GuideAdvanced
n8n in Production and Maintenance
By the end you operate automations that other people depend on: you know what turns a workflow into "production" and audit which ones qualify, you handle errors with five layers of defense, you debug what already failed with the replay engine, you set up observability so you can answer "what happened with order 4471?" nine days later, you harden credentials and secrets, you make a slow workflow fast, you control the cost of AI in production, and you leave behind runbooks and an on-call rotation that work even when you're not around. This guide doesn't set up the server: it assumes you already have a running instance — self-hosted or on n8n Cloud — and works on top of what's already running.
- 64
- lessons
- 8
- modules
- English · Spanish
- available in
- Yes
- certificate
- Free
- access
Outcomes
What you'll be able to do
- Audit what turns a workflow into "production": classify it by criticality, apply the readiness checklist before turning it on, and change it safely while it's live
- Handle errors with five layers of defense: automatic retries, the error branch, the Stop And Error node, global error workflows, and failure notifications
- Debug with the replay engine: reproduce past executions, trace variables in code nodes, and re-run a single node with edited input
- Set up observability: structured logging inside workflows, audit trails, execution metrics, health checks, and alerts for failures and SLA breaches
- Harden credentials and secrets: hygiene and least privilege, external secret stores, securing the instance with HTTPS, and protecting API keys and LLM tokens
- Diagnose and improve performance: where the time actually goes, batching, memory with large payloads, reducing external calls, and handling concurrency and rate limits
- Control the cost of AI in production: choosing current models, using local models at zero cost, and exposing or consuming MCP servers securely
- Write runbooks that work at three in the morning, organize on-call rotations without burning anyone out, and document blameless postmortems
Before you start
What you need to bring
It's for you if...
- Anyone already building n8n workflows who now operates them in production, with people or business processes depending on them
- Teams who need to answer "what happened with this order" weeks after an execution
- Automators using AI in production workflows who need to control spend
- Anyone going on call who needs runbooks that work without that person being present
Requirements and materials
- Knowing how to build workflows in n8n: nodes, triggers, connections, expressions, and basic credentials
- A working n8n instance, self-hosted or on n8n Cloud
- This is not an introduction to n8n: it assumes you already built automations and now want to operate them seriously
Content
The syllabus, module by module
Open any of them to see its lessons.
- 1. Module introduction: what you're about to operate
- 2. What 'production' means for an automation
- 3. Classifying workflows by criticality
- 4. The production-readiness checklist
- 5. Changing a workflow that's running
- 6. Names, tags, and workflow inventory
- 7. The shared responsibility map
- 8. Project: production-readiness audit
- 1. Module introduction: designing for failure
- 2. Anatomy of a failure: node vs workflow
- 3. Automatic retries: Retry On Fail
- 4. Error branch: Continue On Error and error output
- 5. Failing on purpose: the Stop And Error node
- 6. Global error workflows with Error Trigger
- 7. Failure notifications
- 8. Project: workflow with full error handling
- 1. Module introduction: stop guessing
- 2. Reading an execution: the data log
- 3. Execution replay: reproduce a past failure
- 4. Variable tracing in code nodes (JS/Python)
- 5. Pinned data and mocking
- 6. Re-running a single node and editing inputs
- 7. Common failure patterns and how to diagnose them
- 8. Project: debug a broken workflow
- 1. Module introduction: seeing in order to operate
- 2. Structured logging inside workflows
- 3. Traceability and auditing: who changed what
- 4. Execution metrics: success rate, duration, and volume
- 5. Health checks and uptime monitoring
- 6. Alerting on failures and SLA breaches
- 7. Shipping logs and metrics to external tools
- 8. Project: observability layer
- 1. Module introduction: the keys to the kingdom
- 2. How n8n stores credentials
- 3. Credential hygiene and least privilege
- 4. External secrets stores (Enterprise)
- 5. Self-hosted secrets without Enterprise
- 6. Securing the instance
- 7. Protecting API keys and LLM tokens
- 8. Project: harden credentials and secrets
- 1. Module introduction: the AI that won't wreck the budget
- 2. Cost control: where the money leaks
- 3. The native AI Agent node in production
- 4. Choosing current models (July 2026)
- 5. Local models with the AI Starter Kit ($0 per token)
- 6. Expose your instance as an MCP server
- 7. Consuming external MCP servers securely
- 8. Project: cost-controlled AI workflow
Common questions
What people usually ask
No limit. It's a free guide: come in whenever you like, as often as you like.
No. Modules run from easier to harder, but you can jump to the one you need. Progress is saved per lesson.
Whatever is needed is listed under “What you need to bring”, above. If nothing is listed there, you can start from zero.
In the Club's WhatsApp group, and every two weeks there's a live with an instructor where questions get worked through.
Yes. It's issued automatically once you finish every lesson, with a verifiable code you can share on LinkedIn.
Start whenever you like
What students say
These reviews are from enrolled students who completed at least 50% of the course. We moderate reviews only on content grounds (spam, offensive language, personal data), never for being critical or negative.
No approved reviews yet.
Be the first to share your experience!