status
this chapter is in active development
expect live edits and rapid iteration (except for when i am really busy with other stuff) while this material is written.
transport mechanics
reliable delivery and its alternatives.
most production debugging lives here. slow queries, broken streams, mysterious timeouts. tcp state and congestion behavior cause more outages than application logic ever will.
everything here assumes linux. the tools are ss, tcpdump, sysctl, and the occasional strace. if you cannot reproduce a problem with these, you are guessing.