Azure: Suave fails sometimes with HttpPlatfomHandler and doesn't come back up

Question

I'm hosting my web application on azure web app by using suave and following the recommended guide.

---

My app was running like a charm until some days ago something went horribly wrong:

The process threw and exception and crashed, but HttpPlatformHandler failed to restart the process and the eventlog was full of the following errors:

Process '0' failed to start. Port = 30590, Error Code = '-2147023897'.

and

Maximum rapid fail count per minute of '10' exceeded.

---

I got this error for hours and in this time my application responded with

502: The specified CGI application encountered an error and the server terminated the process.

The only way to solve the problem was to manually restart the web application.

---

I don't understand why the platform handler was unable to restart the process. A successful start looks like the following:

Process '5516' started successfully and is listening on port '29676'.

So what's up with this Process '0'?

Ethology

As you mentioned that this issue could be solved by restart your web app manually, I assumed that there be some possible causes (e.g memory usage, CPU, application error, webServer exception etc). You could leverage New Relic or Application Insights to retrieve more details about application exception, server logs and requests. – Syndetic

There wasn't any other error, except the exception that was causing the application crash and then tons oft the other events descripbed above. Also, Application Insights didn't help, because the exception was thrown by IIS (HTTP platform handler), not my application, so no event came through (and my alert rules didn't take effect). – Ethology

My problem isn't, that I couldn't determine the cause of the application crash, it is rather the behavior of the web app after the application crashed (with this strange Process '0' failed to start.). – Ethology