Interface TickSchedulerConfig

interface TickSchedulerConfig {
    maxDrainRounds?: number;
    maxSteps?: number;
    microtaskChainDepthLimit?: number;
    telemetry?: TickSchedulerTelemetryConfig;
    urgentStepCap?: number;
}

Index

Properties

maxDrainRounds? maxSteps? microtaskChainDepthLimit? telemetry? urgentStepCap?

Properties

`Optional` `Readonly`maxDrainRounds

maxDrainRounds?: number

Drain-round cap:

Bounds the number of drain rounds while capturing concurrent commits before committing the tick snapshot.
Exceeding the cap is treated as a cycle (stable=false, degradeReason=cycle_detected).

`Optional` `Readonly`maxSteps

maxSteps?: number

Fixpoint step cap:

Steps count "work acceptance units" within a single tick, not time.
Exceeding the budget triggers a soft degrade (stable=false), deferring nonUrgent backlog to the next tick.

`Optional` `Readonly`microtaskChainDepthLimit

microtaskChainDepthLimit?: number

Microtask starvation protection threshold:

Counts consecutive ticks scheduled on microtask boundaries without yielding to host (best-effort).
Exceeding the limit forces the next tick to start on a macrotask boundary.

`Optional` `Readonly`telemetry

telemetry?: TickSchedulerTelemetryConfig

Optional degraded-tick telemetry (opt-in, sampled):

Runs even when diagnostics=off (Devtools disabled).
Intended for production health signals (frequency of stable=false / forced yield).

`Optional` `Readonly`urgentStepCap

urgentStepCap?: number

Urgent safety cap:

Even urgent work may be cut when the system appears to be in a cycle (avoid freezing).