Skip to main content

Error Strategies

Control what happens when a task fails.

Strategies​

StrategyBehavior
StopAll (default)Fail fast, cancel everything
ContinueSkip dependents, keep running independent tasks
Retry(n)Retry n times immediately before failing
Retry(n, WithRetryBackoff(...))Retry n times with exponential backoff

Usage​

Set at plan level or override per-task:

// Plan-level: don't halt on failure.
plan := orchestrator.NewPlan(client,
    orchestrator.OnError(orchestrator.Continue),
)

// Task-level override: immediate retry.
task.OnError(orchestrator.Retry(3))

// Task-level override: retry with exponential backoff.
task.OnError(orchestrator.Retry(3,
    orchestrator.WithRetryBackoff(1*time.Second, 30*time.Second),
))

With Continue, independent tasks keep running when one fails. With StopAll, the entire plan halts on the first failure.

Failure Recovery​

Use a When guard to trigger recovery tasks when an upstream fails:

alert := plan.TaskFunc("alert",
    func(_ context.Context, _ *client.Client) (*orchestrator.Result, error) {
        return &orchestrator.Result{Changed: true}, nil
    },
)
alert.DependsOn(mightFail)
alert.When(func(results orchestrator.Results) bool {
    r := results.Get("might-fail")
    return r != nil && r.Status == orchestrator.StatusFailed
})

Examples​

See examples/sdk/orchestrator/features/error-strategy.go for a complete working example. For failure-triggered recovery, see Failure Recovery.