Skip to main content

Incidents

Where to find it: Workspace → Operation → Incidents (/operation/incidents), titled Incident Command Center. It is scoped to the currently active project.

An incident is a problem ACE noticed on its own: a failed health check, an alert from a connected tool, or a signal from your deployment pipeline. You never create one by hand. When the same problem keeps firing, ACE groups the repeats under one incident instead of showing duplicates, and bumps an occurrence count. Sometimes two incidents turn out to be the same underlying problem. One gets merged into the other, and the merged one is never shown in this list, so only the surviving incident appears.

For what happens after you open an incident, see Incident detail.

The header​

  • Incident Command Center: the page title, with the subtitle "Monitor, investigate, and coordinate real cloud, infrastructure, deployment, and application incidents with ACE."
  • Refresh button (circular-arrow icon, top right): re-fetches the incident list for the active project. The icon spins while a fetch is in flight.

A footer note at the bottom of the page reads "Incidents are ingested per project via the monitoring webhook · auto-refreshing", a reminder that this list updates itself and that incidents come from the project's webhook, not from anything you type in.

What the tabs filter​

Three tabs sit under the header: Active, Resolved, All. Active is selected by default.

  • Active: every incident whose status is not resolved (that covers open, acknowledged, fixing, resolving, and suppressed).
  • Resolved: incidents whose status is resolved.
  • All: every incident, regardless of status.

Each tab shows a count badge next to its label, but only when that tab's count is greater than zero. Merged incidents are excluded from every tab's count and from the list; they do not have their own row anymore.

How do I filter the list?​

Below the tabs:

  • A search box, placeholder "Search incidents...", filters the visible list by matching your text (case-insensitive) against the incident's title or its source ID.
  • A severity dropdown, defaulting to "All severities". The other options are Critical, High, Medium, Low. Selecting one narrows the list to incidents with exactly that severity.

Both filters apply on top of whichever tab is active, and they combine (search AND severity AND tab must all match).

What's on an incident card?​

Each matching incident renders as one card, newest information first, in a plain list (no pagination on this page, so every matching incident is shown). Clicking a card opens its detail view at /operation/incidents/:projectId/:id.

Each card shows:

  • A colored dot on the left indicating severity: red for Critical, orange for High, yellow for Medium, green for Low.
  • The incident's title, in bold.
  • A second line with, in order: the source ID (monospace), the severity label, the status label, and the type (or "Unknown" if the incident has no type). If the incident has fired more than once, a small ×N badge follows, showing the occurrence count.
  • A third line with the incident's source (the tool or pipeline it came from, or "—" if none), and how long ago it was detected (for example "5m ago", "3h ago", "2d ago").

Loading and empty states​

  • While the list is loading, the page shows "Loading incidents…".
  • If there are no incidents at all for this project, the empty state reads "No incidents found" with the note "No incidents for this project yet — alarms arrive via the ingest webhook."
  • If incidents exist but none match the current tab, search, and severity combination, the same "No incidents found" message shows instead with the note "Try adjusting your filters."
On this page