Macro is a unified workspace for teams: email, chat, docs, tasks, agents, calls, and CRM — @-linked together with shared AI memory.
TypeScript agentaiai-agentsall-in-onecrmcrm-systememaillinearmcpmessagingnotesnotionofficerustslackslack-alternativestartupstartup-toolstypescriptworkspace
system_status
What it is
Platform status banner: Macro staff set the status (operational, degraded, major outage with a message) and every user sees a banner while it is not operational. Each update is kept as history; the newest is current.
How it works
POST /system-status (staff) inserts a row in system_status, then
broadcasts the new CurrentSystemStatus to every open app as a
system_status websocket message through the connection gateway's
/message/broadcast. Apps load GET /system-status (public) on start and
keep whichever of the fetched or pushed status is newer (updatedAt).
Start here
src/domain/models.rs:SystemStatus(message only on non-operational variants, enforced again by the table's CHECK) and the wire shapes.src/domain/service.rs: staff check, store, then best-effort broadcast.src/inbound/axum_router/set_status.rs: the documented endpoint staff call.
Scope
- Owns: status storage, history, the staff check, and the broadcast payload.
- Not here: delivery to sockets →
services/connection_gateway; the banner and dashboard →apps/web/src/features/system-status/.
Boundaries
- Depends on:
macro_authorization(axum),connection_gateway_client. - Used by:
services/authentication_service, which mounts the router at/system-statusand lists it insrc/api/swagger.rs.
Checks
cargo test -p system_status --all-features(SQLX_OFFLINEunset).nix develop --command just prepare_dbfrom the repo root after query changes.
Gotchas
- A failed broadcast does not fail the request: the row is already current, and a retry would record a duplicate. Apps refetch on focus.
Related
- Rules:
.agents/rules/rust/architecture.md· Skill:cloud-storage-hexagonal-architecture