This document is for Celery's development version, which can be significantly different from previous releases. Get old docs here: 2.1.
Queue name/options mapping.
Name of the default queue.
Default exchange.
Default exchange type.
Default delivery mode (“persistent” or “non-persistent”). Default is “persistent”.
Default routing key used when sending tasks.
The timeout in seconds before we give up establishing a connection to the AMQP server.
Name prefix for the queue used when listening for broadcast messages. The workers hostname will be appended to the prefix to create the final queue name.
Default is “celeryctl”.
Name of the exchange used for broadcast messages.
Default is “celeryctl”.
Exchange type used for broadcast messages. Default is “fanout”.
Name of queue used to listen for event messages. Default is “celeryevent”.
Exchange used to send event messages. Default is “celeryevent”.
Exchange type used for the event exchange. Default is “topic”.
Routing key used for events. Default is “celeryevent”.
Type of serialization method used to serialize events. Default is “json”.
Exchange used by the AMQP result backend to publish task results. Default is “celeryresult”.
If set to True, errors in tasks will be sent to ADMINS by e-mail.
List of (name, email_address) tuples for the admins that should receive error e-mails.
The e-mail address this worker sends e-mails from. Default is “celery@localhost”.
The mail server to use. Default is “localhost”.
Username (if required) to log on to the mail server with.
Password (if required) to log on to the mail server with.
The port the mail server is listening on. Default is 25.
Always execute tasks locally, don’t send to the queue.
If set to True, celery.execute.apply() will re-raise task exceptions. It’s the same as always running apply with throw=True.
Task tombstone expire time in seconds.
If enabled, the default behavior will be to not store task results.
If enabled, the default behavior will be to track when tasks starts by storing the STARTED state.
If enabled, the default behavior will be to acknowledge task messages after the task is executed.
If enabled, task errors will be stored even though Task.ignore_result is enabled.
Total number of results to store before results are evicted from the result cache.
A string identifying the default serialization method to use.
Can be pickle (default), json, yaml, msgpack or any custom serialization methods that have been registered with kombu.serialization.registry.
The backend used to store task results (tombstones).
Celery cache backend.
If set, celery will send events that can be captured by monitors like celerymon. Default is: False.
The default rate limit applied to all tasks which doesn’t have a custom rate limit defined. (Default: None)
If True all rate limits will be disabled and all tasks will be executed as soon as possible.
Automatically try to re-establish the connection to the AMQP broker if it’s lost.
Maximum number of retries before we give up re-establishing a connection to the broker.
If this is set to 0 or None, we will retry forever.
Default is 100 retries.
Default log level for celerybeat. Default is: INFO.
Default log file for celerybeat. Default is: None (stderr)
Name of the persistent schedule database file. Default is: celerybeat-schedule.
The maximum number of seconds celerybeat is allowed to sleep between checking the schedule. The default is 5 minutes, which means celerybeat can only sleep a maximum of 5 minutes after checking the schedule run-times for a periodic task to apply. If you change the run_times of periodic tasks at run-time, you may consider lowering this value for changes to take effect faster (A value of 5 minutes, means the changes will take effect in 5 minutes at maximum).
Default log level for celerymon. Default is: INFO.
Default log file for celerymon. Default is: None (stderr)
Mapping of log level names to logging module constants.
The format to use for log messages.
The format to use for task log messages.
Filename of the daemon log file. Default is: None (stderr)
Default log level for daemons. (WARN)
The number of concurrent worker processes. If set to 0 (the default), the total number of available CPUs/cores will be used.
The number of concurrent workers is multipled by this number to yield the wanted AMQP QoS message prefetch count. Default is: 4
Name of the task pool class used by the worker. Default is “celery.concurrency.processes.TaskPool”.
Name of the consumer class used by the worker. Default is “celery.worker.consumer.Consumer”.
Name of the mediator class used by the worker. Default is “celery.worker.controllers.Mediator”.
Name of the ETA scheduler class used by the worker. Default is “celery.worker.controllers.ScheduleController”.