SpiceDB Telemetry

SpiceDB reports metrics that are used to understand how clusters are being configured and the performance they are experiencing. The intent of collecting this information is to prioritize development that will have the most impact on the community.

Telemetry can always be disabled by providing the flag --telemetry-endpoint="".

Implementation

Telemetry is reported via the Prometheus Remote Write protocol. Any metrics prefixed with spicedb_telemetry are reported hourly to telemetry.authzed.com. You can find all of the code in internal/telemetry.

Collected metrics

spicedb_telemetry_info (Gauge)

Information about the SpiceDB environment.

CPU Architecture: architecture of the CPU
vCPUs: number of CPUs reported by the Go runtime
OS: operating system family as reported by the Go Runtime
Datastore Engine: datastore engine configured
Cluster ID: unique identifier for a cluster's datastore
NodeID: unique identifier for the node, usually the hostname

spicedb_telemetry_object_definitions_total (Gauge)

Count of the number of objects defined by the schema.

Cluster ID: unique identifier for a cluster's datastore
NodeID: unique identifier for the node, usually the hostname

spicedb_telemetry_relationships_estimate_total (Gauge)

Rough estimate of the number of relationships stored in the datastore.

Cluster ID: unique identifier for a cluster's datastore
NodeID: unique identifier for the node, usually the hostname

spicedb_telemetry_dispatches (Histogram)

Histogram of cluster dispatches performed by the instance.

Method: gRPC method of the dispatch request
Cached: boolean of whether or not the dispatch was a cache hit
Cluster ID: unique identifier for a cluster's datastore
NodeID: unique identifier for the node, usually the hostname