Cached pages

Pages can be cached in the browser or on a server. A cached page will contain stale (previously generated) application and queue times.

Detecting cached pages

New Relic Browser detects cached pages by comparing the total back-end time against app and queue time. When the sum of app and queue time is greater than the total back-end time, New Relic Browser assumes the page was cached and aggregates the request with app and queue times as 0.

Here are some additional tips for cached pages:

  • Java: Flush the app server’s "work" cache. This forces the app server to recompile.
  • .NET: Make sure your cache directory is clean by using the command flush_dotnet_temp.cmd. This forces the app server to recompile ASPs with page load timing instrumentation.
  • Node.js: By default, Node.js does not cache pages.
  • PHP: If you have a CDN cache your dynamic pages, make sure the cache contains the page load timing instrumentation.
  • Python: Restart your app.
  • Ruby: Look at the header information for automatic vs. manual instrumentation. For example, if you want to ignore specific webpages, you must disable automatic browser monitoring from the user interface, and then manually instrument the webpages that you do want to be tagged.

Browser traces

Browser traces can also be affected by cached pages, as the page load timing JavaScript may contain a reference to an application tier transaction trace, as well as user, account, and production information. Thus, cached page hits may result in browser traces with duplicated information. New Relic Browser attempts to recognize and prevent this as much as possible. For more information, see Browser traces.

For more help

Additional documentation resources include:

Recommendations for learning more: