The Fix
pip install redis==7.1.1
Based on closed redis/redis-py issue #3203 · 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.
@@ -296,7 +296,14 @@ def set_parser(self, parser_class: Type[BaseParser]) -> None:
async def connect(self):
"""Connects to the Redis server if not already connected"""
- await self.connect_check_health(check_health=True)
+ # try once the socket connect with the handshake, retry the whole
+ # connect/handshake flow based on retry policy
def event_handler(msg):
print(msg)
redis_pub_sub.psubscribe(**{"test": event_handler})
def exception_handler(ex, pubsub, thread):
print(traceback.format_exc())
redis_pub_sub.run_in_thread(sleep_time=0.1, exception_handler=exception_handler)
# Close Redis connection
Re-run the minimal reproduction on your broken version, then apply the fix and re-run.
Option A — Upgrade to fixed release\npip install redis==7.1.1\nWhen NOT to use: This fix should not be used if the application cannot tolerate any connection errors.\n\nOption C — Workaround\nis to wrap everything inside a retry call, as shown below. However, this might introduce unexpected side effects.\nWhen NOT to use: This fix should not be used if the application cannot tolerate any connection errors.\n\n
Why This Fix Works in Production
- Trigger: return do()
- Mechanism: The retry functionality in the PubSub-client fails to handle OSError correctly during reconnection attempts
- Why the fix works: Adds retries for both the socket connection and the initial connection handshake, improving the retry behavior for Pub/Sub connection issues. (first fixed release: 7.1.1).
- If left unfixed, retry loops can amplify load and turn a small outage into a cascade (thundering herd).
Why This Breaks in Prod
- Shows up under Python 3.12.2 in real deployments (not just unit tests).
- The retry functionality in the PubSub-client fails to handle OSError correctly during reconnection attempts
- Surfaces as: Traceback (most recent call last):
Proof / Evidence
- GitHub issue: #3203
- Fix PR: https://github.com/redis/redis-py/pull/3863
- First fixed release: 7.1.1
- Reproduced locally: No (not executed)
- Last verified: 2026-02-08
- Confidence: 0.80
- Did this fix it?: Yes (upstream fix exists)
- Own content ratio: 0.39
Discussion
High-signal excerpts from the issue thread (symptoms, repros, edge-cases).
“Btw @petyaslavova , I have a better explanation of the problem here”
“Having the same issue with pipelines. Solved using this hack:”
“Hey, Based on your description, I believe I've encountered the same issue with the synchronous version (as you said)”
Failure Signature (Search String)
- return do()
Error Message
Stack trace
Error Message
-------------
Traceback (most recent call last):
File "/path/to/env/lib/python3.10/site-packages/redis/retry.py", line 62, in call_with_retry
return do()
File "/path/to/env/lib/python3.10/site-packages/redis/client.py", line 842, in <lambda>
lambda: command(*args, **kwargs),
File "/path/to/env/lib/python3.10/site-packages/redis/client.py", line 859, in try_read
if not conn.can_read(timeout=timeout):
File "/path/to/env/lib/python3.10/site-packages/redis/connection.py", line 600, in can_read
return self._parser.can_read(timeout)
File "/path/to/env/lib/python3.10/site-packages/redis/_parsers/hiredis.py", line 79, in can_read
return self.read_from_socket(timeout=timeout, raise_on_timeout=False)
File "/path/to/env/lib/python3.10/site-packages/redis/_parsers/hiredis.py", line 90, in read_from_socket
raise ConnectionError(SERVER_CLOSED_CONNECTION_ERROR)
redis.exceptions.ConnectionError: Connection closed by server.
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "/path/to/env/lib/python3.10/site-packages/redis/connection.py", line 622, in read_response
response = self._parser.read_response(disable_decoding=disable_decoding)
File "/path/to/env/lib/python3.10/site-packages/redis/_parsers/hiredis.py", line 128, in read_response
self.read_from_socket()
File "/path/to/env/lib/python3.10/site-p
... (truncated) ...
Minimal Reproduction
def event_handler(msg):
print(msg)
redis_pub_sub.psubscribe(**{"test": event_handler})
def exception_handler(ex, pubsub, thread):
print(traceback.format_exc())
redis_pub_sub.run_in_thread(sleep_time=0.1, exception_handler=exception_handler)
# Close Redis connection
Environment
- Python: 3.12.2
What Broke
PubSub-client crashes and does not recover when Redis is restarted, causing outages.
Why It Broke
The retry functionality in the PubSub-client fails to handle OSError correctly during reconnection attempts
Fix Options (Details)
Option A — Upgrade to fixed release Safe default (recommended)
pip install redis==7.1.1
Use when you can deploy the upstream fix. It is usually lower-risk than long-lived workarounds.
Option C — Workaround Temporary workaround
is to wrap everything inside a retry call, as shown below. However, this might introduce unexpected side effects.
Use only if you cannot change versions today. Treat this as a stopgap and remove once upgraded.
Fix reference: https://github.com/redis/redis-py/pull/3863
First fixed release: 7.1.1
Last verified: 2026-02-08. Validate in your environment.
When NOT to Use This Fix
- This fix should not be used if the application cannot tolerate any connection errors.
Verify Fix
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
- Make timeouts explicit and test them (unit + integration) to avoid silent behavior changes.
- Instrument retries (attempt count + reason) and alert on spikes to catch dependency slowdowns.
Version Compatibility Table
| Version | Status |
|---|---|
| 7.1.1 | Fixed |
Related Issues
No related fixes found.
Sources
We don’t republish the full GitHub discussion text. Use the links above for context.