Add Prometheus and docker-compose

[Iajrdev]

This PR adds Prometheus and Grafana to the project for monitoring purposes.

• Added docker-compose.yml for setting up Prometheus and Grafana.
• Configured Prometheus to scrape metrics from the Go application.
• Updated documentation with setup instructions.

[alipourhabibi]

I reviewed the PR
I found an error in grafana/provisioning/dashboards/my_dashboard.json
It looks like the json is not valid for dashboard.
I think you should remove the following

 {
-  "dashboard": {
     "id": null,
     "title": "My App Dashboard",
     "tags": [],
@@ -92,7 +91,5 @@
         ]
       }
     ]
-  },
-  "overwrite": false
 }

Apart from that all looks good to me.

[alipourhabibi]

LGMT

[armineyvazi]

global:
  scrape_interval: 5s

scrape_configs:

  - job_name: 'myapp'
    metrics_path: '/v2/metrics'
    static_configs:
-     - targets: ['172.17.0.1:8080']
+     - targets: ['host.docker.internal:8080']

[armineyvazi]

I found a mistake in prometheus/prometheus.yml. It looks like the target IP is outdated. I think you should change the following

[alipourhabibi]

As i understand the 172.17.0.1 refers to the IP address of the default gateway on the default bridge network and is not outdated. However the host.docker.internal is used in Docker desktop and won't work on Linux systems without additional configuration.
If we want to support Windows and Linux together we can add the following to the docker-compose.yaml for prometheus

    extra_hosts:
      - "host.docker.internal:host-gateway"

and use host.docker.internal as our targets in prometheus/prometheus.yml
What do you think?
@Iajrdev

[armineyvazi]

Thank you for completing this. I think this configuration is okay for Linux, macOS, and Windows OS:

prometheus/prometheus.yml

+     - targets: ['host.docker.internal:8080']

docker-compose.yml

+  extra_hosts:
+    - "host.docker.internal:host-gateway"

[Iajrdev]

To clarify, you're suggesting that we should remove '172.17.0.1:8080' from the prometheus.yml file and replace it with 'host.docker.internal:8080'?
@alipourhabibi @armineyvazi

[armineyvazi]

To clarify, the suggestion is to replace 172.17.0.1:8080 with host.docker.internal:8080 in the prometheus.yml file. Additionally, we should add the following line to the docker-compose.yaml file to ensure compatibility across both Windows/Mac and Linux environments:

extra_hosts:
  - "host.docker.internal:host-gateway"

This configuration will make host.docker.internal resolvable within Docker containers on all platforms.

@alipourhabibi @Iajrdev

[Iajrdev]

I have made changes to the files as per your review and pushed them.
Please review them again.

[armineyvazi]

I am not sure, but I think it should look like this, not added in Grafana:

  prometheus:
    image: docker.arvancloud.ir/prom/prometheus:latest
    container_name: prometheus
    volumes:
      - ./prometheus/prometheus.yml:/etc/prometheus/prometheus.yml
      - prometheus_data:/prometheus
    ports:
      - "9090:9090"
+  extra_hosts:
+   - "host.docker.internal:host-gateway"
    restart: unless-stopped

If you have a Linux system, please test this configuration to ensure it works correctly. The extra_hosts directive ensures that host.docker.internal is resolvable within the Prometheus container. Add the Grafana service if needed, without extra_hosts unless required for Grafana.

@Iajrdev

[armineyvazi]

Comment to Approve:

Approved. The configuration is now compatible with Linux, macOS, and Windows. Adding host.docker.internal:host-gateway to extra_hosts ensures that the Docker container can correctly reference the host machine across different operating systems.

Thanks for the update!

[Iajrdev]

• Added healthcheck for Prometheus:

  • Implemented a health check for the Prometheus service to ensure it is ready before proceeding with other services.
  • Health check uses the command: ["CMD", "curl", "-f", "http://localhost:9090/-/ready"].

• Configured depends_on for Grafana:

  • Updated the Grafana service to depend on the Prometheus service being healthy.
  • This guarantees Grafana will only start after Prometheus has successfully passed its health check.

• Added healthcheck for Grafana:

  • Implemented a health check for the Grafana service to monitor its readiness.
  • Health check uses the command: ["CMD", "curl", "-f", "http://localhost:3000/api/health"].

@alipourhabibi @armineyvazi
Please review again.

[alipourhabibi]

LGTM