JupyterHub Default Dashboards

JupyterHub Default Dashboards#

Screenshot of a JupyterHub Grafana dashboard

The “Activity” Grafana dashboard.#

The JupyterHub Default Dashboards folder shows dashboards deployed from the upstream jupyterhub/grafana-dashboards project.

  1. Activity

    This dashboard contains information about the number of running servers, daily, weekly and monthly active users.

  2. Cluster Information

    Contains panels with node-level usage such as:

    • node count

    • memory

    • cpu

    • running users per hub in cluster

    • pod health

  3. Home Directory Usage

    See the disk space occupied by user home directories, as well as last modified and number of files.

  4. JupyterHub Dashboard

    Hub usage stats and hub diagnostics, such as

    • software images pulled by user pods

    • user CPU usage distribution

    • user memory usage distribution

    • server start times

    • hub response latency

    There is also a Panel section about Anomalous user pods where pods with high CPU usage or high memory usage are tracked.

  5. NFS usage, NFS server and Prometheus server diagnostics

    This provides info about the NFS usage diagnostics, such as iops, and monitors things like CPU, memory, disk and network usage of the NFS server and Prometheus instance.

  6. Usage Report

    This provides a report about the memory requests, grouped by username, for notebook nodes and dask-gateway nodes. It also provides a graph that monitors GPU requests per user pod.

  7. User Diagnostics

    This displays cpu and memory usage and requests per user, as well as home directory usage.

  8. User Group Diagnostics

    This displays cpu and memory usage and requests per user group, as well as home directory usage. Only applies if users are assigned user groups at the authentication layer.

Making changes to Grafana dashboards#

If you make any changes to the pre-configured dashboards in the JupyterHub Default Dashboards and Cloud Cost Dashboards folders, then they will not be saved.

We encourage you to create new dashboards outside of these folders (however the configurations will not be backed up), or even contribute to:

Resources#

Grafana is a powerful open source data visualization tool with many features. For more information on how to use Grafana, refer to the Grafana documentation.