Reading Your Telemetry

Your workload’s telemetry is held in your organization’s telemetry stores, one per provider and region it runs in, and read through the platform API. Every query answers your organization’s own records alone.

Queries

Call	Body	Answers
POST /v1/orgs/{org}/vantastar/query/logs	A log query	A page of log records, newest first
POST /v1/orgs/{org}/vantastar/query/traces	A trace query	A page of spans
POST /v1/orgs/{org}/vantastar/query/metrics	A metric query	The series
POST /v1/orgs/{org}/vantastar/query/raw	A query in the store’s own language	The store’s own answer

Every query but the raw one names stores, every store of the organization when absent. from and to are RFC 3339 instants, to now when absent, and a query spans at most 30 days. attributes is a map from an attribute’s name to the value it must equal, so {"vantacloud.workload.name": "api"} reads one workload.

A log query takes severity, the least severity, scope, an instrumentation scope’s name, and text, a substring of the body. A trace query takes traceId, or from with name, status and minDuration in milliseconds. A page holds at most 10,000 records, 100 when limit is absent, and carries a cursor while more remain.

A metric query takes name, aggregation, one of sum, avg, min, max, count, p50, p95 or p99, groupBy, the attributes the series are split by, and step, the width of each point in seconds, at least 60. The answer holds unit and series, each with its labels and points.

Uptime

Read a probe’s availability as the avg of vantacloud.probe.up grouped by vantacloud.probe.name, the share of its probes that succeeded in each step, and its latency as the p99 of vantacloud.probe.duration. An error log record’s traceId reads its whole trace by a trace query naming it.

{
  "name": "vantacloud.probe.up",
  "aggregation": "avg",
  "attributes": { "vantacloud.workload.name": "api" },
  "groupBy": ["vantacloud.probe.name"],
  "from": "2026-11-04T00:00:00Z",
  "step": 60
}

Alerts

An alert rule, a default one or one of your own, is a resource of the kind vantastar/alert-rule at /v1/orgs/{org}/vantastar/alert-rules/{name}, and GET /v1/orgs/{org}/vantastar/alert-rules/{name}/alerts answers the alerts it fired. A rule of your own names a condition, a metric threshold, a log count over a window or a signal’s absence over a window, its severity, every, how often it is evaluated, and notify, its emails and webhooks.