Status

Status Uptime

Daily uptime ribbon for the public status page.

Returns days per-day status cells per service, oldest-first. Each cell is one of operational | degraded | outage | no_data with an optional summary describing the incident span.

Source is always redis (the daily-rollup cache); cells without a cached entry default to no_data. The app.tasks.status_aggregator background loop keeps the cache fresh every 5 min and backfills 90 days on startup.

Performance: single Redis MGET regardless of days value. Sub-millisecond at p99 even with the 90-day window — the old per-request path took ~18 s on cache miss.

All services in SERVICES currently share the same daily roll- up — we don’t (yet) compute per-service status separately. Per-service granularity is a documented follow-up; for now the UI shows the same row for each service which still surfaces actual outages.

GET

status

uptime

Status Uptime

curl --request GET \
  --url https://api.polysimulator.com/v1/status/uptime

{
  "generatedAt": "<string>",
  "source": "<string>",
  "services": [
    {
      "id": "<string>",
      "label": "<string>",
      "days": [
        {
          "date": "<string>",
          "status": "<string>",
          "summary": "<string>",
          "impacted_hours": 123,
          "measured_hours": 123
        }
      ]
    }
  ]
}

Query Parameters

days

integer

default:90

Required range: 1 <= x <= 180

Response

Successful Response

generatedAt

string

required

source

string

required

services

ServiceRow · object[]

required

Show child attributes

Heartbeat dead-man's-switch — auto-cancel orders on missed pings.Register or refresh a dead-man's-switch heartbeat. If the server stops receiving heartbeats for longer than ``interval_ms + grace``, all resting orders for this API key's account are auto-cancelled. Polymarket-compatible: ``POST /heartbeats`` (PM-shape root path, no ``/v1/``) accepts the same body and response. Both paths route to this handler. Bounds: ``interval_ms`` must be ``[1000, 60000]``. Bots should ping at half the interval (or finer) to stay ahead of expiry. Grace window: ``max(1s, 0.25 * interval_ms)``. **Storage**: Redis-backed sorted set + per-registration hash. Survives api-worker restarts; the sweeper runs on the background leader worker with a 60s leader-election TTL so a leader crash promotes a follower automatically. Refresh/sweep TOCTOU race is closed via Redis ``WATCH``/``MULTI`` — a refresh that arrives between the sweeper's ``ZRANGEBYSCORE`` and its ``ZREM`` aborts the dead-man fire and keeps the bot alive.

Status Uptime

curl --request GET \
  --url https://api.polysimulator.com/v1/status/uptime

{
  "generatedAt": "<string>",
  "source": "<string>",
  "services": [
    {
      "id": "<string>",
      "label": "<string>",
      "days": [
        {
          "date": "<string>",
          "status": "<string>",
          "summary": "<string>",
          "impacted_hours": 123,
          "measured_hours": 123
        }
      ]
    }
  ]
}