The Fix
pip install celery==5.5.0
Based on closed celery/celery issue #6067 · PR/commit linked
Production note: This usually shows up under retries/timeouts. Treat it as a side-effect risk until you can verify behavior with a canary + real traffic.
@@ -249,9 +249,13 @@ def Queues(self, queues, create_missing=None,
max_priority = conf.task_queue_max_priority
if not queues and conf.task_default_queue:
+ queue_arguments = None
+ if conf.task_default_queue_type == 'quorum':
+ queue_arguments = {'x-queue-type': 'quorum'}
repro.py
from celery import bootsteps
class NoChannelGlobalQoS(bootsteps.StartStopStep):
requires = {'celery.worker.consumer.tasks:Tasks'}
def start(self, c):
qos_global = False
c.connection.default_channel.basic_qos(
0, c.initial_prefetch_count, qos_global,
)
def set_prefetch_count(prefetch_count):
return c.task_consumer.qos(
prefetch_count=prefetch_count,
apply_global=qos_global,
)
c.qos = QoS(set_prefetch_count, c.initial_prefetch_count)
app.steps['consumer'].add(NoChannelGlobalQoS)
verify
Re-run the minimal reproduction on your broken version, then apply the fix and re-run.
fix.md
Option A — Upgrade to fixed release\npip install celery==5.5.0\nWhen NOT to use: This fix should not be used if backward compatibility with older queue types is required.\n\nOption C — Workaround\nbut I think the expected behaviour is that a celery user, after having setup rabbitmq using [commonly advised queue configurations](https://www.cloudamqp.com/blog/reasons-you-should-switch-to-quorum-queues.html) should be able to connect to the rabbit hosts and start submitting and consuming tasks. If i understand this will currently fail?\nWhen NOT to use: This fix should not be used if backward compatibility with older queue types is required.\n\n
Why This Fix Works in Production
- Trigger: amqp.exceptions.AMQPNotImplementedError: Basic.consume: (540) NOT_IMPLEMENTED - queue 'celery' in vhost '/' does not support global qos
- Mechanism: The library lacked support for configuring quorum queues in RabbitMQ
- Why the fix works: Added support for quorum queues in Celery, allowing users to configure queues with the 'quorum' type for better reliability. (first fixed release: 5.5.0).
Production impact:
- If left unfixed, retries/timeouts can trigger duplicate external side-effects (double charges, duplicate emails, repeated writes).
Why This Breaks in Prod
- Triggered by an upgrade/regression window: 01.405553–52.477317 breaks; 5.5.0 is the first fixed release.
- Shows up under Python 3.12.1 in real deployments (not just unit tests).
- The library lacked support for configuring quorum queues in RabbitMQ
- Surfaces as: amqp.exceptions.AMQPNotImplementedError: Basic.consume: (540) NOT_IMPLEMENTED - queue 'celery' in vhost '/' does not support global qos
Proof / Evidence
- GitHub issue: #6067
- Fix PR: https://github.com/celery/celery/pull/9121
- First fixed release: 5.5.0
- Affected versions: 01.405553–52.477317
- Reproduced locally: No (not executed)
- Last verified: 2026-02-09
- Confidence: 0.85
- Did this fix it?: Yes (upstream fix exists)
- Own content ratio: 0.42
Discussion
High-signal excerpts from the issue thread (symptoms, repros, edge-cases).
“Is there any update on this? Since RabitMQ itself will deprecate classic queues. https://blog.rabbitmq.com/posts/2021/08/4.0-deprecation-announcements/”
“Celery v5.5.0b1 was released, including a fix for this issue. Take note ETA tasks were not handled. See #9149 for possible further updates.”
“Yes, it fails for me. Until celery is updated to support per-consumer QOS it won't work with quorum queues.”
“> RabitMQ itself will deprecate classic queues”
Failure Signature (Search String)
- amqp.exceptions.AMQPNotImplementedError: Basic.consume: (540) NOT_IMPLEMENTED - queue 'celery' in vhost '/' does not support global qos
Error Message
Stack trace
error.txt
Error Message
-------------
amqp.exceptions.AMQPNotImplementedError: Basic.consume: (540) NOT_IMPLEMENTED - queue 'celery' in vhost '/' does not support global qos
Stack trace
error.txt
Error Message
-------------
[2024-09-16 10:33:52,524: CRITICAL/MainProcess] Unrecoverable error: AttributeError("'list' object has no attribute 'startswith'")
Traceback (most recent call last):
File "/var/opt/celery/celery_venv/lib/python3.10/site-packages/celery/worker/worker.py", line 203, in start
self.blueprint.start(self)
File "/var/opt/celery/celery_venv/lib/python3.10/site-packages/celery/bootsteps.py", line 116, in start
step.start(parent)
File "/var/opt/celery/celery_venv/lib/python3.10/site-packages/celery/bootsteps.py", line 365, in start
return self.obj.start()
File "/var/opt/celery/celery_venv/lib/python3.10/site-packages/celery/worker/consumer/consumer.py", line 340, in start
blueprint.start(self)
File "/var/opt/celery/celery_venv/lib/python3.10/site-packages/celery/bootsteps.py", line 116, in start
step.start(parent)
File "/var/opt/celery/celery_venv/lib/python3.10/site-packages/celery/worker/consumer/tasks.py", line 44, in start
qos_global = self.qos_global(c)
File "/var/opt/celery/celery_venv/lib/python3.10/site-packages/celery/worker/consumer/tasks.py", line 93, in qos_global
using_quorum_queues, qname = self.detect_quorum_queues(c)
File "/var/opt/celery/celery_venv/lib/python3.10/site-packages/celery/worker/consumer/tasks.py", line 110, in detect_quorum_queues
is_rabbitmq_broker = c.app.conf.broker_url.startswith(("amqp", "pyamqp"))
Attrib
... (truncated) ...
Minimal Reproduction
repro.py
from celery import bootsteps
class NoChannelGlobalQoS(bootsteps.StartStopStep):
requires = {'celery.worker.consumer.tasks:Tasks'}
def start(self, c):
qos_global = False
c.connection.default_channel.basic_qos(
0, c.initial_prefetch_count, qos_global,
)
def set_prefetch_count(prefetch_count):
return c.task_consumer.qos(
prefetch_count=prefetch_count,
apply_global=qos_global,
)
c.qos = QoS(set_prefetch_count, c.initial_prefetch_count)
app.steps['consumer'].add(NoChannelGlobalQoS)
Environment
- Python: 3.12.1
What Broke
Users experienced failures when attempting to use quorum queues with Celery.
Why It Broke
The library lacked support for configuring quorum queues in RabbitMQ
Fix Options (Details)
Option A — Upgrade to fixed release Safe default (recommended)
pip install celery==5.5.0
When NOT to use: This fix should not be used if backward compatibility with older queue types is required.
Use when you can deploy the upstream fix. It is usually lower-risk than long-lived workarounds.
Option C — Workaround Temporary workaround
but I think the expected behaviour is that a celery user, after having setup rabbitmq using [commonly advised queue configurations](https://www.cloudamqp.com/blog/reasons-you-should-switch-to-quorum-queues.html) should be able to connect to the rabbit hosts and start submitting and consuming tasks. If i understand this will currently fail?
When NOT to use: This fix should not be used if backward compatibility with older queue types is required.
Use only if you cannot change versions today. Treat this as a stopgap and remove once upgraded.
Option D — Guard side-effects with OnceOnly Guardrail for side-effects
Mitigate duplicate external side-effects under retries/timeouts/agent loops by gating the operation before calling external systems.
- Place OnceOnly between your code/agent and real side-effects (Stripe, emails, CRM, APIs).
- Use a stable key per side-effect (e.g., customer_id + action + idempotency_key).
- Fail-safe: configure fail-open vs fail-closed based on blast radius and spend risk.
- This is most useful when retries/timeouts can re-trigger the same external call.
Show example snippet (optional)
onceonly.py
from onceonly import OnceOnly
import os
once = OnceOnly(api_key=os.environ["ONCEONLY_API_KEY"], fail_open=True)
# Stable idempotency key per real side-effect.
# Use a request id / job id / webhook delivery id / Stripe event id, etc.
event_id = "evt_..." # replace
key = f"stripe:webhook:{event_id}"
res = once.check_lock(key=key, ttl=3600)
if res.duplicate:
return {"status": "already_processed"}
# Safe to execute the side-effect exactly once.
handle_event(event_id)
When NOT to use: Do not use this to hide logic bugs or data corruption. Use it to block duplicate external side-effects and enforce tool permissions/spend caps.
Fix reference: https://github.com/celery/celery/pull/9121
First fixed release: 5.5.0
Last verified: 2026-02-09. Validate in your environment.
When NOT to Use This Fix
- This fix should not be used if backward compatibility with older queue types is required.
- Do not use this to hide logic bugs or data corruption. Use it to block duplicate external side-effects and enforce tool permissions/spend caps.
Verify Fix
verify
Re-run the minimal reproduction on your broken version, then apply the fix and re-run.
Did This Fix Work in Your Case?
Quick signal helps us prioritize which fixes to verify and improve.
Prevention
- Capture the exact failing error string in logs and tests so you can reproduce via a minimal script.
- Pin production dependencies and upgrade only with a reproducible test that hits the failing path.
Version Compatibility Table
| Version | Status |
|---|---|
| 01.405553 | Broken |
| 21.576442 | Broken |
| 24.611430 | Broken |
| 26.505187 | Broken |
| 33.588267 | Broken |
| 52.477317 | Broken |
| 5.5.0 | Fixed |
Related Issues
No related fixes found.
Sources
We don’t republish the full GitHub discussion text. Use the links above for context.