (Update and Cancel update actions), ui/src/state/actions.ts, ui/src/screens/fleet/ (row state word, available notice, worker page failure line), ui/src/model/worker-page.ts, docs/CONSOLE.md.
Contract and event changes
None: reads #838's fields, calls #838's routes.
Other code on the same state
The row's health word (F4): updating takes the amber tone and sits after drained in priority, so a drained, updating worker reads drained · updating. Searches: rg "workerHealth" ui/src (model 2, screens 3).
Failure modes
| Call | Outcome | What happens | Limit, counted from |
|---|---|---|---|
POST /v1/workers/{id}/update |
202 / 200 | toast Updating <label> to <to> |
#769's action timeout |
| same | 4xx / 5xx | toast with the error message, as #769's actions | — |
DELETE …/update |
409 UPDATE_NOT_CANCELLABLE |
toast with its message | — |
Rules in play
- ADR 0037 (proposed, PR #762): the console acts through the HTTP routes. It calls #838's routes.
- DESIGN.md: amber for "not taking work", red for failed.
Tests
Seam: the fleet screen tests after #766 (ui/src/screens/fleet/*.test.tsx), modelled on #769's action tests.
- Operator uses the console worker menu: Update in the ⋯ menu for admin (absent for an agent session); confirm dialog with version and lease count, Force; progress on the row (
waiting for 2 leases,installing,restarting, cleared atupdate: null); Cancel update while waiting; failure shown on the worker page;<latest> availablenotice (shown only when newer).
Done when
pnpm test uipasses with the line above.docs/CONSOLE.mddocuments the Update action and the notice.
Out of scope
- A fleet-wide update button.
Depends on
- #838
- #769
Written by an agent.
- Lenguaje dominante
- TypeScript
- Estrellas
- 20
- Forks
- 2
- Merge medio
- 8 h
- PR fusionados (30 d)
- 316
Preparar el entorno
Este proyecto no incluye contenedor de desarrollo, Dockerfile ni guía de contribución, así que la configuración corre por tu cuenta: empieza por su README y consulta nuestra guía para la primera contribución para los pasos generales.
Primeros pasos
- Lee el issue completo y luego la guía de contribución del proyecto.
- Comenta en el issue que vas a ocuparte — evita que dos personas hagan lo mismo.
- Haz un fork del repositorio y trabaja en una rama.
- Abre un pull request que haga referencia al número del issue.
Más de callstackincubator/simlock
-
bug:new
Dificultad 2/5 1-3 horas Aptitud para principiantes 78/100
callstackincubator/simlock#817 ·
Los mantenedores suelen responder en 1 día
-
bug:triage
Dificultad 2/5 1-3 horas Aptitud para principiantes 76/100
callstackincubator/simlock#778 · 1 comentario ·
Los mantenedores suelen responder en 1 día
-
Dificultad 2/5 1-3 horas Aptitud para principiantes 70/100
callstackincubator/simlock#639 · 1 comentario ·
Los mantenedores suelen responder en 1 día
-
A gateway updates a chosen worker, incompatible ones included, and routes around it meanwhileAbiertotask:ready
Dificultad 5/5 Más de una semana Aptitud para principiantes 35/100
callstackincubator/simlock#838 ·
Los mantenedores suelen responder en 1 día
-
Every daemon knows the latest Simlock version, and simlock status says when a newer one existsAbiertotask:ready
Dificultad 4/5 3-5 días Aptitud para principiantes 57/100
callstackincubator/simlock#837 ·
Los mantenedores suelen responder en 1 día
Todos los issues de callstackincubator/simlock
Issues similares
-
agent/scanner help wanted hive/claimed-by-scanner kind/bug stage/020-claimed
Dificultad 2/5 1-3 horas Aptitud para principiantes 90/100
hivecommons/rationguard#315 ·
Los mantenedores suelen responder en 1 día
-
agent-ready area: ui-sdk priority: p3 type: chore
Dificultad 2/5 1-3 horas Aptitud para principiantes 76/100
schlessera/brain-kit#2534 ·
Los mantenedores suelen responder en 1 día
-
bug lane:gh-5
Dificultad 2/5 1-3 horas Aptitud para principiantes 72/100
LLM-4-People/bridlewise-testing#122 ·
Los mantenedores suelen responder en 1 día
-
bug priority: P2
Dificultad 1/5 Menos de una hora Aptitud para principiantes 78/100
Los mantenedores suelen responder en 1 día
-
enhancement priority:low ready-for-dev
Dificultad 2/5 1-3 horas Aptitud para principiantes 67/100
Los mantenedores suelen responder en 1 día