Red Hat OpenShift 4.2 is a significant release that brings a number of great enhancements to the Web Console UI, but you'll notice one of the biggest changes as soon as you log in.

The Cluster Overview Dashboard is the new default landing page of the OpenShift Console and provides a birds-eye view of your cluster's current health, inventory, capacity, utilization, and activity to make identifying problems and resolving issues easier and faster.

This post will briefly cover what this dashboard is made of, but we know from using it ourselves these past few months that static screenshots won't quite do it justice. We're really excited for you to try this new dashboard out in your own clusters, and our User Experience Design team would love to hear any feedback and suggestions you have for future improvements.

A window into your cluster

The Overview Dashboard is designed to answer the following at a quick glance:

  1. What am I looking at?
  2. What's in this cluster?
  3. Is everything okay?
  4. Is there enough capacity?
  5. What is the cluster up to?

The Details card in the top-left corner answers the first question with the cluster’s ID, provider, and current version. The Inventory card directly below it includes the quantity and current statuses of the nodes, pods, and persistent volume claims within the cluster, with clickable status counters that act as quick shortcuts to the full list pages of those objects.

The Health card in the center of the dashboard is (hopefully) pretty mundane most of the time, with a green check mark indicating that no degraded systems or actively-firing alerts from AlertManager require your attention.

The Capacity and Utilization cards visualize how much resource headroom is available at the moment and how the cluster's utilization of CPU, Memory, and Disk space has changed over the last hour.

The Events card in the top-right streams in the same cluster-wide events that can be found in the dedicated Events page, with any warning events highlighted.

Finally, the Top Consumers card in the bottom-right corner helps identify the highest consumers of CPU time, Memory, Storage I/O time, and Network bandwidth so that any outliers greatly affecting the cluster's capacity can be addressed.

Surfacing alerts, warnings, and errors

A perfectly happy and healthy dashboard is always great to see, but the value of including all of this information in one place becomes super clear when things in the cluster go wrong.

Any firing alerts from AlertManager are collected within an Alerts section of the Health card. Resolving those alerts is generally the best first step toward restoring the health of the cluster, but the additional context provided by the object statuses in the Inventory card, errors in the Events card, and resource measurements in the Capacity and Utilization cards can often help to better understand the full scope of each problem.

Funnily enough, we know this because the dashboard started to help our internal OpenShift developers identify and troubleshoot issues almost as soon as the first cards appeared! The errors that bubbled up to the dashboard frequently kicked off discussions that tangibly improved the entire OpenShift product family, and seeing it do so was an absolute thrill (that we probably should have expected).

We can't wait to hear how it helps you and your team as well!

What’s next

The introduction of the Cluster Overview Dashboard in OpenShift 4.2's Console UI is just one of the many ways we're improving the observability and management experience of OpenShift, and there's so much more ahead to look forward to. If you'd like a sneak peek, follow the OpenShift Console and OpenShift Design GitHub repositories to see some of the ideas we're thinking about for future iterations.

Our User Experience Design team would love to hear your ideas and feedback on this new dashboard and any other areas of the Console UI. Please fill out this feedback survey or sign up to participate in future research opportunities.

 


About the author

Red Hatter since 2018, tech historian, founder of themade.org, serial non-profiteer.

Read full bio