core/org.meshtastic.sdk/AutoReconnectConfig

AutoReconnectConfig

data class AutoReconnectConfig(val enabled: Boolean = true, val initialBackoff: Duration = 1.seconds, val maxBackoff: Duration = 60.seconds, val maxAttempts: Int? = null, val backoffMultiplier: Double = 2.0, val jitter: Double = 0.2)(source)

Tunables for the engine's built-in auto-reconnect supervisor.

Configure on the RadioClient.Builder via autoReconnect(...).

Default: Disabled. The 1.0 release will flip the default to enabled = true; pin a value explicitly here if you want today's "host owns reconnect policy" behaviour to survive that bump.

Backoff formula: delay(n) = min(initialBackoff * backoffMultiplier^(n-1), maxBackoff) then multiplied by (1 ± jitter * random()). The supervisor uses kotlinx.coroutines.delay so virtual-time test runners (runTest / advanceTimeBy) drive the schedule deterministically.

Since

0.1.0

Constructors

AutoReconnectConfig

constructor(enabled: Boolean = true, initialBackoff: Duration = 1.seconds, maxBackoff: Duration = 60.seconds, maxAttempts: Int? = null, backoffMultiplier: Double = 2.0, jitter: Double = 0.2)

Types

Companion

object Companion

Properties

backoffMultiplier

val backoffMultiplier: Double

exponential growth factor between attempts (default: 2.0). Coerced to >= 1.0 by the init block.

enabled

val enabled: Boolean

toggle the supervisor entirely. When false the engine surfaces a single ConnectionState.Disconnected after a transport drop.

initialBackoff

val initialBackoff: Duration

delay before the first reconnect attempt (default: 1 s).

jitter

val jitter: Double

symmetric multiplicative randomization, in 0.0..1.0. 0.2 means ±20% on each computed delay. 0.0 disables jitter entirely.

maxAttempts

val maxAttempts: Int?

upper bound on reconnect attempts; null means retry indefinitely. When the cap is reached the engine emits ConnectionState.Disconnected with the original cause and stops.

maxBackoff

val maxBackoff: Duration

hard cap on backoff between successive attempts (default: 60 s).