Boost your cluster operations with Deployment Validation and Insights Advisor for Workloads

Hey admins, are app developers crashing your clusters? Do they cost you additional operations toil while fixing performance and scalability issues? There's a new and exciting way that Red Hat Insights for Red Hat OpenShift can help you address these issues.

What is a Deployment Validation Operator?

Deployment Validation Operator (DVO) is an operator executing kube-linter workload checks and reports compliance failures to Insights. Red Hat has offered this capability for managed OpenShift clusters for more than 6 months and, after careful evaluation and extending features to enable cluster administrators to more flexibly tune cluster policy, it’s offered now to on-premise deployments as well.

Do you prefer not running containers with privileged escalation enabled? Do you want to restrict application resource consumption by limiting requests and limits? What about network policies?  DVO allows you to run a server fully disconnected by allowing its metrics to be forwarded to in-cluster Prometheus instead of sending data to Red Hat. Of course, anonymization and obfuscation of data sent to Red Hat is available, as usual, with Red Hat Insights.

How can I use Deployment Validation Operator?

To take the full advantage of the Deployment Validation Operator, Remote health monitoring must be enabled.

Additionally, the Deployment Validation Operator must be installed from the Red Hat Catalog.

Once installed, results of the analysis of workloads are available at https://console.redhat.com/openshift/insights/advisor/workloads.

What problems does Deployment Validation Operator solve?

Modern development tools provide programmers with flexibility to scale their apps, and to leverage the cloud to build more highly available cloud native applications. But there are many mistakes and potential issues that can result in exactly the opposite, leading to reduced performance, reduced availability, and even reduced application security.  Here are some of the examples that Deployment Validation Operator can help your development team follow Red Hat's recommended best practices:

• Not setting resource requests
• Not adding application health checks
• Overprivileged applications
• Lack of monitoring/logging
• Using default namespaces for all objects
• Missing/Wrong security configurations
• Missing or improper pod disruption budget definitions

A full list is available in the Stackrox Git repository.

Insights Advisor for Workloads

As presented in Red Hat's Hybrid Cloud Console, Insights Advisor for Workloads takes advantage of deployment validation and allows OpenShift administrators to see all runtime  violations of deployment policies.

The UI allows for easier identification of violating components to the level of individual containers, pods, kinds, daemonsets, and so on.

As usual with Red Hat Insights, not only are potential issues flagged, users are also presented with specific guidance on how to address these issues. Documentation on best practice is also provided.

Summary

Deployment Validation paired with Insights Advisor for Workloads can help you review bad practices in a running Deployment so you can better spot issues caused by misconfigured workloads. This can help to prevent cluster performance and stability issues as well as potential security issues exposed by wrong configuration.

We welcome any feedback that can help improve our products. Send your questions and suggestions to insights@redhat.com or directly using the feedback button on console.redhat.com.