Hi@Frode Aarø
apologize for any inconvenience caused on this.
Ensure your application is configured to use the WEBSITE_WARMUP_PATH setting. This helps define paths indicating when your application is ready to serve requests, reducing perceived startup time.
Implement the Health Check feature to manage unexpected code and platform failures during startup. It can remove failing instances from the load balancer, minimizing downtime.
Set up Auto-Heal rules to monitor failed requests and automatically restart the worker process if the application enters an unrecoverable state after startup.
Thoroughly test your application's startup behavior, including simulating dependency failures to understand its response under different conditions. Even a small failure rate can cause significant issues when multiple instances are involved.
Analyze application logs regularly for signs of delays or failures during startup to identify areas needing optimization.
Review and optimize your application code and dependencies to improve startup performance. Reducing dependencies or enhancing their loading times can significantly impact startup duration.
If possible, increase the instance count to distribute the load and improve overall responsiveness during startup.
Since the "Always On" setting is enabled, ensure your application is benefiting from this feature to remain loaded and ready to respond.
references:
https://learn.microsoft.com/en-us/azure/app-service/routine-maintenance-downtime#strategies-for-increased-uptime