Skip to main content

Nameko extension allowing AMQP entrypoints to retry later

Project description

nameko-amqp-retry
=================

Extension for `nameko <http://nameko.readthedocs.org>`_ that allows the built-in AMQP entrypoints to schedule a later retry via redelivery of a message.

RabbitMQ 3.5.4 or later is required.


Installation
------------

Install the library from PyPI::

pip install nameko-amqp-retry


Usage
-----

This library subclasses nameko's built-in entrypoints. Use these subclasses in your service definition, and then raise :class:`nameko_amqp_retry.Backoff` inside an entrypoint you wish to retry later:

.. code-block:: python

from nameko_amqp_retry import Backoff
from nameko_amqp_retry.rpc import rpc

class Service:
name = "service"

@rpc
def calculate(self):
""" Calculate something, or schedule a retry if not ready yet.
"""
if not_ready_yet:
raise Backoff()

return 42

The caller will see the final result, or a :class:`Backoff.Expired` exception if more than the allowed number of retries were made:

.. code-block:: python

>>> n.rpc.service.calculate()
... # blocks for some time
>>> 42

.. code-block:: python

>>> n.rpc.service.calculate()
Traceback (most recent call last):
...
nameko.exceptions.RemoteError: Expired Backoff aborted after ...

The retry schedule is controlled by attributes on the `Backoff` class. You should override them in a subclass to control the behaviour. For example:

Fixed schedule:

.. code-block:: python

class RegularBackoff(Backoff):
""" Retries every 1000ms until limit
"""
schedule = (1000,) # ms

No limit:

.. code-block:: python

class InfiniteBackoff(Backoff):
""" Retries forever
"""
limit = 0


Custom schedule:

.. code-block:: python

class ImpatientBackoff(Backoff):
""" Retries after 100, then 200, then 500 milliseconds
"""
schedule = (100, 200, 500) # ms


Dynamic schedule:

.. code-block:: python

class DynamicBackoff(Backoff):
""" Calculates schedule dynamically
"""
@classmethod
def get_next_schedule_item(cls, index):
...

Alternatively, an entrypoint can be decorated with the `entrypoint_retry` decorator, to automatically retry the method if it raises certain exceptions:

.. code-block:: python

from nameko_amqp_retry import entrypoint_retry
from nameko_amqp_retry.rpc import rpc

class Service:
name = "service"

@rpc
@entrypoint_retry(retry_for=ValueError)
def calculate(self):
""" Calculate something, or schedule a retry if not ready yet.
"""
if not_ready_yet:
raise ValueError()

return 42

@rpc
@entrypoint_retry(
retry_for=(TypeError, ValueError),
limit=5,
schedule=(500, 600, 700, 800, 900),
)
def do_something(self):
""" Calculate something else, or schedule a retry if not ready yet.
"""
if type_not_ready_yet:
raise TypeError()

if value_not_ready_yet:
raise ValueError()

return 24


See docs/examples for more.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

nameko-amqp-retry-0.3.0.tar.gz (5.2 kB view details)

Uploaded Source

Built Distribution

nameko_amqp_retry-0.3.0-py2-none-any.whl (7.5 kB view details)

Uploaded Python 2

File details

Details for the file nameko-amqp-retry-0.3.0.tar.gz.

File metadata

File hashes

Hashes for nameko-amqp-retry-0.3.0.tar.gz
Algorithm Hash digest
SHA256 466c038899c3096752cd5d62050ad9968b128259461d59075f39d2ac3084f302
MD5 5ceb4c72ccabd2f68dd06df32516ff32
BLAKE2b-256 8a243a135e2ce49d9fdb5924d8711e578a8c464c594aa4191f8b283d2d358e41

See more details on using hashes here.

File details

Details for the file nameko_amqp_retry-0.3.0-py2-none-any.whl.

File metadata

File hashes

Hashes for nameko_amqp_retry-0.3.0-py2-none-any.whl
Algorithm Hash digest
SHA256 396595527c29592aa3b25e02f590d86c32f5bc6a91583b94ec1d0043e727df28
MD5 dcfe570ed52eb3a791de5eadae8a2150
BLAKE2b-256 5e33f9696089813d15fdc1a0a21ab03d81947814f30a44673108b7a5f81f2c8b

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page