Class: Unicorn::Worker
Relationships & Source Files | |
Inherits: | Object |
Defined in: | lib/unicorn/worker.rb |
Overview
This class and its members can be considered a stable interface and will not change in a backwards-incompatible fashion between releases of unicorn. Knowledge of this class is generally not not needed for most users of unicorn.
Some users may want to access it in the before_fork/after_fork hooks. See the Configurator
RDoc for examples.
Constant Summary
-
DROPS =
Internal use only
# File 'lib/unicorn/worker.rb', line 19[]
-
PER_DROP =
Internal use only
# File 'lib/unicorn/worker.rb', line 18Raindrops::PAGE_SIZE / Raindrops::SIZE
Class Method Summary
- .new(nr, pipe = nil) ⇒ Worker constructor Internal use only
Instance Attribute Summary
Instance Method Summary
-
#user(user, group = nil, chroot = false)
In most cases, you should be using the Configurator#user directive instead.
-
#==(other_nr)
Internal use only
worker objects may be compared to just plain Integers.
-
#accept_nonblock(*_unused)
Internal use only
this only runs when the Rack app.call is not running act like
Socket#accept_nonblock
(exception: false). - #atfork_child Internal use only
-
#atfork_parent
Internal use only
parent does not read.
-
#close
Internal use only
called in both the master (reaping worker) and worker (SIGQUIT handler).
-
#fake_sig(sig)
Internal use only
call a signal handler immediately without triggering EINTR We do not use the more obvious
Process.kill
(sig, $$) here since that signal delivery may be deferred. -
#quit
Internal use only
master fakes SIGQUIT using this.
-
#soft_kill(sig)
Internal use only
master sends fake signals to children.
Constructor Details
.new(nr, pipe = nil) ⇒ Worker
Instance Attribute Details
#master (readonly)
# File 'lib/unicorn/worker.rb', line 16
attr_reader :master
#nr (rw)
# File 'lib/unicorn/worker.rb', line 14
attr_accessor :nr, :switched
#switched (rw)
# File 'lib/unicorn/worker.rb', line 14
attr_accessor :nr, :switched
#tick (rw)
called in the master process
# File 'lib/unicorn/worker.rb', line 101
def tick # :nodoc: @raindrop[@offset] end
#tick=(value) (rw)
called in the worker process
# File 'lib/unicorn/worker.rb', line 96
def tick=(value) # :nodoc: @raindrop[@offset] = value end
#to_io (readonly)
IO.select-compatible
# File 'lib/unicorn/worker.rb', line 15
attr_reader :to_io # IO.select-compatible
Instance Method Details
#==(other_nr)
worker objects may be compared to just plain Integers
# File 'lib/unicorn/worker.rb', line 91
def ==(other_nr) # :nodoc: @nr == other_nr end
#accept_nonblock(*_unused)
this only runs when the Rack app.call is not running act like Socket#accept_nonblock
(exception: false)
# File 'lib/unicorn/worker.rb', line 76
def accept_nonblock(*_unused) # :nodoc: case buf = @to_io.read_nonblock(4, exception: false) when String # unpack the buffer and trigger the signal handler signum = buf.unpack('l') fake_sig(signum[0]) # keep looping, more signals may be queued when nil # EOF: master died, but we are at a safe place to exit fake_sig(:QUIT) when :wait_readable # keep waiting return :wait_readable end while true # loop, as multiple signals may be sent end
#atfork_child
# File 'lib/unicorn/worker.rb', line 31
def atfork_child # :nodoc: # we _must_ close in child, parent just holds this open to signal @master = @master.close end
#atfork_parent
parent does not read
# File 'lib/unicorn/worker.rb', line 42
def atfork_parent # :nodoc: @to_io = @to_io.close end
#close
called in both the master (reaping worker) and worker (SIGQUIT handler)
# File 'lib/unicorn/worker.rb', line 106
def close # :nodoc: @master.close if @master @to_io.close if @to_io end
#fake_sig(sig)
call a signal handler immediately without triggering EINTR We do not use the more obvious Process.kill
(sig, $$) here since that signal delivery may be deferred. We want to avoid signal delivery while the Rack app.call is running because some database drivers (e.g. ruby-pg) may cancel pending requests.
# File 'lib/unicorn/worker.rb', line 51
def fake_sig(sig) # :nodoc: old_cb = trap(sig, "IGNORE") old_cb.call ensure trap(sig, old_cb) end
#quit
master fakes SIGQUIT using this
# File 'lib/unicorn/worker.rb', line 37
def quit # :nodoc: @master = @master.close if @master end
#soft_kill(sig)
master sends fake signals to children
# File 'lib/unicorn/worker.rb', line 59
def soft_kill(sig) # :nodoc: case sig when Integer signum = sig else signum = Signal.list[sig.to_s] or raise ArgumentError, "BUG: bad signal: #{sig.inspect}" end # writing and reading 4 bytes on a pipe is atomic on all POSIX platforms # Do not care in the odd case the buffer is full, here. @master.write_nonblock([signum].pack('l'), exception: false) rescue Errno::EPIPE # worker will be reaped soon end
#user(user, group = nil, chroot = false)
In most cases, you should be using the Configurator#user directive instead. This method should only be used if you need fine-grained control of exactly when you want to change permissions in your after_fork or after_worker_ready hooks, or if you want to use the chroot support.
Changes the worker process to the specified user
and group
, and chroots to the current working directory if chroot
is set. This is only intended to be called from within the worker process from the after_fork
hook. This should be called in the after_fork
hook after any privileged functions need to be run (e.g. to set per-worker CPU affinity, niceness, etc)
group
can be specified as a string, or as an array of two strings. If an array of two strings is given, the first string is used as the primary group of the process, and the second is used as the group of the log files.
Any and all errors raised within this method will be propagated directly back to the caller (usually the after_fork
hook. These errors commonly include ArgumentError for specifying an invalid user/group and Errno::EPERM
for insufficient privileges.
chroot support is only available in unicorn 5.3.0+ user and group switching appeared in unicorn 0.94.0 (2009-11-05)
# File 'lib/unicorn/worker.rb', line 138
def user(user, group = nil, chroot = false) # we do not protect the caller, checking Process.euid == 0 is # insufficient because modern systems have fine-grained # capabilities. Let the caller handle any and all errors. uid = Etc.getpwnam(user).uid if group if group.is_a?(Array) group, log_group = group log_gid = Etc.getgrnam(log_group).gid end gid = Etc.getgrnam(group).gid log_gid ||= gid end Unicorn::Util.chown_logs(uid, log_gid) if gid && Process.egid != gid Process.initgroups(user, gid) Process::GID.change_privilege(gid) end if chroot chroot = Dir.pwd if chroot == true Dir.chroot(chroot) Dir.chdir('/') end Process.euid != uid and Process::UID.change_privilege(uid) @switched = true end