Skip to main content
AmigoFish

Natural language troubleshooting in OpenShift 5 | Demo

From Red Hat · · 4 min

The player loads from YouTube only when you click. It uses YouTube's privacy-enhanced mode.

If the video does not play, the owner may have turned off embedding. Watch it on YouTube instead.

Watch on YouTube

Not rated yet

Description

You want to diagnose Kubernetes cluster anomalies before they cause service disruptions. But how? Red Hat OpenShift Lightspeed, included within Red Hat OpenShift 5, helps cluster administrators troubleshoot complex issues using natural language, accelerating root-cause analysis while reducing mean time to recovery (MTTR).

Troubleshooting enterprise clusters under pressure often forces operations teams to manually sift through massive volumes of logs, parse raw telemetry, and write complex queries. In this demonstration, Vanessa Martini, Observability Product Manager at Red Hat, shows how OpenShift Lightspeed pairs natural language investigation with the Cluster Observability Operator to resolve cluster issues interactively.

What you will see in this video:

User-initiated troubleshooting: Investigate unalerted anomalies—such as service mesh connection timeouts—by querying a Model Context Protocol (MCP) server equipped with Ops MCP observability toolsets.

Automated evidence correlation: See how the engine correlates in-cluster events and logs to build a clear causal chain, identifying root causes like misconfigured network policies without manual telemetry parsing.

Alert noise reduction: Rapidly triage active alerts, such as a firing PodDisruptionBudget limit, by isolating key signals and pinpointing remediation steps like invalid deployment image tags.

Natural language data visualization: Translate plain English prompts into accurate Prometheus Query Language (PromQL) queries to dynamically render Perses visualization charts and save them directly to investigation dashboards.

A trust-but-verify operational model: Maintain strict human-in-the-loop control with read-only AI recommendations, allowing administrators to inspect evidence summaries before manually executing cluster changes.

Timestamps: 00:00 Introduction 00:23 Architecture and operator prerequisites 00:41 Investigating unalerted network anomalies 01:14 Model Context Protocol and causal chain analysis 01:54 Active alert triage and noise reduction 02:37 Generating PromQL queries and Perses visualizations 03:19 Trust-but-verify workflow summary

Explore more resources: 💡 Discover OpenShift 5 → https://www.redhat.com/en/technologies/cloud-computing/openshift-5?sc_cid=RHCTG0260000501072 ✨ Read the release notes → https://docs.redhat.com/en/documentation/openshift_container_platform/5.0?sc_cid=RHCTG0260000501072

#OpenShiftLightspeed #RedHatOpenShift #Observability #Kubernetes #AIOps #RedHat

Back to Red Hat · Report a problem