App Engine Error 503 - What is the root cause?
I am doing a stress test in my app (Java + Spring) that is deployed on Google App Engine. My instance class is B4. The amount of instances is 1 (because I am testing).
My test sends 200 requests at once. It is my real case (no rump). After this request the next request will be 10 minutes after. Each request spend 200ms(average) to execute (when I send only 1).
The process is light, does not have complex rules or actions on database. Only get a JSON, transform to Java Object and insert on database (No rules).
I will use Queue for balance my requests, but, at moment I am testing amount of requests the app supports at once for choose how many machines (instances) and rate/bucket I will configure (queue).
When requests reaches 180, any errors happen, but I did not know what is the real reason. My target is test until received errors, but I would like to know what is the reason (memory, CPU, limit time, etc).
The most frequently log error is:
"Request was aborted after waiting too long to attempt to service your request."
There is a limit to App Engine hold the requests? I read about this message in others questions here, but I cannot understand the real mean.
The other error message in logs is:
"A problem was encountered with the process that handled this request, causing it to exit. This is likely to cause a new process to be used for the next request to your application. (Error code 204)"
I read about people telling that is problem with limit of memory, but, the memory usage in graph (stackdriver) it is ok. the B4 instance has 512MB and the max reached it was 266MB.
I think when the problem is memory, the message will be: "OutOfMemoryError: Java heap space"