Checklist

• I have verified that the issue exists against the main branch of Celery.
• This has already been asked to the discussions forum first.
• I have read the relevant section in the
  contribution guide
  on reporting bugs.
• I have checked the issues list
  for similar or identical bug reports.
• I have checked the pull requests list
  for existing proposed fixes.
• I have checked the commit log
  to find out if the bug was already fixed in the main branch.
• I have included all related issues and possible duplicate issues
  in this issue (If there are none, check this box anyway).
• I have tried to reproduce the issue with pytest-celery and added the reproduction script below.

Mandatory Debugging Information

• I have included the output of celery -A proj report in the issue.
  (if you are not able to do this, then at least specify the Celery
  version affected).
• I have verified that the issue exists against the main branch of Celery.
• I have included the contents of pip freeze in the issue.
• I have included all the versions of all the external dependencies required
  to reproduce this bug.

Optional Debugging Information

• I have tried reproducing the issue on more than one Python version
  and/or implementation.
• I have tried reproducing the issue on more than one message broker and/or
  result backend.
• I have tried reproducing the issue on more than one version of the message
  broker and/or result backend.
• I have tried reproducing the issue on more than one operating system.
• I have tried reproducing the issue on more than one workers pool.
• I have tried reproducing the issue with autoscaling, retries,
  ETA/Countdown & rate limits disabled.
• I have tried reproducing the issue after downgrading
  and/or upgrading Celery and its dependencies.

Related Issues and Possible Duplicates

Related Issues

I couldn't find any

Possible Duplicates

Couldn't find any

Environment & Settings

❯ uv run celery --version
5.6.1 (recovery)

❯ uv run celery -A app report

software -> celery:5.6.1 (recovery) kombu:5.6.2 py:3.14.0rc2
            billiard:4.2.4 py-amqp:5.3.1
platform -> system:Darwin arch:64bit, Mach-O
            kernel version:25.2.0 imp:CPython
loader   -> celery.loaders.app.AppLoader
settings -> transport:amqp results:disabled

broker_url: 'amqp://guest:********@localhost:5672//'
task_ignore_result: True
deprecated_settings: None

Steps to Reproduce

Required Dependencies

I tested on some old versions just to see if this was related to a recent upgrade, I did not test all intermediate versions, or older than 4.4.7, or individual dependency versions

• Minimal Python Version: 3.10 + 3.14
• Minimal Celery Version: celery:4.4.7 (cliffs) + 5.6.1
• Minimal Kombu Version: 4.6.11
• Minimal Broker Version: N/A
• • Minimal Result Backend Version: N/A
• Minimal OS and/or Kernel Version: Mac OS 26.2
• Minimal Broker Client Version: py-amqp:2.6.1
• Minimal Result Backend Client Version: N/A

Python Packages

altgraph @ file:///AppleInternal/Library/BuildRoots/4~CDlvugAeMYaHcTWIELyAQ2joHAecmlI0GAlKPsw/Library/Caches/com.apple.xbs/Sources/python3/altgraph-0.17.2-py2.py3-none-any.whl
backports-datetime-fromisoformat==2.0.3
cachetools==6.0.0
certifi==2025.4.26
cfgv==3.4.0
chardet==5.2.0
charset-normalizer==3.4.2
colorama==0.4.6
distlib==0.3.7
docutils==0.21.2
exceptiongroup==1.3.0
filelock==3.13.1
flit==3.12.0
flit_core==3.12.0
future @ file:///AppleInternal/Library/BuildRoots/4~CDlvugAeMYaHcTWIELyAQ2joHAecmlI0GAlKPsw/Library/Caches/com.apple.xbs/Sources/python3/future-0.18.2-py3-none-any.whl
identify==2.5.31
idna==3.10
iniconfig==2.1.0
macholib @ file:///AppleInternal/Library/BuildRoots/4~CDlvugAeMYaHcTWIELyAQ2joHAecmlI0GAlKPsw/Library/Caches/com.apple.xbs/Sources/python3/macholib-1.15.2-py2.py3-none-any.whl
# Editable install with no version control (marshmallow==4.0.0)
-e /Users/nate/wave/marshmallow
nodeenv==1.8.0
packaging==25.0
pbr==6.0.0
platformdirs==3.11.0
pluggy==1.6.0
pre-commit==3.5.0
pyproject-api==1.9.1
pytest==8.3.5
PyYAML==6.0.1
requests==2.32.3
simplejson==3.20.1
six @ file:///AppleInternal/Library/BuildRoots/4~CDlvugAeMYaHcTWIELyAQ2joHAecmlI0GAlKPsw/Library/Caches/com.apple.xbs/Sources/python3/six-1.15.0-py2.py3-none-any.whl
stevedore==5.1.0
tomli==2.2.1
tomli_w==1.2.0
tox==4.11.4
typing_extensions==4.13.2
urllib3==2.4.0
virtualenv==20.24.6
virtualenv-clone==0.5.7
virtualenvwrapper==4.8.4

Other Dependencies

Minimally Reproducible Test Case

I have pushed up an example repository here:

Here is a specific gist: https://gist.github.com/compyman/c5c74c7c59088568a39b20203b522be0

There is some other weird behavior that I am trying to explore in that repository, but this is the main issue right now.

Expected Behavior

I would expect that:

• that multiple tasks would be able to co-operatively yield resources
• that gevent monkey patching would effectively patch resources used by celery

Actual Behavior

using the gevent worker, tasks are executing sequentially - it yields 5 tasks into the the message broker, but does not yield the greenthread to let another task so the greenthreads run mostly sequentially

In the gist you can see that sometimes 2 greenthreads will co-operate.

I have been attempting to diagnose this as well and have noticed some things that may be related.

1. The producer queue app.producer_pool._resource is not being monkey patched - it is an instance of a queue.LifoQueue (this appears to be imported during monkey-patching while looking up class on the default celery app which is a little gnarly)
2. if you raise a gevent timeout in the loop - it's possible to get an raised unexpected: RuntimeError('Semaphore released too many times') error (I can raise another issue for this)
3. Some of this behavior seems related to the broker_pool_limit

• if the broker limit is 0, then we do always yield greenthreads (I assume because establishing the connection yields)
• if the broker limit is less than the number of looping tasks, then we will yield more often, I assume while waiting for an available producer?

Let me know if I should raise this issue on another repo, or split it into multiple issues.
Also please let me know if there's any other information I could help out with.

Thank you!