2024-01-15T00:00:00Z).
Time Breakdown
Where request time went — database, cache, outbound HTTP, queue/RPC and application — as a summary and as a time series.Path Parameters
string
required
Account identifier (format:
acc_ followed by 27 alphanumeric characters)Query Parameters
string
required
Start of the window (ISO 8601, UTC)
string
required
End of the window (ISO 8601, UTC)
string
Limit to one bucket
string
Limit to one environment
string
Limit the breakdown to one normalized endpoint
Response
object
Totals for the window:
requestCount, errorCount, totalMs, dbMs, cacheMs (Redis and
other caches combined), httpOutMs, queueMs (queue and RPC combined), appMs (residual
in-process time, floored at 0), spanCount, p50Ms, p95Ms, p99Ms.array
The same fields per bucket, plus
bucket (bucket start) and p95Ms. The bucket width is derived
from the requested range.boolean
True when the answer came from the pre-aggregated metrics tables rather than the request rows.
Endpoints By Time Consumed
Endpoints ranked by the total time they consume, with the same dependency split. Total time, not average latency, is what identifies where your capacity goes.Query Parameters
string
required
Start of the window (ISO 8601, UTC)
string
required
End of the window (ISO 8601, UTC)
string
Limit to one bucket
string
Limit to one environment
number
Maximum endpoints to return (1–1000, default 50)
Response
array
Each entry carries
endpoint, requestCount, errorCount, averageDurationMs, p95Ms,
p99Ms, lastSeen, and the span-time fields (totalMs, dbMs, cacheMs, httpOutMs,
queueMs, appMs, spanCount).Regressions
Endpoints whose latency or error rate moved against a baseline window, ranked by the time the change costs.Query Parameters
string
required
Start of the current window (ISO 8601, UTC)
string
required
End of the current window (ISO 8601, UTC)
string
Limit to one bucket
string
Limit to one environment
string
previous (default) compares against the window immediately before; weekAgo compares against
the same window seven days earlier, which suits traffic with a weekly shape.number
Ignore endpoints below this request count in either window (default 20)
number
Maximum endpoints to return (1–1000, default 20)
Response
string
The baseline that was applied.
array
Each entry carries
endpoint, requestCount, baselineRequestCount, p95Ms, baselineP95Ms,
p95DeltaPct, errorRate, baselineErrorRate, errorRateDelta, totalMs, baselineTotalMs
and score (latency/error movement weighted by traffic — the ranking).If the baseline window falls outside your retention, the comparison is skipped rather than
silently compared against a clamped, overlapping window — the response comes back with no points.
Distributed Trace
Every request that belongs to one trace, in time order. The trace id is stamped on each request at ingest — from the inboundtraceparent header when present, otherwise from the first span — so
this is an equality lookup rather than a walk over span parentage.
Query Parameters
string
required
Trace id shared by the requests in the trace
string
required
Start of the window to search (ISO 8601, UTC)
string
required
End of the window to search (ISO 8601, UTC)
string
Limit to one bucket
string
Limit to one environment
number
Maximum requests to return (1–1000, default 100)
Response
string
The trace that was resolved.
array
Each entry carries
dt, sid, endpoint, method, host, path, direction, bucketSid,
environmentSid, statusCode, durationMs and spanCount.