Self-Hosted Prometheus

Implement self-hosted Prometheus and Grafana in your Cobrowse Enterprise Kubernetes cluster for enhanced monitoring and data visualization.

If you are self-hosting your Cobrowse Enterprise Kubernetes cluster, then deploying your own Prometheus and Grafana instances into your cluster may make the most sense for you.

Prometheus

If you do not have Prometheus running in your cluster already, then you may consider using the Prometheus Community Helm Chart to deploy the components necessary to get you started.

These example configurations and commands are intended to demonstrate the Helm repositories and chart resources you can use as a reference to help deploy a non-production Prometheus instance that is capable of scraping cobrowse redis cluster and service metrics.

For a production-grade deployment with appropriate authentication, persistence, and ingress, we recommend you research the Prometheus configuration options and the full Prometheus helm chart values.yaml file that are available to help you make the appropriate decisions for your organization.

There is one chart that deploys a configurable node-exporter, alertmanager and server component with ingress, which can get you started querying metrics as quickly a possible:

# values.yml
server:
  baseURL: http://<prometheus host>/prometheus/
  prefixURL: /prometheus
  ingress:
    annotations:
      kubernetes.io/ingress.class: nginx
    enabled: true
    hosts:
      - <prometheus host>/prometheus
    path: /prometheus

helm repo add prometheus-community https://prometheus-community.github.io/helm-charts
helm repo update
helm install prometheus prometheus-community/prometheus -f values.yml

Docker Compose

For Docker Compose deployments a Prometheus container can be started as part of the deployment in the following way:

docker-compose -f docker-compose.yml -f metrics/docker-compose.metrics.yml up 

Grafana

If you do not have Grafana installed in your cluster already, then may consider using the Grafana official Helm Chart to deploy the components necessary to get you started.

These example configurations and commands are intended to demonstrate the Helm repositories and chart resources you can use as a reference to help deploy a non-production Grafana server that is capable of querying a local Prometheus server.

For a production-grade deployment with appropriate authentication, persistence, and ingress, we recommend you research the Grafana configuration options and the full Grafana helm chart values.yaml file that are available to make the appropriate decisions for your organization.

There is one chart that deploys the Grafana server with ingress and configuration to query your Prometheus instance, which can get you querying metrics as quickly as possible:

# values.yml
persistence:
  enabled: true
  accessModes:
    - ReadWriteOnce
  size: 5Gi
grafana.ini:
  server:
    domain: <grafana host>
    root_url: "http://<grafana host>/grafana"
    serve_from_sub_path: true
ingress:
  annotations:
    kubernetes.io/ingress.class: nginx
  enabled: true
  hosts:
    - <grafana host>
  path: /grafana
datasources:
  datasources.yaml:
    apiVersion: 1
    datasources:
      - name: Prometheus
        type: prometheus
        url: http://<prometheus host>/prometheus
        access: proxy
        isDefault: true

helm repo add grafana https://grafana.github.io/helm-charts
helm repo update
helm install grafana grafana/grafana -f values.yml