Multiple AppName's for the same Transaction

I have 8 webservers in two data centers Delta and Bravo. The newrelic.yml the app_name field is:
app_name: Prod;ProdDelta
app_name: Prod;ProdBravo

This is useful for when I need to see if a data center is causing an issue.

On the Insights Data Summary page I see that all three applications are reporting. Does this mean that for every transaction there are two events being stored in Insights and thus halving my quota?

I only use Bravo and Delta for a very limited number of charts. If this is killing my storage, what would be a good alternate solution to grouping my servers?

This is a monolithic web app which make the price of Infrastructure overkill.


Every transaction will be stored as a unique event associated with each of the apps the agent is configured to report data to - in your case, two events are stored in Insights for every transaction. There are two main approaches you can take to reduce your Insights data count:

  1. [Recommended] Turn off event capture for specific apps. Account admins can easily do this through the Manage Data page within Insights. This would not require any changes to your existing app_name configuration, allowing you to view the aggregate data in the Prod app and data center-specific performance metrics in the Bravo and Delta apps in APM.

For this approach you could either:
(a) Disable Insights events for the Prod app and write queries with WHERE appName=‘ProdBravo’ OR appName=‘ProdDelta’ clauses to view aggregate application performance.


(b) Alternatively, disable events for the two data center apps (Bravo and Delta). The drawback to this second option is if you want to query the Insights data for just a specific data center- there would no longer be a quick way to segment the data. The workaround would be to use custom attributes to tag events with the data center the transaction hit.

  1. The second option is a riff on (b) above. You could change the app_name setting to only report data into the aggregate Prod app and again leverage custom attributes to tag events with which data center (Bravo or Delta) served the request. This would require the most work to configure and would have the downside of losing the quick visibility of aggregate app performance vs data center-specific performance in APM.