Adding load balancing strategy configuration to cluster clients(replacement for 'read_from_replicas' config) by petyaslavova · Pull Request #3563 · redis/redis-py (original) (raw)

Pull Request check-list

Please make sure to review and check all of these items:

• Do tests and lints pass with this change?
• Do the CI tests pass with this change (enable it first in your forked repo and wait for the github action build to finish)?
• Is the new or changed code fully tested?
• Is a documentation update included (if this change modifies existing APIs, or introduces new ones)?
• Is there an example added to the examples folder (if applicable)?
• Was the change added to CHANGES file?

NOTE: these things are not required to open a PR and can be done afterwards / while the PR is open.

Description of change

Adding load balancing strategy configuration to cluster clients(replacement for 'read_from_replicas' config)
The idea of this setting is to replace the currently existing 'read_from_replicas' configuration - its name is confusing. It leaves the impression that all the reads will be sent to replica nodes, while it actually chooses between the primary and replicas in a round-robin manner.
There was one specific situation in which random replica selection was also exposed - with this change, this configuration will be exposed through the RedisCluster objects for sync and async clients.

• Adding a new configuration field for Redis Cluster objects - load_balancing_strategy
• Adding deprecation for one of the current configs - read_from_replicas
• Adding a new enum for the available load-balancing strategies (LoadBalancingStrategy)
• Adding unit tests for the cluster and cluster pipeline operations with the new load-balancing strategies
• Adding unit test coverage for the added load-balancing strategies
• Adding some setup fixes for multiptocessing unit tests (the issue resurfaced for Mac users with a recent merge)