Your data. Anywhere you go.

New Relic for iOS or Android


Download on the App Store    Android App on Google play


New Relic Insights App for iOS


Download on the App Store


Learn more

Close icon

Relic Solution: Common Causes of Metric Grouping Issues


#1

This post is third in a series of Metric Grouping related posts. Part 1, Part 2.

Metrics are named with “segments” which closely resemble folders of a file structure. e.g. Datastore/statement/MySQL/search/insert

The most common causes of metric grouping issues can be expressed generally in the primary segment:

  • Datastores/ metrics: SQL database calls that utilise temporary tables.

  • External/ metrics: The application makes calls to many various urls, such as a web crawler.

  • ExternalTransaction/ metrics: The application uses Cross Application Tracing to other applications on the account which have Metric Grouping Issues. The issues in the other applications are generally WebTransaction as seen below.

  • Message Broker related metrics (e.g. RabbitMQ) where the message broker utilises GUID’s or unique values to identify each message which are picked up by the agent.

The above issues typically have limited configuration options in the agents, and generally require Support assistance to mitigate. However the occurrence of these issues is dwarfed by the most common cause of Metric Grouping Issues: improper web transaction naming.

  • WebTransaction/, WebTransactionTotalTime/, and Apdex/ metrics. For every web transaction, three metrics are created to track and display properly in the APM Dashboards. Web Transaction MGI’s typically come from improper configuration or lack of instrumentation. Fortunately, there are many options for managing and customising Web Transaction names. These vary by agent and we’ll look into them in the next post.