computeModuleRuns

List compute-module runs (the monitoring + audit trail of executions), newest first, optionally filtered by module and/or outcome. Rows are written by the runtime; empty until modules execute (Phase 2). Requires the org 'view_compute_diagnostics' permission.

computeModuleRuns(
  appId: BigInt!
  moduleName: String
  success: Boolean
  limit: Int
  offset: Int
): [WasmModuleRun!]!

Arguments

computeModuleRuns.appId ● BigInt! non-null scalar

The app (tenant) whose runs to list.

computeModuleRuns.moduleName ● String scalar

Optional: only runs of this module. Omit for all.

computeModuleRuns.success ● Boolean scalar

Optional: filter by outcome (true = succeeded, false = failed).

computeModuleRuns.limit ● Int scalar

Max runs to return (default 50, max 200).

computeModuleRuns.offset ● Int scalar

Number of runs to skip (pagination).

Type

WasmModuleRun object

One recorded compute execution: timing, fuel, host-call counts, outcome. NOT a sample of all executions — a successful tick is deliberately not recorded here, because ticks are high-frequency and aggregate into per-minute usage instead, so this table holds every failure plus demand-driven runs plus the reload that follows a terminated worker. Counting rows here to get a success rate gives roughly 50% against a true 0.1-1.6%; computeModuleStats and computeAppDiagnostics read the per-minute rollups for that. Use this for what happened on an individual run and for flow correlation.