Every metric is green-ok but your users can't log-in

Bolivar Stephen — Fri, 06 Feb 2026 22:18:46 +0000

Your monitoring says one thing. Reality says another. And you're stuck in the middle trying to figure out what's actually broken.

This is the most dangerous blind spot in modern infrastructure: the gap between what your metrics show and what your users experience.

Why Green Dashboards Lie

Here's what we tell ourselves: if CPU is low, memory is available, and HTTP is 200 OK, the system must be working.

This assumption is wrong.

Infrastructure metrics measure potential, not reality. They tell you your system could work. They don't tell you it is working.

It's like saying your car is absolutely fine because the tank is full of gas. However you have 2 flat tires and no steering wheel.

Infrastructure metrics are necessary. But they're not the big picture.

How to Build a Complete Monitoring Strategy

Combine infrastructure metrics with workflow validation:

Infrastructure layer (traditional monitoring):

CPU, memory, disk, network utilization
Process health checks
Resource saturation metrics

Network layer:

TCP port connectivity
DNS resolution
TLS handshake success
Certificate expiry

Application layer:

HTTP response codes
API endpoint availability
Response time percentiles

Business logic layer (workflow monitoring):

User registration completes end-to-end
Login → session → data fetch works
Checkout → payment → confirmation succeeds
Password reset emails actually send

Each layer catches different failure modes. Infrastructure metrics catch capacity issues. Network checks catch connectivity problems. Application metrics catch crashes. Workflow checks catch the subtle breaks where everything looks healthy.

Start With Your Critical Path

You don't need to monitor every possible user journey. Start with the one workflow that would cause panic if it broke. For example Registration and Main value proposition.

Then build a basic check that verify this workflow.
Can a user actually create an account ?
Can a user actually click on that button and it does do what it's supposed to ?

Shift from "are our servers healthy?" to "can users accomplish what they came here to do?"

Conclusion

Your infrastructure metrics will tell you when capacity runs low. When processes crash. When disk fills up.

They won't tell you when authentication tokens expire. When APIs return errors wrapped in 200 responses. When background jobs stop processing.

If you want to know whether your system actually works, test it the way users experience it. Try to do what they do. Verify it works end-to-end.

There's a difference between monitoring infrastructure and monitoring user experience, and that's why I built Monitrics

Send automatic release notes email from Gitlab

Bolivar Stephen — Wed, 09 Jun 2021 18:09:37 +0000

The problem

As a tech lead I'm required to do some communication with the company. Until today, I was sending a recap email about new features and bugfixes on every release. But, I tend to "forget" because it's tedious and boring.

Let automate this !

What you will need:

Sendgrid. Free tier is sufficient
Gitlab

n8n to the rescue

Extendable workflow automation. That's the tagline. Simple, explicit. I like it. I need a no-code platform that can do things for me, but which also offers me the possibility to run code. As an experienced JS/TS developer n8n seems perfect. Let's dive in.

Installation

They provide a Docker image, which is always handy. I'm going to set this up in my cluster, to my liking. Here is a compose file to get you started with docker and traefik 1.7.

version: '3.7'

services:
  n8n:
    image: n8nio/n8n:0.123.1
    environment:
      GENERIC_TIMEZONE: "America/Port_of_Spain"
      TZ: "America/Port_of_Spain"
      N8N_BASIC_AUTH_ACTIVE: true
      N8N_BASIC_AUTH_USER: <CHANGE ME>
      N8N_BASIC_AUTH_PASSWORD: <CHANGE ME>
    volumes:
      - n8n-data:/home/node/.n8n
    networks:
      traefik-public:
      default:
    deploy:
      replicas: 1
      resources:
        limits:
          memory: 512M
      labels:
        - 'traefik.enable=true'
        - 'traefik.backend=n8n'
        - 'traefik.frontend.rule=Host:host.domain.tld'
        - 'traefik.docker.network=traefik-public'
        - 'traefik.port=5678'
        - 'traefik.tags=traefik-public'
        - 'traefik.redirectorservice.frontend.entryPoints=http'
        - 'traefik.redirectorservice.frontend.redirect.entryPoint=https'
        - 'traefik.webservice.frontend.entryPoints=https'
networks:
  traefik-public:
    external: true
volumes:
  n8n-data:

You can also use a cloud hosted version of n8n if setting this up on your own is not something you're interested in.

Now that everything is installed. It's time to dive in what we actually want to do. Which is send and email to the team whenever a new release is published on Gitlab.

I prepared the workflow. Import this json workflow in n8n. And let's make it work on your setup.

After import, you should find a workflow like this.

(Ignore the start button. We're not going to use it).

1. Gitlab webhook

Edit the Webhook node and change the path parameter to a random value, we'll reference this value as $path later.
Go to your project Webhook settings on Gitlab and create a new webhook at the url https://yourdomain.tld/webhook/$path Perfect. Next.

2. Gitlab Markdown to html

The webhook release note is in Markdown. Not very mail friendly if you ask me. Let's convert it to HTML.
Conveniently, Gitlab offers an api for that.
That's the convertMdToHtml node. Let's setup our gitlab api key in n8n.

Create a new Header auth credential that you must name gitlab-api-key.
The header name is "Authorization" and the value is Bearer <your_api_key>. If you don't know how to create a gitlab api key, head over to the wonderful docs.

3. Finally, send the mail.

Create a new api key in Sengrid. Copy it to your clipboard.

Now create a new Sendgrid credential in n8n. You must name it sendgrid-api-key.

Start your workflow. You're done !

DEV Community: Bolivar Stephen