n8n (workflow)

Automatisez avec n8n — Formation Complète 2026 - Home

Community Edition

  • pin data - execute the action once and cache response
  • AI Agent - driven workflow
  • filling table

see also

screenshot

# Setup

Performed by Hermes

cf n8n setup

# Client Setup ⮺

Auto-reconnect with tunnel using autossh + user service

$ systemctl --user status n8n-tunnel
$ journalctl --user -u n8n-tunnel -f

# Comparison

n8n Throughput Explained

Criteria n8n Camunda Azkaban Argo Workflows
Primary job App/automation orchestration Business process (BPM) Batch job scheduling Container/K8s workflow orchestration
Model Visual node graph BPMN 2.0 Job DAGs Steps/DAGs as Kubernetes CRDs (YAML)
Visual designer Strong Strong BPMN Basic ops UI UI to view/submit runs; design is mostly YAML/Git
Execution tracking Run history Process instances / tokens Job/flow status Per-workflow pods, logs, status in Argo UI
Triggers Webhooks, cron, events Messages, timers, APIs, humans Schedules, manual CronWorkflows, sensors (Argo Events), API, GitOps
Human tasks DIY First-class No No (not a BPM engine)
Where it runs App server / Docker App platform Hadoop/JVM cluster era Requires Kubernetes
Unit of work Connector / JS node Service task / user task Shell/Hadoop/Spark job Container (image + command)
On-prem Yes Yes Yes Yes (on your K8s)
Open source Fair-code Edition-dependent Apache 2.0 Apache 2.0
Ops weight Light Heavy Medium / aging Medium–heavy if you already run K8s; heavy if you don’t
Best fit Glue APIs & SaaS Org processes & compliance Legacy batch DAGs CI/CD, data/ML pipelines, infra jobs on K8s
Written on September 9, 2026, Last update on
workflow automation agentic-AI