++ed by:
Graham Barr

NAME

Beanstalk::Client - Client class to talk to beanstalkd server

SYNOPSIS

  use Beanstalk::Client;

  my $client = Beanstalk::Client->new(
    { server       => "localhost",
      default_tube => 'mine',
    }
  );

  # Send a job with explicit data
  my $job = $client->put(
    { data     => "data",
      priority => 100,
      ttr      => 120,
      delay    => 5,
    }
  );

  # Send job, data created by encoding @args. By default with YAML
  my $job2 = $client->put(
    { priority => 100,
      ttr      => 120,
      delay    => 5,
    },
    @args
  );

  # Send job, data created by encoding @args with JSON
  use JSON::XS;
  $client->encoder(sub { encode_json(\@_) });
  my $job2 = $client->put(
    { priority => 100,
      ttr      => 120,
      delay    => 5,
    },
    @args
  );

  # fetch a job
  my $job3 = $client->reserve;

DESCRIPTION

Beanstalk::Client provides a Perl API of protocol version 1.0 to the beanstalkd server, a fast, general-purpose, in-memory workqueue service by Keith Rarick.

METHODS

Constructor

new ($options)

The constructor accepts a single argument, which is a reference to a hash containing options. The options can be any of the accessor methods listed below.

Accessor Methods

server ([$hostname])

Get/set the hostname, and port, to connect to. The port, which defaults to 11300, can be specified by appending it to the hostname with a : (eg "localhost:1234"). (Default: localhost:11300)

socket

Get the socket connection to the server.

delay ([$delay])

Set/get a default value, in seconds, for job delay. A job with a delay will be placed into a delayed state and will not be placed into the ready queue until the time period has passed. This value will be used by put and release as a default. (Default: 0)

ttr ([$ttr])

Set/get a default value, in seconds, for job ttr (time to run). This value will be used by put as a default. (Default: 120)

priority ([$priority])

Set/get a default value for job priority. The highest priority job is the job where the priority value is the lowest (ie jobs with a lower priority value are run first). This value will be used by put, release and bury as a default. (Default: 10000)

encoder ([$encoder])

Set/get serialization encoder. $encoder is a reference to a subroutine that will be called when arguments to put need to be encoded to send to the beanstalkd server. The subroutine should accept a list of arguments and return a string representation to pass to the server. (Default: YAML::Syck::Dump)

decoder ([$decoder])

Set/get the serialization decoder. $decoder is a reference to a subroutine that will be called when data from the beanstalkd server needs to be decoded. The subroutine will be passed the data fetched from the beanstalkd server and should return a list of values the application can use. (Default: YAML::Syck::Load)

error

Fetch the last error that happened.

connect_timeout ([$timeout])

Get/set timeout, in seconds, to use for the connect to the server.

default_tube ([$tube])

Set/get the name of a default tube to put jobs into and fetch from.

By default a connection to a beanstalkd server will put into the default queue and also watch the default queue. If default_tube is set when connect is called the connection will be initialized so that put will put into the given tube and reserve will fetch jobs from the given tube. (Default: none)

debug ([$debug])

Set/get debug value. If set to a true value then all communication with the server will be output with warn

Producer Methods

These methods are used by clients that are placing work into the queue

put ($options [, @args])

Insert job into the currently used tube. Options may be

priority

priority to use to queue the job. Jobs with smaller priority values will be scheduled before jobs with larger priorities. The most urgent priority is 0

Defaults to $self-priority>

delay

An integer number of seconds to wait before putting the job in the ready queue. The job will be in the "delayed" state during this time

Defaults to $self-delay>

ttr

"time to run" - An integer number of seconds to allow a worker to run this job. This time is counted from the moment a worker reserves this job. If the worker does not delete, release, or bury the job within ttr seconds, the job will time out and the server will release the job. The minimum ttr is 1. If the client sends 0, the server will silently increase the ttr to 1.

data

The job body. Defaults to the result of calling the current encoder passing @args

use ($tube)

Change tube that new jobs are inserted into

Worker Methods

reserve ([$timeout])

Reserve a job from the list of tubes currently being watched.

Returns a Beanstalk::Job on success. $timeout is the maximum number of seconds to wait for a job to become ready. If $timeout is not given then the client will wait indefinitely.

Returns undef on error or if $timeout expires.

delete ($id)

Delete the specified job.

release ($id, [, $options])

Release the specified job. Valid options are

priority

New priority to assign to the job

delay

An integer number of seconds to wait before putting the job in the ready queue. The job will be in the "delayed" state during this time

bury ($id [, $options])

The bury command puts a job into the "buried" state. Buried jobs are put into a FIFO linked list and will not be touched by the server again until a client kicks them with the "kick" command.

Valid options are

priority

New priority to assign to the job

touch ($id)

Calling touch with the id of a reserved job will reset the time left for the job to complete back to the original ttr value.

watch ($tube)

Specifies a tube to add to the watch list. If the tube doesn't exist, it will be created

ignore ($tube)

Stop watching $tube

watch_only (@tubes)

Watch only the list of given tube names

Other Methods

connect

Connect to server. If sucessful, set the tube to use and tube to watch if a default_tube was specified.

disconnect

Disconnect from server. socket method will return undef.

quit

Disconnect from server. socket method will return undef.

peek ($id)

Peek at the job id specified. If the job exists returns a Beanstalk::Job object. Returns undef on error or if job does not exist.

peek_ready

Peek at the first job that is in the ready queue. If there is a job in the ready queue returns a Beanstalk::Job object. Returns undef on error or if there are no ready jobs.

peek_delayed

Peek at the first job that is in the delayed queue. If there is a job in the delayed queue returns a Beanstalk::Job object. Returns undef on error or if there are no delayed jobs.

peek_buried

Peek at the first job that is in the buried queue. If there is a job in the buried queue returns a Beanstalk::Job object. Returns undef on error or if there are no buried jobs.

kick ($bound)

The kick command applies only to the currently used tube. It moves jobs into the ready queue. If there are any buried jobs, it will only kick buried jobs. Otherwise it will kick delayed jobs. The server will not kick more than $bound jobs. Returns the number of jobs kicked, or undef if there was an error.

kick_job ($id)

The kick-job command is a variant of kick that operats with a single job identified by its job id. If the given job id exists and is in a buried or delayed state, it will be moved to the ready queue of the same tube where it currently belongs. Returns undef on error.

Note: the kick_job command was only introduced on version 1.8 of beanstalk. If you have a version of beanstalk prior to this then the command will return an error.

stats_job ($id)

Return stats for the specified job $id. Returns undef on error.

If the job exists, the return will be a Stats object with the following methods

  • id - The job id

  • tube - The name of the tube that contains this job

  • state - is "ready" or "delayed" or "reserved" or "buried"

  • pri - The priority value set by the put, release, or bury commands.

  • age - The time in seconds since the put command that created this job.

  • time_left - The number of seconds left until the server puts this job into the ready queue. This number is only meaningful if the job is reserved or delayed. If the job is reserved and this amount of time elapses before its state changes, it is considered to have timed out.

  • reserves - The number of times this job has been reserved

  • timeouts - The number of times this job has timed out during a reservation.

  • releases - The number of times a client has released this job from a reservation.

  • buries - The number of times this job has been buried.

  • kicks - The number of times this job has been kicked.

stats_tube ($tube)

Return stats for the specified tube $tube. Returns undef on error.

If the tube exists, the return will be a Stats object with the following methods

  • name - The tube's name.

  • current_jobs_urgent - The number of ready jobs with priority < 1024 in this tube.

  • current_jobs_ready - The number of jobs in the ready queue in this tube.

  • current_jobs_reserved - The number of jobs reserved by all clients in this tube.

  • current_jobs_delayed - The number of delayed jobs in this tube.

  • current_jobs_buried - The number of buried jobs in this tube.

  • total_jobs - The cumulative count of jobs created in this tube.

  • current_waiting - The number of open connections that have issued a reserve command while watching this tube but not yet received a response.

  • pause - The number of seconds the tube has been paused for.

  • cmd_pause_tube - The cumulative number of pause-tube commands for this tube.

  • pause_time_left - The number of seconds until the tube is un-paused.

stats
  • current_jobs_urgent - The number of ready jobs with priority < 1024.

  • current_jobs_ready - The number of jobs in the ready queue.

  • current_jobs_reserved - The number of jobs reserved by all clients.

  • current_jobs_delayed - The number of delayed jobs.

  • current_jobs_buried - The number of buried jobs.

  • cmd_put - The cumulative number of put commands.

  • cmd_peek - The cumulative number of peek commands.

  • cmd_peek_ready - The cumulative number of peek-ready commands.

  • cmd_peek_delayed - The cumulative number of peek-delayed commands.

  • cmd_peek_buried - The cumulative number of peek-buried commands.

  • cmd_reserve - The cumulative number of reserve commands.

  • cmd_use - The cumulative number of use commands.

  • cmd_watch - The cumulative number of watch commands.

  • cmd_ignore - The cumulative number of ignore commands.

  • cmd_delete - The cumulative number of delete commands.

  • cmd_release - The cumulative number of release commands.

  • cmd_bury - The cumulative number of bury commands.

  • cmd_kick - The cumulative number of kick commands.

  • cmd_stats - The cumulative number of stats commands.

  • cmd_stats_job - The cumulative number of stats-job commands.

  • cmd_stats_tube - The cumulative number of stats-tube commands.

  • cmd_list_tubes - The cumulative number of list-tubes commands.

  • cmd_list_tube_used - The cumulative number of list-tube-used commands.

  • cmd_list_tubes_watched - The cumulative number of list-tubes-watched commands.

  • cmd_pause_tube - The cumulative number of pause-tube commands

  • job_timeouts - The cumulative count of times a job has timed out.

  • total_jobs - The cumulative count of jobs created.

  • max_job_size - The maximum number of bytes in a job.

  • current_tubes - The number of currently-existing tubes.

  • current_connections - The number of currently open connections.

  • current_producers - The number of open connections that have each issued at least one put command.

  • current_workers - The number of open connections that have each issued at least one reserve command.

  • current_waiting - The number of open connections that have issued a reserve command but not yet received a response.

  • total_connections - The cumulative count of connections.

  • pid - The process id of the server.

  • version - The version string of the server.

  • rusage_utime - The accumulated user CPU time of this process in seconds and microseconds.

  • rusage_stime - The accumulated system CPU time of this process in seconds and microseconds.

  • uptime - The number of seconds since this server started running.

  • binlog_oldest_index - The index of the oldest binlog file needed to store the current jobs

  • binlog_current_index - The index of the current binlog file being written to. If binlog is not active this value will be 0

  • binlog_max_size - The maximum size in bytes a binlog file is allowed to get before a new binlog file is opened

list_tubes

Returns a list of tubes

list_tube_used

Returns the current tube being used. This is the tube which put will place jobs.

list_tubes_watched

Returns a list of tubes being watched, or the number of tubes being watched in a scalar context. These are the tubes that reserve will check to find jobs. On error an empty list, or undef in a scalar context, will be returned.

pause_tube ($tube, $delay)

Pause from reserving any jobs in $tube for $delay seconds.

Returns true on success and undef on error.

TODO

More tests

ACKNOWLEDGEMTS

Large parts of this documention were lifted from the documention that comes with beanstalkd

SEE ALSO

http://kr.github.com/beanstalkd/

Beanstalk::Pool, Beanstalk::Job, Beanstalk::Stats

AUTHOR

Graham Barr <gbarr@pobox.com>

CREDITS

  • Ask Bjørn Hansen

  • Rhesa Rozendaal

  • Ian Docherty

BUGS

The Git repository is available at http://github.com/gbarr/perl-beanstalk-client

Please report any bugs or feature requests to the issue tracker at http://github.com/gbarr/perl-beanstalk-client/issues

COPYRIGHT

Copyright (C) 2008-2012 by Graham Barr.

This program is free software; you can redistribute it and/or modify it under the same terms as Perl itself.




Hosting generously
sponsored by Bytemark