Change8

Celery

Backend & Infra

Distributed Task Queue (development branch)

Latest: v5.6.212 releases5 breaking changes6 common errorsView on GitHub

Release History

v5.6.22 fixes
Jan 4, 2026

Celery 5.6.2 resolves a recursive WorkController issue in DjangoWorkerFixup and ensures revoked tasks instantly reflect a REVOKED status in the backend.

v5.6.15 fixes2 features
Dec 29, 2025

Celery 5.6.1 includes several bug fixes, improved prefork handling, and documentation updates.

v5.6.0Breaking4 fixes3 features
Nov 30, 2025

Celery 5.6.0 drops Python 3.8 support, reverts SQS to pycurl, adds ETA task limits and queue‑type configuration, and includes several security and memory‑leak fixes.

v5.6.0rc22 fixes1 feature
Nov 22, 2025

Celery 5.6.0rc2 adds broker URL sanitization, improves shutdown handling, and fixes a return-in-finally issue in asynpool, with no breaking changes.

v5.6.0rc15 fixes3 features
Nov 1, 2025

Celery 5.6.0rc1 adds Django connection‑pool support, updates pymongo and kombu versions, and includes several bug fixes and dependency pinning.

v5.6.0b2Breaking6 fixes6 features
Oct 20, 2025

Celery 5.6.0b2 introduces Redis credential provider support, improved timezone handling, and several bug fixes while raising the minimum Python requirement to 3.9 and removing Python 3.8 support.

v5.6.0b110 fixes8 features
Sep 15, 2025

Celery 5.6.0b1 introduces several new features such as Docker layer caching and a worker_eta_task_limit configuration, along with numerous bug fixes and documentation updates.

v5.5.32 fixes3 features
Jun 1, 2025

Celery 5.5.3 adds support for apply_async without a queue on quorum queues, improves wheel metadata reproducibility, and fixes broker transport option handling and retry callbacks in DelayedDelivery.

v5.5.2Breaking2 fixes
Apr 25, 2025

Celery 5.5.2 includes bug fixes for DST time calculations and documentation links, and removes the deprecated `setup_logger` from COMPAT_MODULES, which may require import updates.

v5.5.11 fix
Apr 7, 2025

Celery v5.5.1 includes a bug fix for an AttributeError occurring with quorum queues and failover brokers.

v5.5.0Breaking3 fixes10 features
Mar 31, 2025

Celery 5.5.0 introduces Redis broker stability fixes, replaces pycurl with urllib3, adds RabbitMQ quorum queue and soft‑shutdown support, native Pydantic task integration, Google Pub/Sub transport, Python 3.13 compatibility, and several configuration enhancements.

v5.5.0rc5Breaking10 fixes8 features
Feb 25, 2025

Celery 5.5.0rc5 adds full Quorum Queue support, a new soft shutdown mechanism, Google Pub/Sub transport, and Pydantic task model support, while bumping the minimum Kombu version to 5.5.0 and fixing several bugs.

Common Errors

ModuleNotFoundError2 reports

The "ModuleNotFoundError" in Celery often arises from missing dependencies or incorrect installation paths. Ensure all necessary packages, especially kombu and any broker-specific libraries (like `py-mssql` for MSSQL), are installed using `pip install -r requirements.txt` or `pip install package_name`. Verify the Python environment Celery is using has these packages available in its `sys.path`.

ParserSyntaxError1 report

ParserSyntaxError in Celery usually arises from malformed `install_requires` entries within `setup.py` or `setup.cfg`, such as invalid version specifiers or non-PEP 440 compliant strings. Correct the problematic entries by ensuring each requirement adheres to valid packaging standards and version constraints (e.g., `pytz>=2020.1`, not `pytz>dev`). After fixing the setup file, rebuild the Celery distribution.

ConnectionResetError1 report

ConnectionResetError in Celery often arises from network instability or the Redis server closing idle connections. To fix this, configure Celery to use connection pooling with a shorter `broker_pool_limit` and implement retry mechanisms using `autoretry_for` in your Celery tasks, ensuring robust recovery from intermittent connection drops. Also, adjust Redis's `timeout` setting to a higher value.

AttributeError1 report

This AttributeError in Celery often arises when a string is incorrectly passed where a module or class is expected, particularly within Django settings or task definitions. Ensure that settings like `CELERY_TASK_SERIALIZER`, `CELERY_RESULT_SERIALIZER`, or task imports are referencing actual module or class objects, not string representations. Review your Celery configuration and task definitions, replacing string literals with direct references to the correct module or class.

NewConnectionError1 report

NewConnectionError in Celery usually arises from the broker (e.g., RabbitMQ, Redis) being unavailable or unreachable during Celery's attempts to establish a connection. To fix this, ensure your broker is running and accessible by Celery workers, and configure retry mechanisms with appropriate backoff strategies within your Celery settings to handle intermittent network issues or broker unavailability. Increasing the `broker_transport_options` retry parameters, such as `max_retries` and `interval_start`, can help.

FileNotFoundError1 report

FileNotFoundError in Celery often arises when Celery tries to access or write to a file path (like a result backend or log file) that doesn't exist or is inaccessible due to permissions. Ensure the directory specified in your Celery configuration (e.g., `CELERY_RESULT_BACKEND` if using a file-based backend or logging paths) exists and the Celery worker process has the necessary read/write permissions. Create the directory or adjust permissions as needed to resolve the issue.

Related Backend & Infra Packages

Subscribe to Updates

Get notified when new versions are released

RSS Feed