Metric Families
Find a metric by name, kind, or label, then use its registered fields in your query.
On this page
A Chio metric family is registered once, with its exact name, kind, labels, and help text, and histogram buckets where declared. The workspace registry holds 60 families.
Families are grouped under the Platform section that owns them, by name prefix. Copy the name and labels into your query as written. For the decision on one request, read its receipt; metrics count and time the work across requests.
Search for a name such as receipt, a kind such as histogram, or a label such as outcome.
Kernel
| Name | Kind | Labels | Help |
|---|---|---|---|
histogram | authority | Capability revocation propagation lag in seconds. Buckets: 1, 5, 15, 30, 60, 120, 300. | |
counter | guard_id, reason_class | Total WASM guard denies by reason class. | |
histogram | guard_id, verdict | WASM guard evaluation duration in seconds. Buckets: 0.0001, 0.0005, 0.001, 0.005, 0.01, 0.025, 0.05, 0.1, 0.25, 0.5, 1. | |
counter | guard, outcome | Total guard evaluation outcomes across native and WASM guards. | |
counter | guard_id | Total WASM guard fuel units consumed. | |
histogram | guard_id, host_fn | WASM guard host-call duration in seconds. Buckets: 0.00001, 0.00005, 0.0001, 0.0005, 0.001, 0.005, 0.01, 0.05, 0.1. | |
gauge | guard_id, epoch | Loaded WASM guard module size in bytes. | |
counter | guard_id, tenant_id | Total WASM guard pool checkout outcomes by tenant. | |
counter | guard_id, tenant_id | Total WASM guard pool evictions by tenant. | |
gauge | guard_id, tenant_id | Warm WASM guard pool size by tenant. | |
counter | guard_id, outcome | Total WASM guard reload outcomes. | |
counter | guard_id, verdict | Total WASM guard verdicts by guard and verdict. | |
histogram | surface, outcome | Kernel mediation decision latency in seconds. Buckets: 0.025, 0.05, 0.075, 0.1, 0.25, 0.5, 1, 2.5. | |
gauge | none | Seconds since the receipt-store checkpoint last advanced while data was pending on the local store. | |
gauge | none | Uncheckpointed receipt entry_seq range (end - start) on the local store. | |
histogram | store, outcome | Receipt write latency at the local store boundary in seconds. Buckets: 0.01, 0.025, 0.05, 0.1, 0.25, 0.5, 1. | |
counter | outcome | Total receipt write outcomes after policy or guard evaluation. |
Node
| Name | Kind | Labels | Help |
|---|---|---|---|
histogram | route, outcome | PagerDuty or OpsGenie alert dispatch latency in seconds. Buckets: 0.25, 0.5, 1, 2.5, 5, 10. | |
counter | route, outcome | Total PagerDuty or OpsGenie alert dispatch outcomes. | |
counter | reconciliation | Total budget or payment holds retained after an ambiguous post-dispatch outcome, labeled by whether durable reconciliation is available. | |
counter | surface, outcome | Total tool-dispatch failures that did not bypass mediation. | |
gauge | exporter | Dead-letter queue depth by exporter. | |
counter | surface | Total suspected fail-open paths detected by SRE guards. | |
counter | outcome | Requests the hosted market edge admitted, refused, or shed. | |
counter | none | Total OTEL ingress batches dropped by bounded queue admission. | |
counter | none | Total OTEL receipt sink batches dropped before append. | |
counter | none | Total settlement observer routing invocations with an unresolved outcome. | |
counter | outcome | Total sidecar request outcomes at the mediation edge. | |
counter | reason | Total receipt signing requests blocked by bounded queue capacity or byte budget. | |
histogram | exporter, severity | SOC export lag in seconds from receipt persistence to sink acknowledgement. Buckets: 30, 60, 120, 300, 600, 1800. | |
counter | exporter, outcome | Total SOC export outcomes for audit rows. |
Cluster
| Name | Kind | Labels | Help |
|---|---|---|---|
gauge | none | Trust-control readiness state, where 1 is ready and 0 is not ready. |
Swarm
| Name | Kind | Labels | Help |
|---|---|---|---|
histogram | witness, outcome | Anchor round latency in seconds. Buckets: 0.1, 0.5, 1, 2.5, 5, 10. | |
counter | treaty, recipient | Total local pheromone gossip queue overflows. | |
counter | outcome | Total local pheromone receiver deposit admission outcomes. | |
histogram | outcome | Local pheromone receiver batch verification latency in seconds. Buckets: 0.001, 0.005, 0.01, 0.025, 0.05, 0.1, 0.25, 0.5, 1. | |
counter | reason | Total local pheromone receiver rejections by bounded reason. | |
counter | responder, treaty | Total bounded catch-up bytes served by relay. | |
counter | reason | Total bounded catch-up requests denied by reason. | |
counter | reason | Total relay outbox batches moved to dead letter. | |
counter | recipient, outcome | Total live pheromone relay delivery outcomes. | |
counter | profile, reason | Total relay endpoints denied by profile lint or delivery policy. | |
histogram | outcome | Live pheromone relay delivery latency in seconds. Buckets: 0.001, 0.005, 0.01, 0.025, 0.05, 0.1, 0.25, 0.5, 1. | |
counter | peer | Total relay nonce replay conflicts. | |
gauge | none | Oldest pending relay outbox age in seconds. | |
gauge | status | Relay outbox depth by bounded status. | |
counter | reason | Total live pheromone relay rejections by bounded reason. | |
counter | profile | Total stale peer-directory bundles or documents rejected. | |
gauge | none | Relay scheduler leases past their expiry. |
Federation
| Name | Kind | Labels | Help |
|---|---|---|---|
histogram | result | Federation hop latency in seconds. Buckets: 0.01, 0.025, 0.05, 0.1, 0.25, 0.5, 1. | |
counter | result | Total federation hop outcomes. | |
histogram | lane | Iroh federation- | |
gauge | lane | Iroh federation- | |
counter | outcome | Total iroh federation- | |
counter | source | Total iroh federation- | |
counter | outcome | Total iroh federation- | |
counter | lane, outcome | Total iroh federation- | |
counter | outcome | Total iroh federation- | |
gauge | none | Iroh federation- | |
counter | seam, reason | Total iroh federation- |
Registry source and grouping
The registry lives in metrics.snapshot, and emitters use its name constants. Every Chio release passes the sre-metrics-registry gate before it is tagged, so a renamed family fails the release before it reaches your dashboards. chio release verify --evidence qualification shows its result for the release you run. The registry carries no section field: each family is filed by its name prefix, and a prefix that matches no section files under Node.