NOTE
Netty Pipeline: Exception Propagation
How handler failures become exceptionCaught events and why exceptions need explicit logging, cleanup, and connection policy.
This is a historical learning note and may contain outdated or incomplete understanding.
Exceptions raised while processing pipeline events can propagate through exceptionCaught handling. An application handler should decide whether the failure is recoverable for that connection/message, should be translated, or should close the channel.
Never silently swallow unexpected exceptions. Include connection/request context without leaking secrets and ensure reference-counted buffers/resources are released on failure paths.
Exception propagation direction and handler internals have version details; rely on the current Netty API contract for exact behavior.