Metric normalization rules

Metric normalization rules help you to manage Metric Grouping Issues (MGIs). This helps to ensure that application performance data remains relevant. The metric normalization applies only to metric timeslice data.

You can create and manage metric normalization rules to:

• Group or filter out metric timeslice data
• Manage the high cardinality metrics that are reported to New Relic
• Prevent MGIs
• Prevent automatic creation of DENY_NEW_METRICS rules
• Reduce the number of unique metric names reported to New Relic

Manage the metric normalization

You can create and manage new metric normalization rules for an entity using the Metric normalization rules page.

View metric normalization rules

To view the Metric normalization rules page go to https://one.newrelic.com > All Entities > Select an entity > Settings > Metric normalization .

Create a metric normalization rule

To create a metric normalization rule:

1. On the Metric normalization rules page, click + Create a new rule.

2. Provide the following details:

   Field	Description
   Time range	Select the required time range from the drop-down menu for the rule.
   Regular expression	Enter a regular expression pattern that matches the metrics to be normalized. This field must start with a ^ and end with a $. It defines a pattern to identify which metrics should be affected by the normalization rule. For example: ^WebTransaction/Uri/RESTfulExample/users/username/[a-zA-Z0-9]+$ The pattern ^[a-zA-Z0-9]+$ matches any string composed of lowercase characters (a-z), uppercase characters (A-Z), and numbers (0-9).
   Regular expression matches	Preview the metrics matched by the Regular expression field.
   What do you want to do with the matches?	Select one of the following action to perform on the metrics that match the Regular expression field: Replace: Replace the matched metrics with value in the Replacement field. Ignore: Block any matched metrics from reporting to New Relic platform. Ignore new metrics: Prevent the creation of new metric names, while allowing existing metric names to continue reporting.
   Replacement	The Replacement field is required only when you select the Replace option in the previous field. Enter the replacement metric to replace the matched metric defined in the Regular expression field.
   Order of rule in sequence	Specify the order in which the rule is applied relative to other rules. Lower the number, higher the evaluated priority. Recommend value is 2000
   Terminate	If enabled, no additional rules will be applied to the metrics after this rule successfully matches a metric.
   Active	Toggle to enable or disable the rule.
   Notes	Notes on the rule for reference.

3. Click Save.

Update a metric normalization rule

Update a metric normalization rule

1. On the Metric normalization rules page, select the rule, and click the icon, then select Edit.
2. Update the necessary details.
3. Click Save.

Disable a metric normalization rule

You can't delete a metric normalization rule, but you can deactivate it. The deactivated rule will not affect any metrics reporting to New Relic. You can deactivate a DENY_NEW_METRICS rule from the MGI details page, when the Denied rate (minute) is below the resolution rate (minute) of 1000.

To deactivate a rule:

1. On the Metric normalization rules page, select the required rule, and click the icon, then click MGI details.
2. Click Deactivate.

MGI details page

The MGI details page provides a comprehensive overview of the metric normalization rules applied to your metrics. It allows you to analyze the impact of these rules on your metric data and make informed decisions about managing high cardinality metrics.

On the MGI details page, you can view the following information:

• Blocked Metrics Overview: Displays the number of metrics affected by the DENY_NEW_METRICS rule, including the rate at which metrics are blocked. Similar insights are available for REPLACE and IGNORE actions.
• Time Picker Functionality: Allows you to view metrics reporting before rule activation, broken down by metric prefix. This feature helps identify the source of metric growth.
• NRQL Query Integration: Offers a 'view in query builder' button to see the NRQL query illustrating the metric explosion just before any rule was applied. This targets Metric Timeslice data, enabling retrospective analysis up to three months.
• Metric Pattern Identification: Enables identification of exploding metric patterns and creation of new regex rules to manage them effectively.
• Real-Time Status Indicators: Includes widgets showing the current status of normalization rules, with indicators changing from 'Active' to 'Mitigated' as rules manage metric growth.
• Rule Modification Capability: Allows users to deactivate their own DENY_NEW_METRICS rules directly from the MGI Details page, as well as modify REPLACE and IGNORE rules without support tickets. For more information, refer to Disable a rule.

Access the MGI details page

On the Metric normalization rules page, select the rule and click the icon, then select MGI details.

Example use case

In the above example, the MGI details page displays that 1.1 million unique metric names are blocked by the DENY_NEW_METRICS rule at a rate of 738 metrics per minute. The majority of this growth is traced to the namespace WebTransaction/Uri/RESTfulExample/users/username/. By viewing the NRQL query, you can pinpoint the source of the explosion, allowing you to create a new rule to manage the metrics effectively.