Method retryFailedWorkers

workers/entity/controller.go:190–223

()

}

func (c *Controller) retryFailedWorkers() {
	c.mux.Lock()
	defer c.mux.Unlock()

	if !c.running {
		return
	}

	c.Entities.Range(func(key, value any) bool {
		entityID := key.(string)
		worker := value.(*Worker)

		if worker.IsRunning() {
			return true
		}

		if !c.backoff.ShouldRetry(entityID) {
			return true
		}

		slog.InfoContext(c.ctx, "retrying failed worker", "entity_id", entityID)
		if err := worker.Start(); err != nil {
			slog.ErrorContext(c.ctx, "retry failed for worker", "entity_id", entityID, "error", err)
			worker.addStatusEvent(fmt.Sprintf("retry failed for worker %s (%s): %s", entityID, worker.Entity.ForgeURL(), err.Error()), params.EventError)
			c.backoff.RecordFailure(entityID)
			return true
		}

		slog.InfoContext(c.ctx, "worker successfully started after retry", "entity_id", entityID)
		worker.addStatusEvent(fmt.Sprintf("worker successfully started after retry for entity: %s (%s)", entityID, worker.Entity.ForgeURL()), params.EventInfo)
		c.backoff.RecordSuccess(entityID)
		return true
	})
}

// loop drains the watcher consumer channel as fast as possible.
// It exits when the context is cancelled or quit is closed, triggering