redis-py Connections are not released to a connection pool,

The Fix

pip install redis==7.1.0

Based on closed redis/redis-py issue #2814 · PR/commit linked

Open PR/Commit

@@ -2143,6 +2143,8 @@ def _send_cluster_commands(
                         connection = get_connection(redis_node, c.args)
                     except ConnectionError:
+                        for n in nodes.values():
+                            n.connection_pool.release(n.connection)
                         # Connection retries are being handled in the node's

fix.md

Option A — Upgrade to fixed release\npip install redis==7.1.0\nWhen NOT to use: Do not apply if the connection pool behavior is expected to change in future versions.\n\n

Why This Fix Works in Production

Trigger: Connections are not released to a connection pool, when ClusterPipeline faces an exception on get_connection
Mechanism: ClusterPipeline does not release connections to the pool after a ConnectionError occurs
Why the fix works: Releases already acquired connections on ClusterPipeline when get_connection raises an exception, addressing a connection leak issue. (first fixed release: 7.1.0).

Production impact:

If left unfixed, this can cause silent data inconsistencies that propagate (bad cache entries, incorrect downstream decisions).

Why This Breaks in Prod

Shows up under Python 3.5.1 in real deployments (not just unit tests).
ClusterPipeline does not release connections to the pool after a ConnectionError occurs
Production symptom (often without a traceback): Connections are not released to a connection pool, when ClusterPipeline faces an exception on get_connection

Proof / Evidence

GitHub issue: #2814
Fix PR: https://github.com/redis/redis-py/pull/3133
First fixed release: 7.1.0
Reproduced locally: No (not executed)
Last verified: 2026-02-08
Confidence: 0.85
Did this fix it?: Yes (upstream fix exists)
Own content ratio: 0.67

Discussion

High-signal excerpts from the issue thread (symptoms, repros, edge-cases).

Jump to Sources Open on GitHub

“Hi @zakaf! First off, thank you for the report, and a potential bug fix. @dvora-h mind checking if this makes sense in the 4.6, and…”

@chayim · 2023-06-28 · source

“@zakaf this is still relevant and the fix makes sense... do you want to open a PR with the fix or wants me to take…”

@dvora-h · 2024-01-21 · source

“@dvora-h sorry for the delayed response. i've created #3133 to fix this issue”

@zakaf · 2024-02-04 · source

Failure Signature (Search String)

Connections are not released to a connection pool, when ClusterPipeline faces an exception on get_connection
**Description**: Description of your issue, stack traces from errors and code that reproduces the issue

Copy-friendly signature

signature.txt

Failure Signature
-----------------
Connections are not released to a connection pool, when ClusterPipeline faces an exception on get_connection
**Description**: Description of your issue, stack traces from errors and code that reproduces the issue

Error Message

Signature-only (no traceback captured)

error.txt

Error Message
-------------
Connections are not released to a connection pool, when ClusterPipeline faces an exception on get_connection
**Description**: Description of your issue, stack traces from errors and code that reproduces the issue

Environment

Python: 3.5.1

What Broke

Connections leak, causing resource exhaustion and potential service outages under high load.

Why It Broke

ClusterPipeline does not release connections to the pool after a ConnectionError occurs

Fix Options (Details)

Option A — Upgrade to fixed release Safe default (recommended)

pip install redis==7.1.0

When NOT to use: Do not apply if the connection pool behavior is expected to change in future versions.

Use when you can deploy the upstream fix. It is usually lower-risk than long-lived workarounds.

Fix reference: https://github.com/redis/redis-py/pull/3133

First fixed release: 7.1.0

Last verified: 2026-02-08. Validate in your environment.

When NOT to Use This Fix

Do not apply if the connection pool behavior is expected to change in future versions.

Did This Fix Work in Your Case?

Quick signal helps us prioritize which fixes to verify and improve.

Prevention

Track RSS + object counts after deployments; alert on monotonic growth and GC pressure.
Add a long-running test that repeats the failing call path and asserts stable memory.

Version Compatibility Table

Version	Status
7.1.0	Fixed

Related Issues

No related fixes found.

Cluster: redis-py:data-consistency redis-py hub redis-py best practices All hubs All clusters

Related clusters: Configuration error Retry storm Race condition

Sources

We don’t republish the full GitHub discussion text. Use the links above for context.

redis-py Connections are not released to a connection pool, when ClusterPipeline faces an exception (Fix)