Azure App Service is a service used to create and deploy scalable, mission-critical web apps.
Web App: Intermittent 502 and ERR_CONNECTION_TIMED_OUT Errors — Investigation and Troubleshooting Steps
Problem description
I am experiencing intermittent 502 errors and ERR_CONNECTION_TIMED_OUT messages with my Linux Web App (atlas-redesign) hosted on Azure. These issues occur several times a day within 15-minute windows and impact my users' ability to access the application reliably.
Environment
Web App (Linux) hosted on Azure, East US.
What I've already tried
I have reviewed diagnostic information, including CPU and memory usage, and examined Azure Front Door metrics. I scaled the app from 2 to 4 instances to reduce CPU pressure. I also checked the App Service plan metrics for CPU percentage, request response times, and HTTP 5xx errors during the incident windows.
For 9-9:30 AM on 11/19 here is the CPU during the window:
and the AFD origin latency:
and the AFD percentage of 5XX:
The 5XX are not visible in the requests metrics or telemetry of our app service.
Current status
I am seeking guidance on how to verify if high CPU is truly the root cause, as claimed by support. I would appreciate any recommended diagnostic steps or best practices to identify the underlying issue.