Triage issues

Narrow a full feed to what's worth acting on, then resolve, acknowledge, or reassign the issues that matter.

Error Feed’s list page, in the left sidebar under Error Feed (see the overview if you haven’t opened it yet), shows every detected issue across your projects, scoped to the last 7 days until you change the range. A full feed is a queue, not a to-do list: some rows need attention today, most don’t.

This guide takes you from a full feed to a handled list: narrow it to what’s worth looking at, scan the table for what actually decides priority, then act on what you find, one issue at a time or several at once.

Narrow the feed

Type into the search box (placeholder Search errors) to match against the error name, issue group, or category. Next to it sit five selects: project, status, severity, and fix layer each open on an All value until you narrow them, while time range opens already scoped to Last 7 days. See Issue fields & filters for the exact set of values each one accepts.

Tip

Start from Severity: Critical and Status: Escalating. Critical narrows to what matters most, and Escalating is the one status of the four that hasn’t settled into Acknowledged, For review, or Resolved. See Severity & Status for when to pick each one.

Once project, status, severity, or fix layer is set, a Clear button appears next to the selects (tooltip: “Clear all filters”) to reset everything in one click instead of undoing each select by hand.

Scan the table

Each row is one issue. Eight columns run left to right: Error, Severity, Status, Events, Users, Fix Layer, Trend (14d), and Last seen.

Three of them decide priority. Severity says how bad it is, Status says where it sits in your workflow, and Trend (14d) shows whether it’s climbing, flat, or settling down.

The rest is context:

  • Error names what’s failing
  • Events and Users size the blast radius
  • Fix Layer points at where in your system the fix belongs
  • Last seen says when it last fired

At the bottom, set Results per page to 10, 25, or 50, and move through the rest with Back and Next.

Act on what you find

Three ways to act, each suited to a different job in the triage workflow:

  • Bulk actions for many rows moving to the same status at once
  • Header buttons for a quick resolve or acknowledge on a single issue
  • Metadata sidebar for a severity or assignee change on a single issue

Note

None of the three ways below show a toast. There’s no confirmation of success and no warning on failure, so the save happens silently either way. To check a change went through, re-check the Status (or Severity) column for that row, or refresh the feed; if the value hasn’t moved, repeat the action.

Handle many at once

Tick the checkbox on any row and a bulk-select toolbar appears above the table. Tick more rows, then open Bulk actions and pick Mark as Resolved, Mark as Acknowledged, Mark as For Review, or Mark as Escalating to move every selected issue to that status in one go.

Resolve or acknowledge one issue

Click a row to open the issue. Its header carries three buttons: Resolve moves the issue to resolved, Acknowledge moves it to acknowledged, and Ignore issue moves it to escalating despite the label. All three disable themselves while the update is in flight.

Change status, severity, or assignee from the sidebar

Every issue also has a metadata sidebar on the right. Its Status and Severity rows each open a menu to set a new value directly. The Assignee row reads Assign until someone’s on it; click it to open a menu headed Assign to, listing everyone in your org plus an Unassign option once someone’s set.

Note

Assigning someone to an issue only records it on the issue itself. It sends no notification of any kind, so tell them yourself if it needs to reach them.

Dive deeper

Was this page helpful?

Questions & Discussion