Metric Explorer provides an interactive query interface for your Managed Service for Prometheus instances. Use it to select metrics, write Prometheus Query Language (PromQL) statements, or combine both approaches to view, verify, and troubleshoot your monitoring data directly in the console.
With Metric Explorer, you can:
Browse and select metrics without writing PromQL
Write and run custom PromQL queries for advanced analysis
Verify that monitoring targets are reachable and reporting data
Quick start
To verify that all monitoring targets are reachable, open Metric Explorer and run the following built-in query:
upA value of 1 means the target is reachable. A value of 0 means the target is down.
For more PromQL queries, see Query metrics.
Prerequisites
Before you begin, make sure that you have:
Activated Application Real-Time Monitoring Service (ARMS)
Open Metric Explorer
Log on to the Managed Service for Prometheus console.
In the left-side navigation pane, choose , then click the name of your target instance.
In the left-side navigation pane, click .
Click the Metrics Explorer tab.
Query metrics
On the Metrics Explorer tab, you can query metrics in two ways:
| Method | When to use |
|---|---|
| Select from list | Browse available metrics and select one from the metric list. Best when you are exploring available data or are unfamiliar with PromQL. |
| Write PromQL | Enter a PromQL statement in the query editor for filtering, aggregation, and arithmetic on metrics. Best for advanced analysis and custom calculations. |
After you select a metric or enter a PromQL statement, click Run Query to display the results.
For the full PromQL syntax reference, see the Prometheus querying documentation.
Common query examples
The following examples demonstrate typical monitoring tasks you can perform with Metric Explorer.
Check target reachability
upReturns 1 for reachable targets and 0 for unreachable targets. Use this query to quickly verify that your monitoring agents are connected and reporting.
Calculate per-second request rate
rate(http_requests_total[5m])Returns the per-second rate of HTTP requests averaged over the last 5 minutes. Use this query to monitor traffic patterns and detect spikes.
Monitor memory usage percentage
(1 - (node_memory_MemAvailable_bytes / node_memory_MemTotal_bytes)) * 100Returns the memory usage percentage for each node. Use this query to identify nodes that are running low on memory.
Count active time series by job
count by (job) ({__name__=~".+"})Returns the number of active time series grouped by job label. Use this query to understand cardinality and identify jobs that produce a high volume of metrics.
Replace metric names in the examples above with the actual metric names reported by your monitoring targets. The available metrics depend on your configured exporters and instrumentation.
What to do next
After you query and verify your metrics, consider these follow-up tasks:
Create alerts: Set up alert rules based on your PromQL queries to receive notifications when metrics exceed thresholds.
Build dashboards: Use dashboards with your Prometheus data source for continuous monitoring and visualization.