Monitor metrics endpoints on services about dockprom HOT 5 CLOSED

I ended up getting this working great using dns-sd... if its useful to put in, add to in the docs (or for anyone googling..)

Before starting dockprom I add a little extra to the config

    echo "
  - job_name: 'services'
    scrape_interval: 5s
    dns_sd_configs:
      - names:
          - scrape.me.prometheus
        type: 'A'
        port: 9999
    " >> prometheus/prometheus.yml

Then, on each of my services in docker-compose that I want to be monitored i provide the metrics at ":9999/metrics" and add the right alias...

  my-service:
    image: "super-special-service:latest"
    networks:
      default:
        aliases:
         - scrape.me.prometheus

And add prometheus to the network docker-compose made for my app....

docker network connect myapp_default prometheus

And, done!

from dockprom.

Hello, I have a question, I have multiple services from the same web application but each is a different environments, I'm using this config:

  - job_name: 'app'
    metrics_path: /services/app/metrics
    scheme: https
    scrape_interval:  5s
    tls_config:
      insecure_skip_verify: true
    dns_sd_configs:
    - names:
      - 'tasks.app'
      type: 'A'
      port: 443

How can I get like a Swarm Service Label || Container Label || Task Label in order to differentiate each task?

from dockprom.

Your config is similar to what I've used for Docker Swarm in this project swarmprom.

How will you differentiate the services since the job name will be the same for all? The containers IPs will change after each deploy.

from dockprom.

Hmm. I... don't know. Something fun to play with next.

from dockprom.

My approche was this https://github.com/stefanprodan/swarmprom#monitoring-applications-and-backend-services

from dockprom.