Find resources and activity

Use Search resources in Citadel's header, or press Ctrl+K on Windows/Linux or ⌘K on macOS. The shortcut is inactive while typing in an editor, input, or another dialog; use the header button in that case.

Search across Citadel

1. Enter at least two characters from the resource name. Search is case-insensitive.
2. Check the result's type, status, and parent context before opening it.
3. Refine the name if there are many matches; the popup shows a limited selection per resource type. With an empty query, select a category to open its list.

Search includes Platforms, Stacks, Deployments, managed Swarm Services, Git Repositories, Registries, Automation Actions, Backup Policies and Repositories, Build Projects, and Build Pools. Some types also match a useful field such as a Platform address, Git URL, Registry host, or Build branch.

Results require Read access. A linked parent is shown only when you can also read it. If a resource is missing, try its list page and review access assignments before assuming it was deleted.

Narrow a resource list

Use a list's search and available Platform or Tag filters to narrow the results. Clear filters when a known resource appears to be missing. On Deployments and Stacks, Updates available shows detected updates; an empty result does not prove that every resource has been checked successfully.

Global search does not search container logs, Compose contents, secret values, or run history. Open the owning Platform for Docker containers, images, networks, and volumes. Use the resource's Runs, Releases, or Activities view for its history.

Investigate an operation

Find the resource and time in Activities, then open the event to inspect its details. Use the event's copy-link control when sharing it with a teammate who has access. For a failed job, also inspect its run logs; for an alert, follow the investigation workflow.

Saved configuration, a recorded operation, and current runtime state answer different questions. Check the running workload and application endpoint before retrying a deployment whose progress connection was interrupted.