Skip to content

Fix memory leak in JedisClusterInfoCache - replica nodes not cleared #4191

@cvanwage

Description

@cvanwage

If the client is configured with isReadOnlyForRedisClusterReplicas == true, calls to reset() within JedisClusterInfoCache will clear out the slots and slotNodes, but will not wipe out replicaSlots. This means that any calls to assignSlotsToReplicaNode, will cause the lists to grow indefinitely.

This can cause a memory leak when a node experiences a failure due to repeated calls to refresh the slots.

Metadata

Metadata

Assignees

No one assigned

    Labels

    Type

    No type

    Projects

    No projects

    Milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions