Troubleshooting Queries
Use troubleshooting information to identify and address common issues that can occur while working with queries in Monitoring.
Error: Exceeded Maximum Metric Streams
Troubleshoot too many metric streams when querying metric data.
An error indicates that the metric query exceeded the maximum number of metric streams .
This issue occurs when the query evaluates too many metric streams.
To remedy this issue, update the query to evaluate a number of metric streams that's within the limit.
For example, select dimensions to reduce the number of metric streams. See Selecting Dimensions for a Query.
To evaluate all metric streams that were in the original query, spread the metric streams across multiple queries.
Missing Resources or Metrics
Troubleshoot missing resources or metrics when querying metric data.
When querying metric data, you don't see all expected resources or metrics. In the Console, you might notice this issue when viewing a default chart, custom chart, or chart for an alarm.
Cause: Incorrect time range
The metric data returned from the query doesn't include times when the resources emit data points.
Remedy: Change the time range
See the relevant instructions:
- Service Metrics or resource page in the Console: Changing the Time Range for Default Metric Charts.
- Metrics Explorer page in the Console: Changing the Time Range for a Custom Metric Chart.
- Query (CLI, API, or Console): Selecting a Nondefault Time Range for a Query
Cause: Incorrect compartment
Metric namespaces are shown only when associated resources exist in the selected compartment. For example, the oci_autonomous_database
namespace is shown only when Autonomous Databases exist in the selected compartment.
Remedy: Change the compartment
Select the compartment that contains the resource that you want.
See the relevant instructions:
- Service Metrics page in the Console: Viewing Default Metric Charts for a Metric Namespace (Multiple Resources).
- Query (CLI, API, or the Metrics Explorer page in the Console): Creating a Query.
- Alarm query (CLI, API, or the Create Alarm page in the Console): Creating a Basic Alarm.
Cause: Compute instance isn't enabled for monitoring
The compute instance doesn't emit metric data because the instance isn't enabled for monitoring.