The service‑scoped list of individual sampled traces/spans — the same Root/All Spans experience as the Traces Explorer, filtered to the current service and time window.

Get here: Service Overview → Distributed Tracing tab.

Two tabs

• Root Entry Spans — one row per trace, keyed on the request that started it, so you browse whole requests at a glance. Columns: Service · Operation · Start Time · Spans Count · Service Count · Error Count · Duration · Trace Duration (with duration bars; any trace containing an error is highlighted red).
• All Spans — one row per individual span (the trace flattened), for when you need to hunt down a specific operation. Columns: Service · Operation · Start Time · Duration. A row’s … → View details opens the span; toggle Worst Spans (threshold, default 500 ms) to surface only the slow ones.

Data source: sampled spans.

Configure columns (⚙ gear)

The gear icon at the top‑right of the spans table (next to Worst Spans) opens column settings. Columns are grouped into Default (built‑in fields — Service, Operation, Start Time, Duration, and the Root Entry Spans counts) and Custom (any span attribute from your instrumentation — e.g. http.route, user_id, region, db.system). Check an attribute to add it as a column, then sort/reorder it or toggle line wrap. Your choices are remembered with the view (and carry into saved/shared views).

Open a trace

Click a row → the full Trace Detail waterfall.

Limit

The spans list shows a capped set (footer reads e.g. Showing 500 of 2689 results). Narrow the filters/time range to reach the specific spans. (The standalone Trace Explorer caps at 1000.)