Topology page: right-click "Deploy all" on a site/area node (that layer and below) #40

New Issue

2026-10-22T13:46:42-04:00

dohertj2 commented

2026-10-22 13:46:42 -04:00

Request

On the Central UI Topology page, add a context-menu (right-click) action Deploy all on a site or area node. It deploys every instance in that node and its descendants: sub-areas recursively, plus instances directly in the node.

Example (QA): right-click Aisle001 → Deploy all → deploys every instance under Aisle001.

Considerations

Role: Deployer-gated, and limited to the operator's site scope, like the existing per-instance deploy actions.
Reuse existing behaviour: build on the bulk site deploy path (DeploymentService.DeploySiteAsync) and keep its semantics:
- it skips Disabled and up-to-date instances unless forced;
- it uses bounded parallel Send;
- it is not all-or-nothing;
- it reports a summary including SkippedInstances.
  This likely needs an area-scoped variant (or an instance-id filter) on that path.
Before running: confirm in a dialog showing the instance count. Optionally offer a "force" option to redeploy up-to-date instances, mirroring deploy site --force.
After running: show the outcome as a toast/summary (deployed / skipped / failed). Rows update live on the Deployments page.
CLI parity: worth adding, e.g. deploy area --id <areaId> [--force].

## Request On the Central UI **Topology** page, add a context-menu (right-click) action **Deploy all** on a site or area node. It deploys every instance in that node and its descendants: sub-areas recursively, plus instances directly in the node. Example (QA): right-click `Aisle001` → Deploy all → deploys every instance under Aisle001. ## Considerations - **Role:** Deployer-gated, and limited to the operator's site scope, like the existing per-instance deploy actions. - **Reuse existing behaviour:** build on the bulk site deploy path (`DeploymentService.DeploySiteAsync`) and keep its semantics: - it skips Disabled and up-to-date instances unless forced; - it uses bounded parallel Send; - it is not all-or-nothing; - it reports a summary including `SkippedInstances`. This likely needs an area-scoped variant (or an instance-id filter) on that path. - **Before running:** confirm in a dialog showing the instance count. Optionally offer a "force" option to redeploy up-to-date instances, mirroring `deploy site --force`. - **After running:** show the outcome as a toast/summary (deployed / skipped / failed). Rows update live on the Deployments page. - **CLI parity:** worth adding, e.g. `deploy area --id <areaId> [--force]`.

dohertj2 commented

2026-10-22 14:58:17 -04:00

Implemented on main (b9751473..e029dc1c, local — not yet pushed). Right-click a site or area on Topology → "Deploy all…" → dialog with in-scope / disabled counts and a Force checkbox → summary toast (deployed / skipped / failed). Backend: DeploymentService.DeployAreaAsync runs the same DeploySiteAsync core narrowed to the area + all descendant areas (Disabled always skipped, up-to-date skipped unless forced, bounded-parallel send, not all-or-nothing, SkippedInstances in the summary). MgmtDeployArea is Deployer-only and authorised by the area's owning site; CLI: scadabridge deploy area --id [--force] (exit 1 on any failure). Instances with no area are only reachable via a site deploy. Live-gated on the docker rig: a parent-area deploy picked up the child area's instance, a second run skipped it as up-to-date, --force redeployed it, an unknown id exits 1; Playwright DeployAll_OnArea_ShowsSummaryToast passes. Not deployed to QA/prod.

Implemented on main (b9751473..e029dc1c, local — not yet pushed). Right-click a site or area on Topology → "Deploy all…" → dialog with in-scope / disabled counts and a Force checkbox → summary toast (deployed / skipped / failed). Backend: DeploymentService.DeployAreaAsync runs the same DeploySiteAsync core narrowed to the area + all descendant areas (Disabled always skipped, up-to-date skipped unless forced, bounded-parallel send, not all-or-nothing, SkippedInstances in the summary). MgmtDeployArea is Deployer-only and authorised by the area's owning site; CLI: scadabridge deploy area --id <areaId> [--force] (exit 1 on any failure). Instances with no area are only reachable via a site deploy. Live-gated on the docker rig: a parent-area deploy picked up the child area's instance, a second run skipped it as up-to-date, --force redeployed it, an unknown id exits 1; Playwright DeployAll_OnArea_ShowsSummaryToast passes. Not deployed to QA/prod.

dohertj2 closed this issue

2026-10-22 14:58:17 -04:00

dohertj2 referenced this issue from a commit

2026-10-22 14:59:33 -04:00

feat(deploymentmanager): seam for subtree deploy — DeployAreaAsync / PreviewSubtreeDeployAsync signatures + SubtreeDeployPreview (Gitea #40)

dohertj2 referenced this issue from a commit

2026-10-22 14:59:33 -04:00

feat(deploymentmanager): DeployAreaAsync bulk-deploys an area subtree through the shared DeploySiteAsync core; PreviewSubtreeDeployAsync returns confirm counts (Gitea #40)

dohertj2 referenced this issue from a commit

2026-10-22 14:59:33 -04:00

feat(managementservice): MgmtDeployArea — Deployer-only, scoped by the area's owning site (fail-closed on an unknown area), long-running like MgmtDeploySite (Gitea #40)

dohertj2 referenced this issue from a commit