Documentation Grafana Cloud Observe and act Observability home page

Observability home page

The Observability home page is the landing page for the Observability section in Grafana Cloud. It brings together your services, databases, frontend applications, Pods, and nodes in one view so you can see what needs attention across your entire stack without switching between individual products.

The home page is powered by the knowledge graph, which automatically discovers entities from your telemetry and surfaces insights when something is wrong. Each entity type appears in its own section with key metrics, an insight ring showing severity, and an attention badge when entities have active issues. You can customize which sections appear, reorder them, and adjust the time range to focus on what matters most to your team.

Before you begin

To use the Observability home page, you must have:

• A Grafana Cloud stack with the knowledge graph activated.
• At least one data source sending telemetry that the knowledge graph can discover (for example, Application Observability, Kubernetes Monitoring, or Cloud Provider Observability).
• Viewer or higher access to Grafana Cloud.

Open the Observability home page

In the Grafana main menu, select Observability. The home page is the default landing page for this section.

Navigate the page layout

The home page is organized into the following areas:

Assistant prompt

At the top of the page, an embedded Assistant prompt lets you ask questions about your environment directly from the home page, for example, “Why is payments-svc returning errors?”

Select Ask to open a conversation with the Assistant without leaving the page.

Toolbar

Below the Assistant prompt, the toolbar provides controls to:

• Customize: Opens the page editor to add, remove, or reorder sections.
• Navigation arrows and zoom: Step forward and backward through time, or zoom in and out on the current time range.
• Time range: Adjusts the time window for all sections on the page. The default is Last 1 hour.
• Refresh: Manually refreshes all section data.
• RCA Workbench: Opens RCA workbench with a count of active investigations.

Entity sections

Each section displays a table of entities for a given type. Sections appear in the following default order:

1. Frontend (RUM): Web applications monitored with Frontend Observability, showing LCP and page loads.
2. Services: Application services discovered by the knowledge graph, showing technology, latency (p95), error ratio, and request rate.
3. Databases: Database entities showing technology, latency avg, error ratio, and request rate.
4. Pods: Kubernetes Pods showing CPU cores and memory.
5. Nodes: Kubernetes nodes showing technology, CPU, memory, and disk utilization percentages.

Each section includes:

• Insights column: An insight ring showing the severity of active insights (red for critical, yellow for warning, blue for informational).
• Need attention badge: A count of entities with active insights that require investigation.
• View all: A link that opens the full entity catalog filtered to that entity type.
• RCA Workbench actions: Hover over an entity row to select + RCA Workbench to add the entity to an investigation. If the entity is already in the active investigation, select Remove to remove it.

Customize the home page

To change which sections appear on the home page and how they’re arranged:

1. Select Customize in the toolbar. The page enters edit mode and a Widget settings panel opens on the right.
2. To add a section, select the + button next to a section type in the panel. Available types are grouped by category, for example, Services, Service instances, and External services under Services, or Nodes, Processes, and Volumes under Infrastructure.
3. To reorder sections, drag them on the canvas.
4. To remove a section, select it on the canvas and remove it from the panel.
5. Select Save to apply your changes, or Cancel to discard them.

To return to the default layout, select Restore defaults while in edit mode.

Your customizations persist across sessions so you see the same layout each time you return.

Triage entities that need attention

When a section displays a need attention badge, entities within that category have active knowledge graph insights. To investigate:

1. Look at the insight rings in the Insights column to identify which entities have critical (red) or warning (yellow) issues.
2. Select an entity name to open its entity detail page with metrics, logs, and traces.
3. Alternatively, select View all to open the entity catalog filtered to that entity type, where you can apply additional filters.

Next steps

• Explore the entity catalog to filter and search across all discovered entities.
• Use RCA workbench to investigate incidents involving multiple entities.
• Configure the knowledge graph to adjust entity discovery and insight thresholds.