Class: Unicorn::Configurator

Inherits:
Struct
  • Object
show all
Defined in:
lib/unicorn/configurator.rb

Overview

Implements a simple DSL for configuring a Unicorn server.

See unicorn.bogomips.org/examples/unicorn.conf.rb for an example config file. An example config file for use with nginx is also available at unicorn.bogomips.org/examples/nginx.conf

Constant Summary collapse

DEFAULTS =

Default settings for Unicorn

{
  # Backward compatibility soft timeout (disabled in default configuration)
  :soft_timeout => 60,
  :timeout => 60,
  :logger => Logger.new($stderr),
  :worker_processes => 1,
  :after_fork => lambda { |server, worker|
      server.logger.info("worker=#{worker.nr} spawned pid=#{$$}")
    },
  :before_fork => lambda { |server, worker|
      server.logger.info("worker=#{worker.nr} spawning...")
    },
  :before_exec => lambda { |server|
      server.logger.info("forked child re-executing...")
    },
  :pid => nil,
  :preload_app => false,
}

Instance Attribute Summary collapse

Instance Method Summary collapse

Constructor Details

#initialize(defaults = {}) ⇒ Configurator

:nodoc:



35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
 
# File 'lib/unicorn/configurator.rb', line 35

def initialize(defaults = {}) #:nodoc:
  self.set = Hash.new(:unset)
  use_defaults = defaults.delete(:use_defaults)
  self.config_file = defaults.delete(:config_file)

  # after_reload is only used by unicorn_rails, unsupported otherwise
  self.after_reload = defaults.delete(:after_reload)

  set.merge!(DEFAULTS) if use_defaults
  defaults.each { |key, value| self.send(key, value) }
  Hash === set[:listener_opts] or
      set[:listener_opts] = Hash.new { |hash,key| hash[key] = {} }
  Array === set[:listeners] or set[:listeners] = []
  reload
end

Instance Attribute Details

#after_reload ⇒ Object

Returns the value of attribute after_reload

Returns:

  • (Object) —

    the current value of after_reload



13
14
15
 
# File 'lib/unicorn/configurator.rb', line 13

def after_reload
  @after_reload
end

#config_file ⇒ Object

Returns the value of attribute config_file

Returns:

  • (Object) —

    the current value of config_file



13
14
15
 
# File 'lib/unicorn/configurator.rb', line 13

def config_file
  @config_file
end

#set ⇒ Object

Returns the value of attribute set

Returns:

  • (Object) —

    the current value of set



13
14
15
 
# File 'lib/unicorn/configurator.rb', line 13

def set
  @set
end

Instance Method Details

#[](key) ⇒ Object

:nodoc:



76
77
78
 
# File 'lib/unicorn/configurator.rb', line 76

def [](key) # :nodoc:
  set[key]
end

#after_fork(*args, &block) ⇒ Object

sets after_fork hook to a given block. This block will be called by the worker after forking. The following is an example hook which adds a per-process listener to every worker:

after_fork do |server,worker|
  # per-process listener ports for debugging/admin:
  addr = "127.0.0.1:#{9293 + worker.nr}"

  # the negative :tries parameter indicates we will retry forever
  # waiting on the existing process to exit with a 5 second :delay
  # Existing options for Unicorn::Configurator#listen such as
  # :backlog, :rcvbuf, :sndbuf are available here as well.
  server.listen(addr, :tries => -1, :delay => 5, :backlog => 128)

  # drop permissions to "www-data" in the worker
  # generally there's no reason to start Unicorn as a priviledged user
  # as it is not recommended to expose Unicorn to public clients.
  worker.user('www-data', 'www-data') if Process.euid == 0
end



111
112
113
 
# File 'lib/unicorn/configurator.rb', line 111

def after_fork(*args, &block)
  set_hook(:after_fork, block_given? ? block : args[0])
end

#before_exec(*args, &block) ⇒ Object

sets the before_exec hook to a given Proc object. This Proc object will be called by the master process right before exec()-ing the new unicorn binary. This is useful for freeing certain OS resources that you do NOT wish to share with the reexeced child process. There is no corresponding after_exec hook (for obvious reasons).



128
129
130
 
# File 'lib/unicorn/configurator.rb', line 128

def before_exec(*args, &block)
  set_hook(:before_exec, block_given? ? block : args[0], 1)
end

#before_fork(*args, &block) ⇒ Object

sets before_fork got be a given Proc object. This Proc object will be called by the master process before forking each worker.



118
119
120
 
# File 'lib/unicorn/configurator.rb', line 118

def before_fork(*args, &block)
  set_hook(:before_fork, block_given? ? block : args[0])
end

#commit!(server, options = {}) ⇒ Object

:nodoc:



67
68
69
70
71
72
73
74
 
# File 'lib/unicorn/configurator.rb', line 67

def commit!(server, options = {}) #:nodoc:
  skip = options[:skip] || []
  set.each do |key, value|
    value == :unset and next
    skip.include?(key) and next
    server.__send__("#{key}=", value)
  end
end

#expand_addr(address) ⇒ Object

expands “unix:path/to/foo” to a socket relative to the current path expands pathnames of sockets if relative to “~” or “~username” expands “*:port and ”:port“ to ”0.0.0.0:port“



373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
 
# File 'lib/unicorn/configurator.rb', line 373

def expand_addr(address) #:nodoc
  return "0.0.0.0:#{address}" if Integer === address
  return address unless String === address

  case address
  when %r{\Aunix:(.*)\z}
    File.expand_path($1)
  when %r{\A~}
    File.expand_path(address)
  when %r{\A(?:\*:)?(\d+)\z}
    "0.0.0.0:#$1"
  when %r{\A(.*):(\d+)\z}
    # canonicalize the name
    packed = Socket.pack_sockaddr_in($2.to_i, $1)
    Socket.unpack_sockaddr_in(packed).reverse!.join(':')
  else
    address
  end
end

#listen(address, opt = {}) ⇒ Object

adds an address to the existing listener set.

The following options may be specified (but are generally not needed):

:backlog: this is the backlog of the listen() syscall.

Some operating systems allow negative values here to specify the maximum allowable value. In most cases, this number is only recommendation and there are other OS-specific tunables and variables that can affect this number. See the listen(2) syscall documentation of your OS for the exact semantics of this.

If you are running unicorn on multiple machines, lowering this number can help your load balancer detect when a machine is overloaded and give requests to a different machine.

Default: 1024

:rcvbuf, :sndbuf: maximum receive and send buffer sizes of sockets

These correspond to the SO_RCVBUF and SO_SNDBUF settings which can be set via the setsockopt(2) syscall. Some kernels (e.g. Linux 2.4+) have intelligent auto-tuning mechanisms and there is no need (and it is sometimes detrimental) to specify them.

See the socket API documentation of your operating system to determine the exact semantics of these settings and other operating system-specific knobs where they can be specified.

Defaults: operating system defaults

:tcp_nodelay: disables Nagle’s algorithm on TCP sockets

This has no effect on UNIX sockets.

Default: operating system defaults (usually Nagle’s algorithm enabled)

:tcp_nopush: enables TCP_CORK in Linux or TCP_NOPUSH in FreeBSD

This will prevent partial TCP frames from being sent out. Enabling tcp_nopush is generally not needed or recommended as controlling tcp_nodelay already provides sufficient latency reduction whereas Unicorn does not know when the best times are for flushing corked sockets.

This has no effect on UNIX sockets.

:tries: times to retry binding a socket if it is already in use

A negative number indicates we will retry indefinitely, this is useful for migrations and upgrades when individual workers are binding to different ports.

Default: 5

:delay: seconds to wait between successive tries

Default: 0.5 seconds

:umask: sets the file mode creation mask for UNIX sockets

Typically UNIX domain sockets are created with more liberal file permissions than the rest of the application. By default, we create UNIX domain sockets to be readable and writable by all local users to give them the same accessibility as locally-bound TCP listeners.

This has no effect on TCP listeners.

Default: 0 (world read/writable)



279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
 
# File 'lib/unicorn/configurator.rb', line 279

def listen(address, opt = {})
  address = expand_addr(address)
  if String === address
    [ :umask, :backlog, :sndbuf, :rcvbuf, :tries ].each do |key|
      value = opt[key] or next
      Integer === value or
        raise ArgumentError, "not an integer: #{key}=#{value.inspect}"
    end
    [ :tcp_nodelay, :tcp_nopush ].each do |key|
      (value = opt[key]).nil? and next
      TrueClass === value || FalseClass === value or
        raise ArgumentError, "not boolean: #{key}=#{value.inspect}"
    end
    unless (value = opt[:delay]).nil?
      Numeric === value or
        raise ArgumentError, "not numeric: delay=#{value.inspect}"
    end
    set[:listener_opts][address].merge!(opt)
  end

  set[:listeners] << address
end

#listeners(addresses) ⇒ Object

sets listeners to the given addresses, replacing or augmenting the current set. This is for the global listener pool shared by all worker processes. For per-worker listeners, see the after_fork example This is for internal API use only, do not use it in your Unicorn config file. Use listen instead.



201
202
203
204
205
 
# File 'lib/unicorn/configurator.rb', line 201

def listeners(addresses) # :nodoc:
  Array === addresses or addresses = Array(addresses)
  addresses.map! { |addr| expand_addr(addr) }
  set[:listeners] = addresses
end

#logger(new) ⇒ Object

sets object to the new Logger-like object. The new logger-like object must respond to the following methods:

+debug+, +info+, +warn+, +error+, +fatal+, +close+



83
84
85
86
87
88
89
90
 
# File 'lib/unicorn/configurator.rb', line 83

def logger(new)
  %w(debug info warn error fatal close).each do |m|
    new.respond_to?(m) and next
    raise ArgumentError, "logger=#{new} does not respond to method=#{m}"
  end

  set[:logger] = new
end

#pid(path) ⇒ Object

sets the path for the PID file of the unicorn master process



303
 
# File 'lib/unicorn/configurator.rb', line 303

def pid(path); set_path(:pid, path); end

#preload_app(bool) ⇒ Object

Enabling this preloads an application before forking worker processes. This allows memory savings when using a copy-on-write-friendly GC but can cause bad things to happen when resources like sockets are opened at load time by the master process and shared by multiple children. People enabling this are highly encouraged to look at the before_fork/after_fork hooks to properly close/reopen sockets. Files opened for logging do not have to be reopened as (unbuffered-in-userspace) files opened with the File::APPEND flag are written to atomically on UNIX.

In addition to reloading the unicorn-specific config settings, SIGHUP will reload application code in the working directory/symlink when workers are gracefully restarted.



318
319
320
321
322
323
324
325
 
# File 'lib/unicorn/configurator.rb', line 318

def preload_app(bool)
  case bool
  when TrueClass, FalseClass
    set[:preload_app] = bool
  else
    raise ArgumentError, "preload_app=#{bool.inspect} not a boolean"
  end
end

#reload ⇒ Object

:nodoc:



51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
 
# File 'lib/unicorn/configurator.rb', line 51

def reload #:nodoc:
  instance_eval(File.read(config_file), config_file) if config_file

  # working_directory binds immediately (easier error checking that way),
  # now ensure any paths we changed are correctly set.
  [ :pid, :stderr_path, :stdout_path ].each do |var|
    String === (path = set[var]) or next
    path = File.expand_path(path)
    test(?w, path) || test(?w, File.dirname(path)) or \
          raise ArgumentError, "directory for #{var}=#{path} not writable"
  end

  # unicorn_rails creates dirs here after working_directory is bound
  after_reload.call if after_reload
end

#soft_timeout(seconds) ⇒ Object

sets the timeout of worker processes to seconds. Workers handling the request/app.call/response cycle taking longer than this time period will be softly killed (via SIGABRT). This timeout is enforced by the master process itself and not subject to the scheduling limitations by the worker process. Due the low-complexity, low-overhead implementation, timeouts of less than 3.0 seconds can be considered inaccurate and unsafe. ABORT is handled by the worker and raise an exception, offering a way to log the stack trace in your rails application.



142
143
144
145
146
147
148
 
# File 'lib/unicorn/configurator.rb', line 142

def soft_timeout(seconds)
  Numeric === seconds or raise ArgumentError,
                              "not numeric: timeout=#{seconds.inspect}"
  seconds >= 3 or raise ArgumentError,
                              "too low: timeout=#{seconds.inspect}"
  set[:soft_timeout] = seconds
end

#stderr_path(path) ⇒ Object

Allow redirecting $stderr to a given path. Unlike doing this from the shell, this allows the unicorn process to know the path its writing to and rotate the file if it is used for logging. The file will be opened with the File::APPEND flag and writes synchronized to the kernel (but not necessarily to disk) so multiple processes can safely append to it.



333
334
335
 
# File 'lib/unicorn/configurator.rb', line 333

def stderr_path(path)
  set_path(:stderr_path, path)
end

#stdout_path(path) ⇒ Object

Same as stderr_path, except for $stdout



338
339
340
 
# File 'lib/unicorn/configurator.rb', line 338

def stdout_path(path)
  set_path(:stdout_path, path)
end

#timeout(seconds) ⇒ Object

sets the timeout of worker processes to seconds. Workers handling the request/app.call/response cycle taking longer than this time period will be forcibly killed (via SIGKILL). This timeout is enforced by the master process itself and not subject to the scheduling limitations by the worker process. Due the low-complexity, low-overhead implementation, timeouts of less than 3.0 seconds can be considered inaccurate and unsafe.

For running Unicorn behind nginx, it is recommended to set “fail_timeout=0” for in your nginx configuration like this to have nginx always retry backends that may have had workers SIGKILL-ed due to timeouts.

# See http://wiki.nginx.org/NginxHttpUpstreamModule for more details
# on nginx upstream configuration:
upstream unicorn_backend {
  # for UNIX domain socket setups:
  server unix:/path/to/unicorn.sock fail_timeout=0;

  # for TCP setups
  server 192.168.0.7:8080 fail_timeout=0;
  server 192.168.0.8:8080 fail_timeout=0;
  server 192.168.0.9:8080 fail_timeout=0;
}



173
174
175
176
177
178
179
 
# File 'lib/unicorn/configurator.rb', line 173

def timeout(seconds)
  Numeric === seconds or raise ArgumentError,
                              "not numeric: timeout=#{seconds.inspect}"
  seconds >= 3 or raise ArgumentError,
                              "too low: timeout=#{seconds.inspect}"
  set[:timeout] = seconds
end

#user(user, group = nil) ⇒ Object

Runs worker processes as the specified user and group. The master process always stays running as the user who started it. This switch will occur after calling the after_fork hook, and only if the Worker#user method is not called in the after_fork hook



363
364
365
366
367
368
 
# File 'lib/unicorn/configurator.rb', line 363

def user(user, group = nil)
  # raises ArgumentError on invalid user/group
  Etc.getpwnam(user)
  Etc.getgrnam(group) if group
  set[:user] = [ user, group ]
end

#worker_processes(nr) ⇒ Object

sets the current number of worker_processes to nr. Each worker process will serve exactly one client at a time. You can increment or decrement this value at runtime by sending SIGTTIN or SIGTTOU respectively to the master process without reloading the rest of your Unicorn configuration. See the SIGNALS document for more information.



188
189
190
191
192
193
194
 
# File 'lib/unicorn/configurator.rb', line 188

def worker_processes(nr)
  Integer === nr or raise ArgumentError,
                         "not an integer: worker_processes=#{nr.inspect}"
  nr >= 0 or raise ArgumentError,
                         "not non-negative: worker_processes=#{nr.inspect}"
  set[:worker_processes] = nr
end

#working_directory(path) ⇒ Object

sets the working directory for Unicorn. This ensures USR2 will start a new instance of Unicorn in this directory. This may be a symlink.



345
346
347
348
349
350
351
352
353
354
355
356
357
 
# File 'lib/unicorn/configurator.rb', line 345

def working_directory(path)
  # just let chdir raise errors
  path = File.expand_path(path)
  if config_file &&
     config_file[0] != ?/ &&
     ! test(?r, "#{path}/#{config_file}")
    raise ArgumentError,
          "config_file=#{config_file} would not be accessible in" \
          " working_directory=#{path}"
  end
  Dir.chdir(path)
  HttpServer::START_CTX[:cwd] = ENV["PWD"] = path
end