puma 5.2.2 → 6.3.0

data/docs/plugins.md

@@ -3,22 +3,22 @@
 Puma 3.0 added support for plugins that can augment configuration and service
 operations.
 
-2 canonical plugins to look to aid in development of further plugins:
+There are two canonical plugins to aid in the development of new plugins:
 
 * [tmp\_restart](https://github.com/puma/puma/blob/master/lib/puma/plugin/tmp_restart.rb):
   Restarts the server if the file `tmp/restart.txt` is touched
 * [heroku](https://github.com/puma/puma-heroku/blob/master/lib/puma/plugin/heroku.rb):
-  Packages up the default configuration used by puma on Heroku (being sunset with the release of Puma 5.0)
+  Packages up the default configuration used by Puma on Heroku (being sunset
+  with the release of Puma 5.0)
 
-Plugins are activated in a puma configuration file (such as `config/puma.rb'`)
+Plugins are activated in a Puma configuration file (such as `config/puma.rb'`)
 by adding `plugin "name"`, such as `plugin "heroku"`.
 
-Plugins are activated based simply on path requirements so, activating the
-`heroku` plugin will simply be doing `require "puma/plugin/heroku"`. This
-allows gems to provide multiple plugins (as well as unrelated gems to provide
-puma plugins).
+Plugins are activated based on path requirements so, activating the `heroku`
+plugin is much like `require "puma/plugin/heroku"`. This allows gems to provide
+multiple plugins (as well as unrelated gems to provide Puma plugins).
 
-The `tmp_restart` plugin is bundled with puma, so it can always be used.
+The `tmp_restart` plugin comes with Puma, so it is always available.
 
 To use the `heroku` plugin, add `puma-heroku` to your Gemfile or install it.
 
@@ -26,13 +26,13 @@ To use the `heroku` plugin, add `puma-heroku` to your Gemfile or install it.
 
 ## Server-wide hooks
 
-Plugins can use a couple of hooks at server level: `start` and `config`.
+Plugins can use a couple of hooks at the server level: `start` and `config`.
 
-`start` runs when the server has started and allows the plugin to start other
-functionality to augment puma.
+`start` runs when the server has started and allows the plugin to initiate other
+functionality to augment Puma.
 
-`config` runs when the server is being configured and is passed a `Puma::DSL`
-object that can be used to add additional configuration.
+`config` runs when the server is being configured and receives a `Puma::DSL`
+object that is useful for additional configuration.
 
-Any public methods in `Puma::Plugin` are the public API that any plugin may
-use.
+Public methods in [`Puma::Plugin`](../lib/puma/plugin.rb) are treated as a
+public API for plugins.

data/docs/rails_dev_mode.md

@@ -2,16 +2,15 @@
 
 ## "Loopback requests" 
 
-Be cautious of "loopback requests", where a Rails application executes a request to a server that in turn, results in another request back to the same Rails application before the first request is completed. Having a loopback request will trigger [Rails' load interlock](https://guides.rubyonrails.org/threading_and_code_execution.html#load-interlock) mechanism. The load interlock mechanism prevents a thread from using Rails autoloading mechanism to load constants while the application code is still running inside another thread.
+Be cautious of "loopback requests," where a Rails application executes a request to a server that, in turn, results in another request back to the same Rails application before the first request completes. Having a loopback request will trigger [Rails' load interlock](https://guides.rubyonrails.org/threading_and_code_execution.html#load-interlock) mechanism. The load interlock mechanism prevents a thread from using Rails autoloading mechanism to load constants while the application code is still running inside another thread.
 
 This issue only occurs in the development environment as Rails' load interlock is not used in production environments. Although we're not sure, we believe this issue may not occur with the new `zeitwerk` code loader.
 
 ### Solutions
 
-
 #### 1. Bypass Rails' load interlock with `.permit_concurrent_loads`
 
-Wrap the first request inside a block that will allow concurrent loads, [`ActiveSupport::Dependencies.interlock.permit_concurrent_loads`](https://guides.rubyonrails.org/threading_and_code_execution.html#permit-concurrent-loads). Anything wrapped inside the `.permit_concurrent_loads` block will bypass the load interlock mechanism, allowing new threads to access the Rails environment and boot properly. 
+Wrap the first request inside a block that will allow concurrent loads: [`ActiveSupport::Dependencies.interlock.permit_concurrent_loads`](https://guides.rubyonrails.org/threading_and_code_execution.html#permit-concurrent-loads). Anything wrapped inside the `.permit_concurrent_loads` block will bypass the load interlock mechanism, allowing new threads to access the Rails environment and boot properly. 
 
 ###### Example 
 

data/docs/restart.md

@@ -1,8 +1,8 @@
-Puma provides three distinct kinds of restart operations, each for different use cases. Hot restarts and phased restarts are described here. The third kind of restart operation is called "refork" and is described in the documentation for [`fork_worker`](fork_worker.md).
+Puma provides three distinct kinds of restart operations, each for different use cases. This document describes "hot restarts" and "phased restarts." The third kind of restart operation is called "refork" and is described in the documentation for [`fork_worker`](fork_worker.md).
 
 ## Hot restart
 
-To perform a "hot" restart, Puma performs an `exec` operation to start the process up again, so no memory is shared between the old process and the new process. As a result, it is safe to issue a restart any place where you would manually stop Puma and start it again. In particular, it is safe to upgrade Puma itself using a hot restart.
+To perform a "hot" restart, Puma performs an `exec` operation to start the process up again, so no memory is shared between the old process and the new process. As a result, it is safe to issue a restart at any place where you would manually stop Puma and start it again. In particular, it is safe to upgrade Puma itself using a hot restart.
 
 If the new process is unable to load, it will simply exit. You should therefore run Puma under a process monitor when using it in production.
 
@@ -16,14 +16,14 @@ Any of the following will cause a Puma server to perform a hot restart:
 
 ### Supported configurations
 
-* Works in cluster mode and in single mode
+* Works in cluster mode and single mode
 * Supported on all platforms
 
 ### Client experience
 
-* All platforms: for clients with an in-flight request, those clients will be served responses before the connection is closed gracefully. Puma gracefully disconnects any idle HTTP persistent connections before restarting.
+* All platforms: clients with an in-flight request are served responses before the connection is closed gracefully. Puma gracefully disconnects any idle HTTP persistent connections before restarting.
 * On MRI or TruffleRuby on Linux and BSD: Clients who connect just before the server restarts may experience increased latency while the server stops and starts again, but their connections will not be closed prematurely.
-* On Windows and on JRuby: Clients who connect just before a restart may experience "connection reset" errors.
+* On Windows and JRuby: Clients who connect just before a restart may experience "connection reset" errors.
 
 ### Additional notes
 
@@ -32,7 +32,7 @@ Any of the following will cause a Puma server to perform a hot restart:
 
 ## Phased restart
 
-Phased restarts replace all running workers in a Puma cluster. This is a useful way to gracefully upgrade the application that Puma is serving. A phased restart works by first killing an old worker, then starting a new worker, waiting until the new worker has successfully started before proceeding to the next worker. This process continues until all workers have been replaced. The master process is not restarted.
+Phased restarts replace all running workers in a Puma cluster. This is a useful way to upgrade the application that Puma is serving gracefully. A phased restart works by first killing an old worker, then starting a new worker, waiting until the new worker has successfully started before proceeding to the next worker. This process continues until all workers are replaced. The master process is not restarted.
 
 ### How-to
 
@@ -45,7 +45,7 @@ Any of the following will cause a Puma server to perform a phased restart:
 ### Supported configurations
 
 * Works in cluster mode only
-* To support upgrading the application that Puma is serving, ensure `prune_bundler` is enabled and that `preload_app` is disabled (it is disabled by default).
+* To support upgrading the application that Puma is serving, ensure `prune_bundler` is enabled and that `preload_app!` is disabled
 * Supported on all platforms where cluster mode is supported
 
 ### Client experience

data/docs/signals.md

@@ -1,8 +1,8 @@
-The [unix signal](https://en.wikipedia.org/wiki/Unix_signal) is a method of sending messages between [processes](https://en.wikipedia.org/wiki/Process_(computing)). When a signal is sent, the operating system interrupts the target process's normal flow of execution. There are standard signals that are used to stop a process but there are also custom signals that can be used for other purposes. This document is an attempt to list all supported signals that Puma will respond to. In general, signals need only be sent to the master process of a cluster.
+The [unix signal](https://en.wikipedia.org/wiki/Unix_signal) is a method of sending messages between [processes](https://en.wikipedia.org/wiki/Process_(computing)). When a signal is sent, the operating system interrupts the target process's normal flow of execution. There are standard signals that are used to stop a process, but there are also custom signals that can be used for other purposes. This document is an attempt to list all supported signals that Puma will respond to. In general, signals need only be sent to the master process of a cluster.
 
 ## Sending Signals
 
-If you are new to signals it can be useful to see how they can be used. When a process is created in a *nix like operating system it will have a [PID - or process identifier](https://en.wikipedia.org/wiki/Process_identifier) that can be used to send signals to the process. For demonstration we will create an infinitely running process by tailing a file:
+If you are new to signals, it can be helpful to see how they are used. When a process starts in a *nix-like operating system, it will have a [PID - or process identifier](https://en.wikipedia.org/wiki/Process_identifier) that can be used to send signals to the process. For demonstration, we will create an infinitely running process by tailing a file:
 
 ```sh
 $ echo "foo" >> my.log
@@ -10,7 +10,7 @@ $ irb
 > pid = Process.spawn 'tail -f my.log'
 ```
 
-From here we can see that the tail process is running by using the `ps` command:
+From here, we can see that the tail process is running by using the `ps` command:
 
 ```sh
 $ ps aux | grep tail
@@ -27,7 +27,7 @@ Process.detach(pid) # https://ruby-doc.org/core-2.1.1/Process.html#method-c-deta
 Process.kill("TERM", pid)
 ```
 
-Now you will see via `ps` that there is no more `tail` process. Sometimes when referring to signals the `SIG` prefix will be used for instance `SIGTERM` is equivalent to sending `TERM` via `Process.kill`.
+Now you will see via `ps` that there is no more `tail` process. Sometimes when referring to signals, the `SIG` prefix will be used. For example, `SIGTERM` is equivalent to sending `TERM` via `Process.kill`.
 
 ## Puma Signals
 
@@ -35,13 +35,14 @@ Puma cluster responds to these signals:
 
 - `TTIN` increment the worker count by 1
 - `TTOU` decrement the worker count by 1
-- `TERM` send `TERM` to worker. Worker will attempt to finish then exit.
-- `USR2` restart workers. This also reloads puma configuration file, if there is one.
-- `USR1` restart workers in phases, a rolling restart. This will not reload configuration file.
-- `HUP ` reopen log files defined in stdout_redirect configuration parameter. If there is no stdout_redirect option provided it will behave like `INT`
-- `INT ` equivalent of sending Ctrl-C to cluster. Will attempt to finish then exit.
+- `TERM` send `TERM` to worker. The worker will attempt to finish then exit.
+- `USR2` restart workers. This also reloads the Puma configuration file, if there is one.
+- `USR1` restart workers in phases, a rolling restart. This will not reload the configuration file.
+- `HUP ` reopen log files defined in stdout_redirect configuration parameter. If there is no stdout_redirect option provided, it will behave like `INT`
+- `INT ` equivalent of sending Ctrl-C to cluster. Puma will attempt to finish then exit.
 - `CHLD`
-- `URG ` refork workers in phases from worker 0, if `fork_workers` option is enabled.
+- `URG ` refork workers in phases from worker 0 if `fork_workers` option is enabled.
+- `INFO` print backtraces of all puma threads
 
 ## Callbacks order in case of different signals
 

data/docs/stats.md

@@ -1,4 +1,4 @@
-## accessing stats
+## Accessing stats
 
 Stats can be accessed in two ways:
 
@@ -47,18 +47,18 @@ end
 
 ## Explanation of stats
 
-`Puma.stats` returns different information and a different structure depending on if Puma is in single vs cluster mode. There is one top-level attribute that is common to both modes:
+`Puma.stats` returns different information and a different structure depending on if Puma is in single vs. cluster mode. There is one top-level attribute that is common to both modes:
 
-* started_at: when puma was started
+* started_at: when Puma was started
 
 ### single mode and individual workers in cluster mode
 
-When Puma is run in single mode, these stats ar available at the top level. When Puma is run in cluster mode, these stats are available within the `worker_status` array in a hash labeled `last_status`, in an array of hashes, one hash for each worker.
+When Puma runs in single mode, these stats are available at the top level. When Puma runs in cluster mode, these stats are available within the `worker_status` array in a hash labeled `last_status`, in an array of hashes where one hash represents each worker.
 
 * backlog: requests that are waiting for an available thread to be available. if this is above 0, you need more capacity [always true?]
 * running: how many threads are running
-* pool_capacity: the number of requests that the server is capable of taking right now. For example if the number is 5 then it means there are 5 threads sitting idle ready to take a request. If one request comes in, then the value would be 4 until it finishes processing. If the minimum threads allowed is zero, this number will still have a maximum value of the maximum threads allowed.
-* max_threads: the maximum number of threads puma is configured to spool up per worker 
+* pool_capacity: the number of requests that the server is capable of taking right now. For example, if the number is 5, then it means there are 5 threads sitting idle ready to take a request. If one request comes in, then the value would be 4 until it finishes processing. If the minimum threads allowed is zero, this number will still have a maximum value of the maximum threads allowed.
+* max_threads: the maximum number of threads Puma is configured to spool per worker
 * requests_count: the number of requests this worker has served since starting
 
 
@@ -72,9 +72,9 @@ When Puma is run in single mode, these stats ar available at the top level. When
 
 ### worker status
 
-* started_at: when the worker was started
+* started_at: when the worker started
 * pid: the process id of the worker process
-* index: each worker gets a number. if puma is configured to have 3 workers, then this will be 0, 1, or 2
+* index: each worker gets a number. if Puma is configured to have 3 workers, then this will be 0, 1, or 2
 * booted: if it's done booting [?]
 * last_checkin: Last time the worker responded to the master process' heartbeat check.
 * last_status: a hash of info about the worker's state handling requests. See the explanation for this in "single mode and individual workers in cluster mode" section above.

data/docs/systemd.md

@@ -1,19 +1,18 @@
 # systemd
 
-[systemd](https://www.freedesktop.org/wiki/Software/systemd/) is a
-commonly available init system (PID 1) on many Linux distributions. It
-offers process monitoring (including automatic restarts) and other
-useful features for running Puma in production.
+[systemd](https://www.freedesktop.org/wiki/Software/systemd/) is a commonly
+available init system (PID 1) on many Linux distributions. It offers process
+monitoring (including automatic restarts) and other useful features for running
+Puma in production.
 
 ## Service Configuration
 
-Below is a sample puma.service configuration file for systemd, which
-can be copied or symlinked to /etc/systemd/system/puma.service, or if
-desired, using an application or instance specific name.
+Below is a sample puma.service configuration file for systemd, which can be
+copied or symlinked to `/etc/systemd/system/puma.service`, or if desired, using
+an application or instance-specific name.
 
-Note that this uses the systemd preferred "simple" type where the
-start command remains running in the foreground (does not fork and
-exit).
+Note that this uses the systemd preferred "simple" type where the start command
+remains running in the foreground (does not fork and exit).
 
 ~~~~ ini
 [Unit]
@@ -25,8 +24,7 @@ After=network.target
 
 [Service]
 # Puma supports systemd's `Type=notify` and watchdog service
-# monitoring, if the [sd_notify](https://github.com/agis/ruby-sdnotify) gem is installed,
-# as of Puma 5.1 or later.
+# monitoring, as of Puma 5.1 or later.
 # On earlier versions of Puma or JRuby, change this to `Type=simple` and remove
 # the `WatchdogSec` line.
 Type=notify
@@ -37,8 +35,8 @@ WatchdogSec=10
 # Preferably configure a non-privileged user
 # User=
 
-# The path to the your application code root directory.
-# Also replace the "<YOUR_APP_PATH>" place holders below with this path.
+# The path to your application code root directory.
+# Also replace the "<YOUR_APP_PATH>" placeholders below with this path.
 # Example /home/username/myapp
 WorkingDirectory=<YOUR_APP_PATH>
 
@@ -64,33 +62,31 @@ Restart=always
 WantedBy=multi-user.target
 ~~~~
 
-See [systemd.exec](https://www.freedesktop.org/software/systemd/man/systemd.exec.html)
+See
+[systemd.exec](https://www.freedesktop.org/software/systemd/man/systemd.exec.html)
 for additional details.
 
 ## Socket Activation
 
-systemd and puma also support socket activation, where systemd opens
-the listening socket(s) in advance and provides them to the puma
-master process on startup. Among other advantages, this keeps
-listening sockets open across puma restarts and achieves graceful
-restarts, including when upgraded puma, and is compatible with both
-clustered mode and application preload.
-
-**Note:** Any wrapper scripts which `exec`, or other indirections in
-`ExecStart`, may result in activated socket file descriptors being closed
-before they reach the puma master process. For example, if using `bundle exec`,
-pass the `--keep-file-descriptors` flag. `bundle exec` can be avoided by using a
-`puma` executable generated by `bundle binstubs puma`. This is tracked in
-[#1499].
-
-**Note:** Socket activation doesn't currently work on JRuby. This is
-tracked in [#1367].
-
-To use socket activation, configure one or more `ListenStream` sockets
-in a companion `*.socket` unit file. Also uncomment the associated
-`Requires` directive for the socket unit in the service file (see
-above.) Here is a sample puma.socket, matching the ports used in the
-above puma.service:
+systemd and Puma also support socket activation, where systemd opens the
+listening socket(s) in advance and provides them to the Puma master process on
+startup. Among other advantages, this keeps listening sockets open across puma
+restarts and achieves graceful restarts, including when upgraded Puma, and is
+compatible with both clustered mode and application preload.
+
+**Note:** Any wrapper scripts which `exec`, or other indirections in `ExecStart`
+may result in activated socket file descriptors being closed before reaching the
+puma master process. For example, if using `bundle exec`, pass the
+`--keep-file-descriptors` flag. `bundle exec` can be avoided by using a `puma`
+executable generated by `bundle binstubs puma`. This is tracked in [#1499].
+
+**Note:** Socket activation doesn't currently work on JRuby. This is tracked in
+[#1367].
+
+Configure one or more `ListenStream` sockets in a companion `*.socket` unit file
+to use socket activation. Also, uncomment the associated `Requires` directive
+for the socket unit in the service file (see above.) Here is a sample
+puma.socket, matching the ports used in the above puma.service:
 
 ~~~~ ini
 [Unit]
@@ -113,31 +109,32 @@ Backlog=1024
 WantedBy=sockets.target
 ~~~~
 
-See [systemd.socket](https://www.freedesktop.org/software/systemd/man/systemd.socket.html)
+See
+[systemd.socket](https://www.freedesktop.org/software/systemd/man/systemd.socket.html)
 for additional configuration details.
 
-Note that the above configurations will work with Puma in either
-single process or cluster mode.
+Note that the above configurations will work with Puma in either single process
+or cluster mode.
 
 ### Sockets and symlinks
 
-When using releases folders, you should set the socket path using the
-shared folder path (ex. `/srv/projet/shared/tmp/puma.sock`), not the
-release folder path (`/srv/projet/releases/1234/tmp/puma.sock`).
+When using releases folders, you should set the socket path using the shared
+folder path (ex. `/srv/projet/shared/tmp/puma.sock`), not the release folder
+path (`/srv/projet/releases/1234/tmp/puma.sock`).
 
 Puma will detect the release path socket as different than the one provided by
-systemd and attempt to bind it again, resulting in the exception
- `There is already a server bound to:`.
+systemd and attempt to bind it again, resulting in the exception `There is
+already a server bound to:`.
 
 ### Binding
 
-By default you need to configure puma to have binds matching with all
+By default, you need to configure Puma to have binds matching with all
 ListenStream statements. Any mismatched systemd ListenStreams will be closed by
-puma.
+Puma.
 
 To automatically bind to all activated sockets, the option
 `--bind-to-activated-sockets` can be used. This matches the config DSL
-`bind_to_activated_sockets` statement. This will cause puma to create a bind
+`bind_to_activated_sockets` statement. This will cause Puma to create a bind
 automatically for any activated socket. When systemd socket activation is not
 enabled, this option does nothing.
 
@@ -146,8 +143,8 @@ binds that's not socket activated.
 
 ## Usage
 
-Without socket activation, use `systemctl` as root (e.g. via `sudo`) as
-with other system services:
+Without socket activation, use `systemctl` as root (i.e., via `sudo`) as with
+other system services:
 
 ~~~~ sh
 # After installing or making changes to puma.service
@@ -156,35 +153,35 @@ systemctl daemon-reload
 # Enable so it starts on boot
 systemctl enable puma.service
 
-# Initial start up.
+# Initial startup.
 systemctl start puma.service
 
 # Check status
 systemctl status puma.service
 
-# A normal restart. Warning: listeners sockets will be closed
+# A normal restart. Warning: listener's sockets will be closed
 # while a new puma process initializes.
 systemctl restart puma.service
 ~~~~
 
-With socket activation, several but not all of these commands should
-be run for both socket and service:
+With socket activation, several but not all of these commands should be run for
+both socket and service:
 
 ~~~~ sh
 # After installing or making changes to either puma.socket or
 # puma.service.
 systemctl daemon-reload
 
-# Enable both socket and service so they start on boot.  Alternatively
-# you could leave puma.service disabled and systemd will start it on
-# first use (with startup lag on first request)
+# Enable both socket and service, so they start on boot.  Alternatively
+# you could leave puma.service disabled, and systemd will start it on
+# the first use (with startup lag on the first request)
 systemctl enable puma.socket puma.service
 
-# Initial start up. The Requires directive (see above) ensures the
+# Initial startup. The Requires directive (see above) ensures the
 # socket is started before the service.
 systemctl start puma.socket puma.service
 
-# Check status of both socket and service.
+# Check the status of both socket and service.
 systemctl status puma.socket puma.service
 
 # A "hot" restart, with systemd keeping puma.socket listening and
@@ -197,8 +194,8 @@ systemctl restart puma.service
 systemctl restart puma.socket puma.service
 ~~~~
 
-Here is sample output from `systemctl status` with both service and
-socket running:
+Here is sample output from `systemctl status` with both service and socket
+running:
 
 ~~~~
 ● puma.socket - Puma HTTP Server Accept Sockets
@@ -231,14 +228,12 @@ Apr 07 08:40:19 hx puma[28320]: Use Ctrl-C to stop
 
 ### capistrano3-puma
 
-By default,
-[capistrano3-puma](https://github.com/seuros/capistrano-puma) uses
-`pumactl` for deployment restarts, outside of systemd.  To learn the
-exact commands that this tool would use for `ExecStart` and
-`ExecStop`, use the following `cap` commands in dry-run mode, and
-update from the above forking service configuration accordingly. Note
-also that the configured `User` should likely be the same as the
-capistrano3-puma `:puma_user` option.
+By default, [capistrano3-puma](https://github.com/seuros/capistrano-puma) uses
+`pumactl` for deployment restarts outside of systemd. To learn the exact
+commands that this tool would use for `ExecStart` and `ExecStop`, use the
+following `cap` commands in dry-run mode, and update from the above forking
+service configuration accordingly. Note also that the configured `User` should
+likely be the same as the capistrano3-puma `:puma_user` option.
 
 ~~~~ sh
 stage=production # or different stage, as needed
@@ -248,3 +243,4 @@ cap $stage puma:stop  --dry-run
 
 [Restart]: https://www.freedesktop.org/software/systemd/man/systemd.service.html#Restart=
 [#1367]: https://github.com/puma/puma/issues/1367
+[#1499]: https://github.com/puma/puma/issues/1499

data/docs/testing_benchmarks_local_files.md

@@ -0,0 +1,150 @@
+# Testing - benchmark/local files
+
+These files generate data that shows request-per-second (RPS), etc. Typically, files are in 
+pairs, a shell script and a Ruby script. The shell script starts the server, then runs the 
+Ruby file, which starts client request stream(s), then collects and logs metrics.
+
+## response_time_wrk.sh
+
+This uses [wrk] for generating data. One or more wrk runs are performed. Summarizes RPS and 
+wrk latency times. The default for the `-b` argument runs 28 different client request streams, 
+and takes a bit over 5 minutes.  See 'Request Stream Configuration' below for `-b` argument
+description.
+
+<details>
+  <summary>Summary output for<br/><code>benchmarks/local/response_time_wrk.sh -w2 -t5:5 -s tcp6</code>:</summary>
+
+```
+Type   req/sec    50%     75%     90%     99%    100%  Resp Size
+─────────────────────────────────────────────────────────────────    1kB
+array   13710    0.74    2.52    5.23    7.76   37.45      1024
+chunk   13502    0.76    2.55    5.28    7.84   11.23      1042
+string  13794    0.74    2.51    5.20    7.75   14.07      1024
+io       9615    1.16    3.45    7.13   10.57   15.75      1024
+─────────────────────────────────────────────────────────────────   10kB
+array   13458    0.76    2.57    5.31    7.93   13.94     10239
+chunk   13066    0.78    2.64    5.46    8.18   38.48     10320
+string  13500    0.76    2.55    5.29    7.88   11.42     10240
+io       9293    1.18    3.59    7.39   10.94   16.99     10240
+─────────────────────────────────────────────────────────────────  100kB
+array   11315    0.96    3.06    6.33    9.49   17.69    102424
+chunk    9916    1.10    3.48    7.20   10.73   15.14    103075
+string  10948    1.00    3.17    6.57    9.83   17.88    102378
+io       8901    1.21    3.72    7.48   11.27   59.98    102407
+─────────────────────────────────────────────────────────────────  256kB
+array    9217    1.15    3.82    7.88   11.74   17.12    262212
+chunk    7339    1.45    4.76    9.81   14.63   22.70    264007
+string   8574    1.19    3.81    7.73   11.21   15.80    262147
+io       8911    1.19    3.80    7.55   15.25   60.01    262183
+─────────────────────────────────────────────────────────────────  512kB
+array    6951    1.49    5.03   10.28   15.90   25.08    524378
+chunk    5234    2.03    6.56   13.57   20.46   32.15    527862
+string   6438    1.55    5.04   10.12   16.28   72.87    524275
+io       8533    1.15    4.62    8.79   48.15   70.51    524327
+───────────────────────────────────────────────────────────────── 1024kB
+array    4122    1.80   15.59   41.87   67.79  121.00   1048565
+chunk    3158    2.82   15.22   31.00   71.39   99.90   1055654
+string   4710    2.24    6.66   13.65   20.38   70.44   1048575
+io       8355    1.23    3.95    7.94   14.08   68.54   1048498
+───────────────────────────────────────────────────────────────── 2048kB
+array    2454    4.12   14.02   27.70   43.48   88.89   2097415
+chunk    1743    6.26   17.65   36.98   55.78   92.10   2111358
+string   2479    4.38   12.52   25.65   38.44   95.62   2097502
+io       8264    1.25    3.83    7.76   11.73   65.69   2097090
+
+Body    ────────── req/sec ──────────   ─────── req 50% times ───────
+ KB     array   chunk  string      io   array   chunk  string      io
+1       13710   13502   13794    9615   0.745   0.757   0.741   1.160
+10      13458   13066   13500    9293   0.760   0.784   0.759   1.180
+100     11315    9916   10948    8901   0.960   1.100   1.000   1.210
+256      9217    7339    8574    8911   1.150   1.450   1.190   1.190
+512      6951    5234    6438    8533   1.490   2.030   1.550   1.150
+1024     4122    3158    4710    8355   1.800   2.820   2.240   1.230
+2048     2454    1743    2479    8264   4.120   6.260   4.380   1.250
+─────────────────────────────────────────────────────────────────────
+wrk -t8 -c16 -d10s
+benchmarks/local/response_time_wrk.sh -w2 -t5:5 -s tcp6 -Y
+Server cluster mode -w2 -t5:5, bind: tcp6
+Puma repo branch 00-response-refactor
+ruby 3.2.0dev (2022-06-14T01:21:55Z master 048f14221c) +YJIT [x86_64-linux]
+
+[2136] - Gracefully shutting down workers...
+[2136] === puma shutdown: 2022-06-13 21:16:13 -0500 ===
+[2136] - Goodbye!
+
+ 5:15 Total Time
+```
+</details><br/>
+
+## bench_base.sh, bench_base.rb
+
+These two files setup parameters for the Puma server, which is normally started in a shell 
+script. It then starts a Ruby file (a subclass of BenchBase), passing arguments to it. The 
+Ruby file is normally used to generate a client request stream(s).
+
+### Puma Configuration
+
+The following arguments are used for the Puma server:
+
+* **`-C`** - configuration file
+* **`-d`** - app delay
+* **`-r`** - rackup file, often defaults to test/rackup/ci_select.ru
+* **`-s`** - bind socket type, default is tcp/tcp4, also tcp6, ssl/ssl4, ssl6, unix, or aunix
+  (unix & abstract unix are not available with wrk).
+* **`-t`** - threads, expressed as '5:5', same as Puma --thread
+* **`-w`** - workers, same as Puma --worker
+* **`-Y`** - enable Ruby YJIT
+
+### Request Stream Configuration
+
+The following arguments are used for request streams:
+
+* **`-b`** - response body configuration. Body type options are a array, c chunked, s string,
+  and i for File/IO. None or any combination can be specified, they should start the option.
+  Then, any combination of comma separated integers can be used for the response body size
+  in kB. The string 'ac50,100' would create four runs, 50kb array, 50kB chunked, 100kB array,
+  and 100kB chunked. See 'Testing - test/rackup/ci-*.ru files' for more info.
+* **`-c`** - connections per client request stream thread, defaults to 2 for wrk.
+* **`-D`** - duration of client request stream in seconds.
+* **`-T`** - number of threads in the client request stream. For wrk, this defaults to
+  80% of Puma workers * max_threads.
+
+### Notes - Configuration
+
+The above lists script arguments.
+
+`bench_base.sh` contains most server defaults. Many can be set via ENV variables.
+
+`bench_base.rb` contains the client request stream defaults. The default value for
+`-b` is `acsi1,10,100,256,512,1024,2048`, which is a 4 x 7 matrix, and hence, runs
+28 jobs. Also, the i body type (File/IO) generates files, they are placed in the
+`"#{Dir.tmpdir}/.puma_response_body_io"` directory, which is created.
+
+### Notes - wrk
+
+The shell scripts use `-T` for wrk's thread count, since `-t` is used for Puma
+server threads.  Regarding the `-c` argument, wrk has an interesting behavior.
+The total number of connections is set by `(connections/threads).to_i`. The scripts
+here use `-c` as connections per thread.  Hence, using `-T4 -c2` will yield a total
+of eight wrk connections, two per thread. The equivalent wrk arguments would be `-t4 -c8`.
+
+Puma can only process so many requests, and requests will queue in the backlog
+until Puma can respond to them. With wrk, if the number of total connections is
+too high, one will see the upper latency times increase, pushing into the lower
+latency times as the connections are increased. The default values for wrk's
+threads and connections were chosen to minimize requests' time in the backlog.
+
+An example with four wrk runs using `-b s10`.  Notice that `req/sec` varies by
+less than 1%, but the `75%` times increase by an order of magnitude:
+```
+req/sec    50%     75%     90%     99%    100%  Resp Size   wrk cmd line
+─────────────────────────────────────────────────────────────────────────────
+ 13597   0.755   2.550   5.260   7.800  13.310     12040    wrk -t8  -c16 -d10
+ 13549   0.793   4.430   8.140  11.220  16.600     12002    wrk -t10 -c20 -d10
+ 13570   1.040  25.790  40.010  49.070  58.300     11982    wrk -t8  -c64 -d10
+ 13684   1.050  25.820  40.080  49.160  66.190     12033    wrk -t16 -c64 -d10
+```
+Finally, wrk's output may cause rounding errors, so the response body size calculation is
+imprecise.
+
+[wrk]: <https://github.com/ioquatix/wrk>