Question 231923 in Docker Question posted in
The official documentation can be found here.

How to use prometheus to monitor all containers in a docker-compose

I have a docker-compose as the following:

version: '2'
services:

  rabbitmq:
    image: rabbitmq:management
    ports:
      - 5672:5672
      - 15672:15672
    networks:
      - my-network

  zipkin:
    image: openzipkin/zipkin
    ports:
      - 9411:9411
    networks:
      - my-network

  redis:
    image: redis
    ports:
      - 6379:6379
    networks:
      - my-network

  ts-ui-dashboard:
    build: ts-ui-dashboard
    image: ts/ts-ui-dashboard
    restart: always
    ports:
      - 80:8080
    networks:
      - my-network

  ts-route-service:
    build: ts-route-service
    image: ts/ts-route-service
    restart: always
    ports:
         - 11178:11178
    networks:
      - my-network

  ts-route-mongo:
    image: mongo
    networks:
        - my-network
  ...
networks:
    my-network:
      # driver: overlay
      driver: bridge

It contains 68 containers and I want to use Prometheus to monitor all of them.

I follow the docker official tutorial to deploy the Prometheus service but it only shows two targets(i.e., docker and prometheus).

I guess here "docker" refers to the whole docker-desktop but what I need is the detailed metrics of each container in my docker-compose.

What’s the potential way to implement it? Ans as mentioned above there are some containers that are MongoDB, do I need to config Prometheus for them?

Tags:

2

Answers


  1. Chosen as BEST ANSWER
    0

    It can be solved by using cadvisor+prometheus. I follow the tutorial here and deploy it successfully.


  2. 0

    To monitor all containers in your Docker Compose setup using Prometheus, you need to configure Prometheus to scrape metrics from each individual container. Here’s a general outline of the steps you can take:

    1. Instrument Your Containers:
      Ensure that the containers you want to monitor expose Prometheus-compatible metrics. This typically involves configuring the applications within your containers to expose metrics in a format that Prometheus can understand. Many applications have built-in metrics endpoints, and for others, you might need to use exporters.

    2. Configure Prometheus:
      In your Prometheus configuration file (usually named prometheus.yml), you will define your targets and scraping configuration. Each service/container you want to monitor will have its own scrape job configuration. Here’s an example of how you can configure Prometheus to scrape metrics from your services:

      global:
  scrape_interval: 15s

scrape_configs:
  - job_name: 'ts-ui-dashboard'
    static_configs:
      - targets: ['ts-ui-dashboard:8080'] # Replace with actual container hostname and port

  - job_name: 'ts-route-service'
    static_configs:
      - targets: ['ts-route-service:11178'] # Replace with actual container hostname and port

  # Add similar scrape job configurations for other containers...

    3. Container Hostnames:
      The target hostnames (ts-ui-dashboard, ts-route-service, etc.) in the Prometheus configuration should match the service names defined in your Docker Compose file. Docker’s built-in DNS resolution will ensure that Prometheus can reach these containers.

    4. MongoDB and Other Databases:
      For databases like MongoDB, you can use exporters that expose database-specific metrics in a Prometheus-compatible format. For MongoDB, you can use the mongodb_exporter exporter. Similar to other services, you’ll need to define a scrape job for the MongoDB exporter in your prometheus.yml file.

    5. Scrape Interval:
      Adjust the scrape_interval in the global section of your Prometheus configuration to control how often Prometheus scrapes metrics from the configured targets.

    6. Start Prometheus:
      Run Prometheus with the configuration you created using a command like prometheus --config.file=prometheus.yml.

    7. Access Prometheus UI:
      After starting Prometheus, you can access its web interface at http://localhost:9090 (if you’re running Prometheus locally).

    8. Explore and Build Queries:
      Use the Prometheus UI to explore metrics and build queries to monitor your services. You can create graphs, alerts, and dashboards to visualize the collected data.

    Remember that each container you want to monitor needs to have the necessary metrics available and accessible over the network. You might need to adjust your application configurations to enable metrics endpoints or use exporters for some services.

    Additionally, make sure that the ports you use for metrics scraping are exposed and reachable within your Docker network.

    Note: The Prometheus configuration file and the setup might vary based on your specific setup and the metrics endpoints provided by the services you are using. Always refer to the official documentation of your applications and Prometheus for detailed setup instructions.

    Login or Signup to reply.
Please signup or login to give your own answer.

Back To Top
Search