The APM Agent is software that runs on your application server. It collects traces and sends them to Raygun, where you can analyze your application's performance.
Application Performance Index (Apdex) is an open standard that benchmarks user experience and allows you to identify poor user experience. Learn more about Apdex on the overview page.
The overview page provides a customizable summary of application performance over time. Use it as a health monitor or a starting point for a deeper investigation.
The visualizations and metrics show what's running slowly, what can be improved, and user satisfaction.
Applications are containers used to segment data inside Raygun. With unlimited applications, you can monitor many different software projects and development environments within one Raygun account.
The call tree shows where your application is spending the most time in aggregate. Click the parent name to expand and view details about its child traces. To learn about a specific call click the magnifying glass icon on the right hand side to display execution time, parent %, and if it is in the hotpath.
Code filtering lets you control which methods appear in traces. This filters out unwanted data like short running helper functions which are already performant, but can clog up traces as they are regularly called.
The discover page shows a list of your application's slowest and most popular requests, methods, queries and API calls. Use this page to identify slow transactions that could benefit from a performance fine-tuning.
An error is an exception that Raygun APM has recorded but has not been sent to Raygun Crash Reporting.
An exception refers to the detection of an unhandled exception as a result of your code. Raygun APM will track when they occur and correlate them to the associated crash data in Raygun Crash Reporting for complete visibility.
On the API call detail page you will see the performance details of API calls including execution duration over time and sample count over time. This is helpful for identifying if you're relying on third party who has poor performance that could be impacting your customer experience.
Click on a trace name to view a trace on a flame chart. This shows the state of the stack at every millisecond during the performance profile, how long each segment ran for, and where it was called from.
Click the "show hot path" toggle to view the hot path of a trace on a flame chart. This changes the colors of the flame chart to show the series of calls which made up the majority of the total time taken.
Integrations are a powerful way to make the most of your Raygun APM experience and workflow.
Issues are created each time a trace violates a performance rule. Violations of the same rule are grouped together as an issue group on the issues page. Issue groups are sorted by the last seen date but can be sorted by type, issue name, first seen, or number of traces by clicking the header row.
Manual issues can be created or added to from a trace or method, so you can be alerted if the violation of this rule occurs in the future.
On the method detail page you will see the execution duration over time, sample count over time, and traces associated with each method.
On the query detail page you will see the execution duration over time, sample count over time, and traces associated with each query.
On the request detail page you will see the execution duration over time, sample count over time, page execution over time, call tree distribution, and the traces associated with each request.
Requests per minute (RPM)
The Requests per minute (RPM) tile shows how many http requests your application receives each minute.
Rules are customizable policies that create an issue when a performance threshold is reached. Use the default rules or create your own to proactively monitor application performance before an actual issue or outage happens.
Sampling gives you control over the rate at which traces are accepted into Raygun APM. Sampling can be defined as a single rate applied to all traces or a rate per URL for a web application.
A trace is a profiling trace of a given request. The traces page allows you to drill into traces and visualize them on a flame chart. Traces can also be combined with your source control system to show the code inside the methods to assist engineers in identifying why methods may be slow.