Observability

Measure everything.

OpenTelemetry traces, Prometheus metrics, structured JSON logs. Built into the request path — not bolted on after.

Booting…

zaps://monitoring

Ops / sec

—

P99 Latency

—

Memory

—

Active Principals

—

Audit Chain

—

Uptime

—

Initializing…0/6 panels

Distributed trace

One SET. Nine spans.

Total: 0.00ms · 0/9 spans

0ms0.5ms1.0ms1.6ms

TLS handshake

—

Auth resolve DID

—

Scope check

—

Route to shard

—

Audit pre-log

—

Engine SET

—

Journal append

—

Audit post-log

—

Response

—

Protocol

Edge

Engine

Persist

Cross-cut

Integrations

Plug into your stack.

OpenTelemetry

W3C Trace Context propagation. OTLP export to any collector.

Prometheus

/metrics endpoint. Scrape-ready. Custom zaps_* metric namespace.

Structured Logs

JSON logs to stdout. Trace ID, principal, operation on every line.

If you can't see it, you can't fix it.

Every request traced. Every metric exported. Every log structured.

Next: Performance →Back to Durability