n8n + Argo = ❤️: Offloading CPU-Heavy Steps to Argo Workflows

说明与用途

在 n8n 中，CPU 密集型 Code 节点会阻塞工作器并导致任务中止。将这些任务卸载到 Argo Workflows 可以释放工作器、消除失败并保持队列短，但整体作业时间会更长。

问题

CPU‑heavy Code nodes in n8n block worker slots and cause task aborts.

结果

Offloading CPU‑heavy steps to Argo frees n8n workers, eliminates task aborts, and keeps the queue backlog low, though overall job time increases due to serial execution within the runner.

工作流

• Webhook triggers n8n workflow
• HTTP Request node submits Argo WorkflowTemplate with job parameters and n8n resume URL
• Wait node pauses execution until webhook callback
• Argo pod runs heavy job in node:26‑alpine container
• onExit handler posts status to n8n resume URL
• n8n resumes execution after callback

证据

A CPU‑heavy step does not belong in an n8n worker's Node.js event loop. Let n8n hand the job to Argo, release its worker slot with a Wait node, and resume when the Argo pod calls back.

来源

At 30 CPU‑seconds per job, 9 of the 10 executions failed with Task execution aborted because runner became unresponsive.

来源

The offloaded workflow has three nodes. The flow: Submit, Wait, Callback.

来源