Overview
Collect, query, and alert on the telemetry your workloads emit. Observability unifies the three signals, logs, metrics, and traces, behind one workspace-scoped API, so you can search logs, run metric queries, follow a request across services, and act on what you find without stitching together separate tools.
The service speaks open query standards rather than a proprietary dialect: log queries use
LogQL, metric queries use
PromQL, and
trace search uses TraceQL on the qibdo
engine, so the expressions you already know keep working and nothing locks you in. Telemetry
queries and per-engine configuration take an engine path segment that selects the provider:
qibdo is available today, with aws, gcp, and azure planned behind the same query surface,
so one API will cover telemetry wherever it lives. Silences, activities, and operations are
engine-neutral and skip the segment.
Core resources
Section titled “Core resources”- Logs: query a workspace’s log lines with a bounded, paginated
query(LogQL on theqibdoengine), tail them in near real time, or list the available log streams. Every bounded query is scoped to a workspace and a required time window; tailing is live and takes no window. - Metrics: run instant (
query) and range (queryRange) metric queries in PromQL, and discover what is queryable through the metric descriptors and label catalogues. - Traces: search for traces with TraceQL, fetch a single trace by id, or read the service graph the spans imply (call and error counts on the edges between services), so you can see how requests flow across the platform.
- Alert rules: named rules that evaluate a PromQL or LogQL expression. An alert rule notifies when its condition holds; a recording rule precomputes a derived series instead. Create, update, and delete them per workspace and engine.
- Silences: time-bounded mutes that suppress alert notifications during a maintenance window or a known incident, without disabling the underlying rule.
- Retention policies: per-workspace rules that decide how long each signal is kept before it ages out.
- Log sinks: routing rules that select matching log lines and deliver them to a destination: an archive, a downstream stream, or an external endpoint.
- Ingestion configs: the settings that govern how incoming telemetry is accepted and shaped before it is stored.
- Activities and operations: an audit trail of what changed, and the operation records that mutating calls return, already completed.