Find data to alert on

Before creating an alert rule, you need to know the exact metric or log query you want to monitor. Grafana’s Drilldown apps make this easy—you can explore your data, find interesting metrics or logs, and then use those queries in your alert.

You don’t need an existing dashboard. The Drilldown apps let you discover what’s available and identify what’s worth alerting on.

To find data to alert on, complete the following steps:

1. Sign in to your Grafana Cloud account.

2. In Grafana Cloud, select Drilldown from the main menu.

3. Choose the appropriate Drilldown app based on what you want to alert on:

   Expand table

   If you want to alert on	Select
   Infrastructure metrics (CPU, memory, disk)	Metrics Drilldown
   Application or system logs	Logs Drilldown

4. In the Drilldown app, explore the available data:

   • For Metrics Drilldown: Browse metric names, filter by labels (like job or instance), and view current values.
   • For Logs Drilldown: Search log content, filter by labels, and identify patterns worth alerting on.

5. Identify a specific metric or log pattern you want to alert on.

   Metrics examples:

   Expand table

   Metric	Use case
   node_cpu_seconds_total	CPU usage per host
   node_memory_MemAvailable_bytes	Available memory
   node_filesystem_avail_bytes	Disk space

   Logs examples:

   Expand table

   Log query	Use case
   {job="myapp"} |= "error"	Application errors
   {job="nginx"} | json | status >= 500	HTTP 5xx errors
   {job="syslog"} |= "OOM"	Out-of-memory events

6. Note the metric name or log query—you’ll use this in your alert rule.

   Tip: You can also copy the query directly from the Drilldown app to paste into your alert rule later.

In the next milestone, you’ll build your alert query.