Enhance Service Metrics Tab

[dannycroft]

APM agents collect multiple runtime metrics today, but only JVM metrics are available in the UI. We should increase the feature scope to include other available runtimes. Improving the availability of runtime metrics on a service's metrics tab will allow our customers a unified place to go an quickly analyze for each monitored service.

General Tasks

• Investigate the feasibility of access to metrics for each runtime (Agent/Server)
• Research most valuable metrics available for each runtime
• Implement new or enhance current APIs for extra runtime metrics
• Add runtime metrics (with corresponding components) to service metrics tab

Available Metrics

• https://www.elastic.co/guide/en/apm/agent/go/current/metrics.html
• https://www.elastic.co/guide/en/apm/agent/ruby/current/metrics.html
• https://www.elastic.co/guide/en/apm/agent/python/current/metrics.html
• https://www.elastic.co/guide/en/apm/agent/nodejs/current/metrics.html
• https://www.elastic.co/guide/en/apm/agent/dotnet/current/metrics.html
• PHP doesn't report runtime metrics today.

Screenshots

Current Metrics Tab

Current JVMs Tab

Links

• One idea was to experiment with the Dashboard Container
• OTEL -Semantic Conventions for System Metrics
• Schema for metrics ecs#474

[elasticmachine]

Pinging @elastic/apm-ui (Team:apm)

[glucaci]

Hi, there are any plans for this?

I think will be very usefull for other runtimes like .NET CLR to display other metrics than CPU and Memory.
Will make more sens to use the OTEL runtime metrics instead of system metrics?

Or another more flexible option will be to have on the Services configuration a posibility to enable/disable each graphic and map it to a desired field.

Thanks!

[sorenlouv]

@glucaci Yes, while I cannot provide exact release date I can say we are currently looking into this.

Thanks for the feedback!

[smith]

Closing this. You now should be able to accomplish most of what's here by using dashboards.