architecture.md

Architecture

Overview

Puma is a threaded Ruby HTTP application server, processing requests across a TCP or UNIX socket.

Puma processes (there can be one or many) accept connections from the socket via a thread (in the Reactor class). The connection, once fully buffered and read, moves in to the todo list, where it will be picked up by a free/waiting thread in the threadpool (the ThreadPool class).

Puma works in two main modes: cluster and single. In single mode, only one Puma process is booted. In cluster mode, a master process is booted, which prepares (and may boot) the application, and then uses the fork() system call to create 1 or more child processes. These child processes all listen to the same socket. The master process does not listen to the socket or process requests - its purpose is mostly to manage and listen for UNIX signals and possibly kill or boot child processes.

We sometimes call child processes (or Puma processes in single mode) workers, and we sometimes call the threads created by Puma's ThreadPool worker threads.

How Requests Work

• Upon startup, Puma listens on a TCP or UNIX socket.
  • The backlog of this socket is configured (with a default of 1024). This determines the size of the queue for unaccepted connections. Generally, this setting is unimportant and will never be hit in production use. If the backlog is full, the connection will be refused by the operating system.
  • This socket backlog is distinct from the backlog of work as reported by Puma.stats or the control server. The backlog as reported by Puma is the number of connections in the process' todo set waiting for a thread from the ThreadPool.
• By default, a single, separate thread (created by the Reactor class) is used to read and buffer requests from the socket.
  • When at least one worker thread is available for work, the reactor thread listens to the socket and accepts a request, if one is waiting.
  • The reactor thread waits for the entire HTTP request to be received.
    • The time spent waiting for the HTTP request body to be received is exposed to the Rack app as env['puma.request_body_wait'] (milliseconds).
  • Once fully buffered and received, the connection is pushed into the "todo" set.
• Worker threads pop work off the "todo" set for processing.
  • The worker thread processes the request via calling the configured Rack application. The Rack application generates the HTTP response.
  • The worker thread writes the response to the connection. Note that while Puma buffers requests via a separate thread, it does not use a separate thread for responses.
  • Once done, the thread become available to process another connection in the "todo" set.

queue_requests

The queue_requests option is true by default, enabling the separate reactor thread used to buffer requests as described above.

If set to false, this buffer will not be used for connections while waiting for the request to arrive.

In this mode, when a connection is accepted, it is added to the "todo" queue immediately, and a worker will synchronously do any waiting necessary to read the HTTP request from the socket.