Containers

A rolling deploy drops a small number of requests every time. Why?

Whether the candidate understands the process and signal model inside a container and the timing relationship between routing and termination.

The situation behind the question

Every deploy produces a 15-second burst of 502s affecting roughly 0.3% of requests. The container entrypoint is sh -c "node server.js", the termination grace period is 30 seconds, and the application has no signal handler.