Lab: the anatomy of a production gRPC server¶

One in-process grpc.aio server, built six ways, and a client that reports what it got: what an unhandled handler exception tells the caller, what a reporting interceptor sees from either side of the exception handler, what a payload over the size limit gets, what a shutdown does to a request that is still running, what the health service says while a dependency is down, and what TLS and mTLS refuse. The certificates are generated with openssl into a temporary directory.

uv venv --python 3.13 .venv
uv pip install --python .venv/bin/python "grpc-server-kit[health,settings]==0.1.1"
.venv/bin/python server_lab.py