Fix RuntimeError in ClusterPubSub sharded message generator

[petyaslavova]

Fix RuntimeError in ClusterPubSub sharded message generator

Problem: RuntimeError: dictionary changed size during iteration when subscribing to multiple sharded channels on different cluster nodes.

Solution: Create a snapshot of node_pubsub_mapping.values() before iteration in _pubsubs_generator() to prevent errors from concurrent modifications.

Changes:
Fixed _pubsubs_generator() in ClusterPubSub class
Added 3 cluster-only tests validating multi-node sharded pubsub and the concurrent mapping changes fix

[Copilot]

Pull request overview

This PR fixes a RuntimeError: dictionary changed size during iteration that occurs when subscribing to multiple sharded channels across different cluster nodes. The issue arose because the generator was iterating directly over node_pubsub_mapping.values() while concurrent subscriptions could modify the dictionary.

Key changes:

• Modified _pubsubs_generator() to create a snapshot of pubsub instances before iteration
• Introduced cycle boundaries using None markers to properly track iteration completion
• Added comprehensive tests validating multi-node sharded pubsub behavior

Reviewed changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated 2 comments.

File	Description
redis/cluster.py	Fixed generator to snapshot mapping values before iteration and added cycle markers
tests/test_pubsub.py	Added three cluster-specific tests for multi-node sharded pubsub scenarios

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Copilot]

Pull request overview

Copilot reviewed 2 out of 2 changed files in this pull request and generated 3 comments.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.