Monitoring

Cluster monitoring via Grafana/Prometheus comes built in with QHub, and is enabled by default.

Accessing the Grafana Dashboards

The monitoring dashboards can be accessed via Grafana at: your-qhub-domain.com/monitoring. The initial login credentials are username: admin and password: prom-operator, but users should change the admin password immediately after the first log in.

More than 25 prebuilt dashboards come with Qhub. To access them, hover over the dashboards icon, then click on “Manage” as indicated in the image below.

For example, the General/Kubernetes/Compute Resources/ Node (Pods) dashboard allows you to easily see which pods are using the most compute and memory on a particular node of your kubernetes cluster.

Additionally, the General/Kubernetes/Networking/Cluster dashboard gives a nice overview of network traffic in your kubernetes cluster.

Please check out the Grafana documentation for further details on Grafana.

Disabling Cluster Monitoring

1. To disable cluster monitoring on QHub deployments, simply disable the feature flag within your qhub-config.yaml file. For example:

monitoring:
  enabled: false

previous

Day-to-day Maintenance

next

ClearML