nobody declared these artifacts as captured from a real run, so nothing here may be read as measured, whatever it was built from
VoiceGateway · run ramp-500 · provenance synthetic
Run window (UTC) 2026-07-31T18:00:00Z to 2026-07-31T19:00:00Z · artifacts no artifact checksum
At least one gate could NOT be evaluated. This is NOT a pass: the run did not demonstrate a healthy deployment, it failed to measure something. Read the gate rows below for which one, and treat the un-measured half as unknown rather than fine.
The verdict is the worst gate below (PASS < WAIVED < WARN < UNKNOWN < FAIL), decided when the run executed, not when this report was exported.
| Test | Peak concurrency | Duration | Established | Peak CPU | Peak memory | Failed | By cause |
|---|---|---|---|---|---|---|---|
| ramp-500 | 492 | 60.0 min | 99.900% | 68.0% | 61.2% | 15 | timeout 3, unexpected_sip 9 |
Peak CPU and memory are the worst node’s peak over each test’s window, correlated by time overlap. They describe the fleet during the test, not any particular call.
No capacity table, and none was attempted: whoever built this payload supplied no capacity block, so nothing here is a statement about the run. Sizing without a derivation would mean inventing the one number the whole table rests on.
| Status | Gate | What it found | Metric | Value | Threshold |
|---|---|---|---|---|---|
| PASS | g0 | d | not recorded | not measured | not measured |
| UNKNOWN | g1 | d | not recorded | not measured | not measured |
What it takes to run this again. Scenario files and generator configuration are referenced by name rather than reproduced here: they are the work of whoever authored them.
This is background, not a description of the run above. Each item is a fact somebody recorded when the Source column says, and nothing here is re-verified against this run. Where an item names a fleet size, a version or a limit, read it as what was true when it was recorded: the deployment may have changed since, and the measured sections above are the authority on what actually ran.
As run. Absolute URLs are reduced to a host label, so a command here may need its endpoint restored before it will execute.
| Item | Detail | Source |
|---|---|---|
| ramp | gossipper sipp -sf uac-media.xml -rsa media.example.com:5060 -l 500 -r 4.1667 -m 15000 -pause_ms 118000 -trace_stat | operator notebook, step 4.1 |
Verify these against the binary you are holding. A generator's own documentation can disagree with its behaviour, and a flag whose default is wrong produces a run that completes and measures the wrong thing.
| Item | Detail | Source |
|---|---|---|
| -l | Max concurrent calls. Defaults to 1: omitting it caps the whole run at one concurrent call. | gossipper sipp -h, binary 0.1.62 |
What it takes to get a working binary.
| Item | Detail | Source |
|---|---|---|
| build | Does not build on macOS; cross-compile for linux/amd64. | observed on 0.1.61 and 0.1.62 |
Read this section before acting on anything above. A load run measures a fleet from outside, through a generator placing real calls, and these are its structural limits, not this run’s bad luck.