Jump to content

Connect SuperML | Leeroopedia MCP: Equip your AI agents with best practices, code verification, and debugging knowledge. Powered by Leeroo — building Organizational Superintelligence. Contact us at founders@leeroo.com.

Principle:Apache Dolphinscheduler Task Reassignment

From Leeroopedia
Revision as of 18:04, 16 February 2026 by Admin (talk | contribs) (Auto-imported from principles/Apache_Dolphinscheduler_Task_Reassignment.md)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)


Knowledge Sources
Domains Distributed_Systems, Fault_Tolerance
Last Updated 2026-02-10 00:00 GMT

Overview

A task reassignment mechanism that migrates in-flight tasks from a failed worker to healthy workers within the same worker group, maintaining execution continuity.

Description

The Task Reassignment principle defines how tasks running on a crashed worker are recovered and re-dispatched. When a worker fails, the FailoverCoordinator identifies all tasks assigned to that worker using the ITaskExecutionRunnable registry. Each affected task is delegated to TaskFailover which re-queues the task to WorkerGroupDispatcherCoordinator. The dispatcher then selects a healthy worker from the same worker group using its load balancer, ensuring the task runs on an appropriate node.

Usage

Automatically triggered as part of worker failover processing. Tasks are re-dispatched to the same worker group they were originally assigned to.

Theoretical Basis

Task reassignment follows a Migration Pattern:

doWorkerFailover(workerAddress):
    tasks = getTasksOnWorker(workerAddress)
    for task in tasks:
        taskFailover.failoverTask(task)
        // TaskFailover re-queues to WorkerGroupDispatcherCoordinator
        // which selects a new healthy worker from the same group

Related Pages

Implemented By

Page Connections

Double-click a node to navigate. Hold to expand connections.
Principle
Implementation
Heuristic
Environment