tuandClaude Opus 4.7
985cd39b2a
Clusters with no state-transition events (e.g. a clean rollout that didn't cause a degradation) now show a green dot instead of blue. Rationale: absence of a transition means the workload stayed in its existing state — the rollout went through unscathed, which is the healthy outcome a reader is scanning for. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
oumta-dashboard plugin
Fullscreen view for the Oumta cluster dashboard (Go backend in oumta-dashboard/).
Open
- Oumta button in the hyprcrd top bar (next to Tasks)
window._oumtaToggle()from the console / other plugins
Close with ESC or the × button.
Views
- Overview — cluster state, workload counts, attention list
- Workloads — filterable table (name, state, replicas, cpu/mem, image, reason)
- Timeline — event stream; clicking a workload filters down to its own events
Namespaces
Tab row at the top. Currently two mocks (oumta-dev, oumta-beta). Real list comes from the backend once /api/namespaces exists.
Configuration
Hardwired at the top of oumta-dashboard.js in the CONFIG block:
- baseUrl — where the Oumta Go backend is reachable (default
http://localhost:8085). - authToken — optional, sent as
Authorization: Bearer …when non-empty. - pollIntervalSec — polling fallback cadence when SSE is unavailable. Default 15 s.
No UI settings panel: the dashboard only runs against one specific deployment, so exposing these as editable fields would be noise.
State today
Everything is mocked. The live indicator in the top right shows polling once the overlay is open, but no real HTTP call fires yet. See TODO.md for the integration plan.
Disabling
Add "oumta-dashboard" to the disabled array in data/plugins.json.