Scroll to navigation

Minion::Backend(3) User Contributed Perl Documentation Minion::Backend(3)

NAME

Minion::Backend - Backend base class

SYNOPSIS

  package Minion::Backend::MyBackend;
  use Mojo::Base 'Minion::Backend';
  sub broadcast          {...}
  sub dequeue            {...}
  sub dispatch_schedules {...}
  sub enqueue            {...}
  sub fail_job           {...}
  sub finish_job         {...}
  sub history            {...}
  sub list_jobs          {...}
  sub list_locks         {...}
  sub list_schedules     {...}
  sub list_workers       {...}
  sub lock               {...}
  sub note               {...}
  sub pause_schedule     {...}
  sub receive            {...}
  sub register_worker    {...}
  sub remove_job         {...}
  sub repair             {...}
  sub reset              {...}
  sub resume_schedule    {...}
  sub retry_job          {...}
  sub schedule           {...}
  sub stats              {...}
  sub unlock             {...}
  sub unregister_worker  {...}
  sub unschedule         {...}

DESCRIPTION

Minion::Backend is an abstract base class for Minion backends, like Minion::Backend::Pg.

ATTRIBUTES

Minion::Backend implements the following attributes.

minion

  my $minion = $backend->minion;
  $backend   = $backend->minion(Minion->new);

Minion object this backend belongs to. Note that this attribute is weakened.

METHODS

Minion::Backend inherits all methods from Mojo::Base and implements the following new ones.

auto_retry_job

  my $bool = $backend->auto_retry_job($job_id, $retries, $attempts);

Automatically "retry" job with "backoff" in Minion if there are attempts left, used to implement backends like Minion::Backend::Pg.

broadcast

  my $bool = $backend->broadcast('some_command');
  my $bool = $backend->broadcast('some_command', [@args]);
  my $bool = $backend->broadcast('some_command', [@args], [$id1, $id2, $id3]);

Broadcast remote control command to one or more workers. Meant to be overloaded in a subclass.

dequeue

  my $job_info = $backend->dequeue($worker_id, 0.5);
  my $job_info = $backend->dequeue($worker_id, 0.5, {queues => ['important']});

Wait a given amount of time in seconds for a job, dequeue it and transition from "inactive" to "active" state, or return "undef" if queues were empty. Meant to be overloaded in a subclass.

These options are currently available:

  id => '10023'
    

Dequeue a specific job.

  min_priority => 3
    

Do not dequeue jobs with a lower priority.

  queues => ['important']
    

One or more queues to dequeue jobs from, defaults to "default".

  tasks => ['foo', 'bar']
    

One or more tasks to dequeue jobs for, defaults to all.

These fields are currently available:

  args => ['foo', 'bar']
    

Job arguments.

  id => '10023'
    

Job ID.

  retries => 3
    

Number of times job has been retried.

  task => 'foo'
    

Task name.

dispatch_schedules

  my $dispatched = $backend->dispatch_schedules;

Enqueue jobs for all schedules whose firing time has been reached, advance their firing times to the next match, and return information about each dispatch as an array reference. Coordinates across multiple workers so a single dispatch cycle never enqueues a schedule twice. Meant to be overloaded in a subclass.

Each dispatched entry contains these fields:

  id => 23
    

Schedule id.

  job => '10025'
    

Id of the job that was just enqueued.

  name => 'daily'
    

Schedule name.

enqueue

  my $job_id = $backend->enqueue('foo');
  my $job_id = $backend->enqueue(foo => [@args]);
  my $job_id = $backend->enqueue(foo => [@args] => {priority => 1});

Enqueue a new job with "inactive" state. Meant to be overloaded in a subclass.

These options are currently available:

  attempts => 25
    

Number of times performing this job will be attempted, with a delay based on "backoff" in Minion after the first attempt, defaults to 1.

  delay => 10
    

Delay job for this many seconds (from now), defaults to 0.

  expire => 300
    

Job is valid for this many seconds (from now) before it expires.

  lax => 1
    

Existing jobs this job depends on may also have transitioned to the "failed" state to allow for it to be processed, defaults to "false".

  notes => {foo => 'bar', baz => [1, 2, 3]}
    

Hash reference with arbitrary metadata for this job.

  parents => [$id1, $id2, $id3]
    

One or more existing jobs this job depends on, and that need to have transitioned to the state "finished" before it can be processed.

  priority => 5
    

Job priority, defaults to 0. Jobs with a higher priority get performed first. Priorities can be positive or negative, but should be in the range between 100 and -100.

  queue => 'important'
    

Queue to put job in, defaults to "default".

fail_job

  my $bool = $backend->fail_job($job_id, $retries);
  my $bool = $backend->fail_job($job_id, $retries, 'Something went wrong!');
  my $bool = $backend->fail_job(
    $job_id, $retries, {whatever => 'Something went wrong!'});

Transition from "active" to "failed" state with or without a result, and if there are attempts remaining, transition back to "inactive" with a delay based on "backoff" in Minion. Meant to be overloaded in a subclass.

finish_job

  my $bool = $backend->finish_job($job_id, $retries);
  my $bool = $backend->finish_job($job_id, $retries, 'All went well!');
  my $bool = $backend->finish_job(
    $job_id, $retries, {whatever => 'All went well!'});

Transition from "active" to "finished" state with or without a result. Meant to be overloaded in a subclass.

history

  my $history = $backend->history;

Get history information for job queue. Meant to be overloaded in a subclass.

These fields are currently available:

  daily => [{epoch => 12345, finished_jobs => 95, failed_jobs => 2}, ...]
    

Hourly counts for processed jobs from the past day.

list_jobs

  my $results = $backend->list_jobs($offset, $limit);
  my $results = $backend->list_jobs($offset, $limit, {states => ['inactive']});

Returns the information about jobs in batches. Meant to be overloaded in a subclass.

  # Get the total number of results (without limit)
  my $num = $backend->list_jobs(0, 100, {queues => ['important']})->{total};
  # Check job state
  my $results = $backend->list_jobs(0, 1, {ids => [$job_id]});
  my $state = $results->{jobs}[0]{state};
  # Get job result
  my $results = $backend->list_jobs(0, 1, {ids => [$job_id]});
  my $result  = $results->{jobs}[0]{result};

These options are currently available:

  before => 23
    

List only jobs before this id.

  ids => ['23', '24']
    

List only jobs with these ids.

  notes => ['foo', 'bar']
    

List only jobs with one of these notes.

  queues => ['important', 'unimportant']
    

List only jobs in these queues.

  states => ['inactive', 'active']
    

List only jobs in these states.

  tasks => ['foo', 'bar']
    

List only jobs for these tasks.

These fields are currently available:

  args => ['foo', 'bar']
    

Job arguments.

  attempts => 25
    

Number of times performing this job will be attempted.

  children => ['10026', '10027', '10028']
    

Jobs depending on this job.

  created => 784111777
    

Epoch time job was created.

  delayed => 784111777
    

Epoch time job was delayed to.

  expires => 784111777
    

Epoch time job is valid until before it expires.

  finished => 784111777
    

Epoch time job was finished.

  id => 10025
    

Job id.

  lax => 0
    

Existing jobs this job depends on may also have failed to allow for it to be processed.

  notes => {foo => 'bar', baz => [1, 2, 3]}
    

Hash reference with arbitrary metadata for this job.

  parents => ['10023', '10024', '10025']
    

Jobs this job depends on.

  priority => 3
    

Job priority.

  queue => 'important'
    

Queue name.

  result => 'All went well!'
    

Job result.

  retried => 784111777
    

Epoch time job has been retried.

  retries => 3
    

Number of times job has been retried.

  started => 784111777
    

Epoch time job was started.

  state => 'inactive'
    

Current job state, usually "active", "failed", "finished" or "inactive".

  task => 'foo'
    

Task name.

  time => 78411177
    

Server time.

  worker => '154'
    

Id of worker that is processing the job.

list_locks

  my $results = $backend->list_locks($offset, $limit);
  my $results = $backend->list_locks($offset, $limit, {names => ['foo']});

Returns information about locks in batches. Meant to be overloaded in a subclass.

  # Get the total number of results (without limit)
  my $num = $backend->list_locks(0, 100, {names => ['bar']})->{total};
  # Check expiration time
  my $results = $backend->list_locks(0, 1, {names => ['foo']});
  my $expires = $results->{locks}[0]{expires};

These options are currently available:

  names => ['foo', 'bar']
    

List only locks with these names.

These fields are currently available:

  expires => 784111777
    

Epoch time this lock will expire.

  id => 1
    

Lock id.

  name => 'foo'
    

Lock name.

list_schedules

  my $results = $backend->list_schedules($offset, $limit);
  my $results = $backend->list_schedules($offset, $limit, {names => ['daily']});

Returns information about schedules in batches. Meant to be overloaded in a subclass.

  # Get the total number of results (without limit)
  my $num = $backend->list_schedules(0, 100)->{total};
  # Check next firing time
  my $results = $backend->list_schedules(0, 1, {names => ['daily']});
  my $next    = $results->{schedules}[0]{next_run};

These options are currently available:

  before => 23
    

List only schedules before this id.

  ids => ['23', '24']
    

List only schedules with these ids.

  names => ['foo', 'bar']
    

List only schedules with these names.

These fields are currently available:

  args => ['foo', 'bar']
    

Job arguments used for each enqueued job.

  attempts => 25
    

Number of attempts each enqueued job will get.

  created => 784111777
    

Epoch time the schedule was created.

  cron => '0 9 * * 1-5'
    

Cron expression.

  expire => 300
    

Expiration in seconds for each enqueued job.

  id => 23
    

Schedule id.

  last_job => '10025'
    

Id of the most recently enqueued job, or "undef" if the schedule has not fired yet.

  last_run => 784111777
    

Epoch time the schedule last fired, or "undef" if it has not fired yet.

  lax => 0
    

Lax dependency setting for each enqueued job.

  name => 'daily'
    

Schedule name.

  next_run => 784111777
    

Epoch time the schedule will fire next.

  notes => {foo => 'bar'}
    

Hash reference with arbitrary metadata applied to each enqueued job.

  paused => 0
    

True if the schedule is paused and will not fire.

  priority => 0
    

Priority of each enqueued job.

  queue => 'default'
    

Queue each enqueued job is placed in.

  task => 'foo'
    

Task name.

list_workers

  my $results = $backend->list_workers($offset, $limit);
  my $results = $backend->list_workers($offset, $limit, {ids => [23]});

Returns information about workers in batches. Meant to be overloaded in a subclass.

  # Get the total number of results (without limit)
  my $num = $backend->list_workers(0, 100)->{total};
  # Check worker host
  my $results = $backend->list_workers(0, 1, {ids => [$worker_id]});
  my $host    = $results->{workers}[0]{host};

These options are currently available:

  before => 23
    

List only workers before this id.

  ids => ['23', '24']
    

List only workers with these ids.

These fields are currently available:

  id => 22
    

Worker id.

  host => 'localhost'
    

Worker host.

  jobs => ['10023', '10024', '10025', '10029']
    

Ids of jobs the worker is currently processing.

  notified => 784111777
    

Epoch time worker sent the last heartbeat.

  pid => 12345
    

Process id of worker.

  started => 784111777
    

Epoch time worker was started.

  status => {queues => ['default', 'important']}
    

Hash reference with whatever status information the worker would like to share.

lock

  my $bool = $backend->lock('foo', 3600);
  my $bool = $backend->lock('foo', 3600, {limit => 20});

Try to acquire a named lock that will expire automatically after the given amount of time in seconds. An expiration time of 0 can be used to check if a named lock already exists without creating one. Meant to be overloaded in a subclass.

These options are currently available:

  limit => 20
    

Number of shared locks with the same name that can be active at the same time, defaults to 1.

note

  my $bool = $backend->note($job_id, {mojo => 'rocks', minion => 'too'});

Change one or more metadata fields for a job. Setting a value to "undef" will remove the field. Meant to be overloaded in a subclass.

pause_schedule

  my $bool = $backend->pause_schedule('daily');

Pause a schedule by name so it stops firing until resumed. Returns true on success, false if the schedule does not exist. Meant to be overloaded in a subclass.

receive

  my $commands = $backend->receive($worker_id);

Receive remote control commands for worker. Meant to be overloaded in a subclass.

register_worker

  my $worker_id = $backend->register_worker;
  my $worker_id = $backend->register_worker($worker_id);
  my $worker_id = $backend->register_worker(
    $worker_id, {status => {queues => ['default', 'important']}});

Register worker or send heartbeat to show that this worker is still alive. Meant to be overloaded in a subclass.

These options are currently available:

  status => {queues => ['default', 'important']}
    

Hash reference with whatever status information the worker would like to share.

remove_job

  my $bool = $backend->remove_job($job_id);

Remove "failed", "finished" or "inactive" job from queue. Meant to be overloaded in a subclass.

repair

  $backend->repair;

Repair worker registry and job queue if necessary. Meant to be overloaded in a subclass.

reset

  $backend->reset({all => 1});

Reset job queue. Meant to be overloaded in a subclass.

These options are currently available:

  all => 1
    

Reset everything.

  locks => 1
    

Reset only locks.

resume_schedule

  my $bool = $backend->resume_schedule('daily');

Resume a previously paused schedule. Returns true on success, false if the schedule does not exist. Meant to be overloaded in a subclass.

retry_job

  my $bool = $backend->retry_job($job_id, $retries);
  my $bool = $backend->retry_job($job_id, $retries, {delay => 10});

Transition job back to "inactive" state, already "inactive" jobs may also be retried to change options. Meant to be overloaded in a subclass.

These options are currently available:

  attempts => 25
    

Number of times performing this job will be attempted.

  delay => 10
    

Delay job for this many seconds (from now), defaults to 0.

  expire => 300
    

Job is valid for this many seconds (from now) before it expires.

  lax => 1
    

Existing jobs this job depends on may also have transitioned to the "failed" state to allow for it to be processed, defaults to "false".

  parents => [$id1, $id2, $id3]
    

Jobs this job depends on.

  priority => 5
    

Job priority.

  queue => 'important'
    

Queue to put job in.

schedule

  my $id = $backend->schedule('daily', '0 4 * * *', 'cleanup');
  my $id = $backend->schedule('daily', '0 4 * * *', 'cleanup', [@args]);
  my $id = $backend->schedule(
    'daily', '0 4 * * *', 'cleanup', [@args], {priority => 5});

Create or replace a schedule by unique name. The cron expression is parsed up front and the next firing time is computed; the entry is rejected if the expression is invalid. Existing schedules with the same name are updated, but the firing time is preserved if the cron expression has not changed. Meant to be overloaded in a subclass.

These options are currently available:

  attempts => 25
    

Number of times performing each enqueued job will be attempted, with a delay based on "backoff" in Minion after the first attempt, defaults to 1.

  expire => 300
    

Each enqueued job is valid for this many seconds (from enqueue time) before it expires.

  lax => 1
    

Existing jobs each enqueued job depends on may also have transitioned to the "failed" state to allow for it to be processed, defaults to "false".

  notes => {foo => 'bar', baz => [1, 2, 3]}
    

Hash reference with arbitrary metadata for each enqueued job.

  priority => 5
    

Priority of each enqueued job, defaults to 0.

  queue => 'important'
    

Queue to put each enqueued job in, defaults to "default".

stats

  my $stats = $backend->stats;

Get statistics for the job queue. Meant to be overloaded in a subclass.

These fields are currently available:

  active_jobs => 100
    

Number of jobs in "active" state.

  active_locks => 100
    

Number of active named locks.

  active_workers => 100
    

Number of workers that are currently processing a job.

  delayed_jobs => 100
    

Number of jobs in "inactive" state that are scheduled to run at specific time in the future or have unresolved dependencies.

  enqueued_jobs => 100000
    

Rough estimate of how many jobs have ever been enqueued.

  failed_jobs => 100
    

Number of jobs in "failed" state.

  finished_jobs => 100
    

Number of jobs in "finished" state.

  inactive_jobs => 100
    

Number of jobs in "inactive" state.

  inactive_schedules => 100
    

Number of schedules that are currently paused.

  inactive_workers => 100
    

Number of workers that are currently not processing a job.

  schedules => 100
    

Number of schedules that are currently active.

  uptime => 1000
    

Uptime in seconds.

  workers => 200;
    

Number of registered workers.

unlock

  my $bool = $backend->unlock('foo');

Release a named lock. Meant to be overloaded in a subclass.

unregister_worker

  $backend->unregister_worker($worker_id);

Unregister worker. Meant to be overloaded in a subclass.

unschedule

  my $bool = $backend->unschedule('daily');

Remove a schedule by name. Returns true on success, false if the schedule does not exist. Meant to be overloaded in a subclass.

SEE ALSO

Minion, Minion::Guide, <https://minion.pm>, Mojolicious::Guides, <https://mojolicious.org>.

2026-05-22 perl v5.42.1