Metrics Need Shared Dashboards? Connect Managed Grafana to Prometheus

A shared dashboard gives a team the same view of system behavior. Azure Managed Grafana hosts the visualization service, while Azure Monitor workspace stores the Prometheus time series. Grafana queries those metrics using its managed identity.

Reuse amw-ctmonitorweu and the running AKS cluster from Monitor AKS. Confirm up already returns data there. Creating Managed Grafana introduces a separate service charge; review its tier and cost before deploying.

Create and Connect Grafana

Create Azure Managed Grafana in the Azure portal:

Resource group: rg-cloudtrips-monitor-grafana-test-weu
Name: graf-ctmonitorweu
Region: West Europe
Pricing plan: Standard
Zone redundancy: Disabled for this lab
System-assigned managed identity: On
Include myself under Grafana administrator role: Selected

Clear the option that automatically assigns Monitoring Reader on the target subscription. This lab only needs access to one metrics workspace. Create the resource, then open amw-ctmonitorweu → Access control (IAM) and assign:

Role: Monitoring Data Reader
Assign access to: Managed identity
Member: graf-ctmonitorweu
Scope: this Azure Monitor workspace

Open graf-ctmonitorweu → Integrations → Azure Monitor workspaces → Add, select amw-ctmonitorweu, and add it. Wait for role propagation. The integration creates the Prometheus data source inside Grafana.

Screenshot 1: Managed Grafana integration listing amw-ctmonitorweu.

Build a Dashboard

From Grafana’s Azure Overview page, open its Endpoint and sign in. In Grafana, open Explore, choose the newly added Prometheus data source, switch the query editor to Code, and run up for the last 30 minutes. Series appearing here confirm both the connection and read permission.

Create a dashboard through Dashboards → New → New dashboard → Add visualization. Choose that data source and use:

sum by (job) (up)

Choose Time series, set the panel title to Successful scrape targets by job, and apply/save the panel. Save the dashboard as CloudTrips AKS monitoring. Set its time range to Last 30 minutes and refresh interval to 1 minute.

Each plotted value counts successful scrape targets for one job. It describes monitoring target availability. To investigate application CPU, memory, or pod state, open the Kubernetes dashboards installed with the integration and select your cluster and namespace.

Screenshot 2: Saved CloudTrips AKS monitoring dashboard with Prometheus data.

Share the View

In the Azure portal, open graf-ctmonitorweu → Access control (IAM). Assign a teammate or test user Grafana Viewer at this resource’s scope. Copy the saved dashboard’s URL and have that user sign in to view it. A Viewer can read dashboards; Grafana Editor permits dashboard editing, and Grafana Admin permits administration. The user accesses Grafana with their identity; the data source accesses Prometheus with Grafana’s managed identity.

If you have no second user, review the role assignment options and keep your own Admin role. The dashboard URL still opens your saved view. A copied URL alone grants no access.

Finish

After testing, delete rg-cloudtrips-monitor-grafana-test-weu if you no longer need Managed Grafana. Remove its Monitoring Data Reader assignment from amw-ctmonitorweu if it remains. Delete the dedicated AKS lab resource group when finished with that cluster. Keep shared monitoring resources only for the remaining Trips.