wu-sheng commented on code in PR #13502: URL: https://github.com/apache/skywalking/pull/13502#discussion_r2451926789
########## docs/en/setup/backend/backend-go-app-profiling.md: ########## @@ -0,0 +1,107 @@ +# Go App Profiling + +Go App Profiling uses the Pprof for sampling + +Pprof is bound within the auto-instrument agent and corresponds to [In-Process Profiling](../../concepts-and-designs/profiling.md#in-process-profiling). + +It is delivered to the agent in the form of a task, allowing it to be enabled or disabled dynamically. +When service encounters performance issues (cpu usage, memory allocation, etc.), Pprof task can be created. +When the agent receives a task, it enables Pprof for sampling. +After sampling is completed, the sampling results are analyzed by requesting the server to render a flame graph for performance +analysis to determine the specific business code lines that cause performance problems. +Note, tracing profiling in the Go agent relies on the Go runtime’s global CPU sampling used by pprof. +Since only one CPU profiler can run at a time within the same instance, tracing and pprof CPU profiling cannot be enabled simultaneously. +If both are activated on the same instance, one task may fail to start. Review Comment: Is this about time ranges overlapping verification? -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
