Resumable Transfer (Checkpoint)

RedisShake 4.x does NOT support resumable transfer or cluster topology change awareness.

Resumable transfer means a data synchronization tool can continue from where it was interrupted. For example, Alibaba Cloud DTS (Data Transmission Service) and Tair Global Active-Active both support resumable transfer - sync tasks can resume from the checkpoint after restart, and automatically recover from network fluctuations or node failovers.

However, implementing production-grade resumable transfer and topology awareness is extremely complex, requiring Redis itself, the sync tool, and surrounding components to be modified together for robust implementation. RedisShake alone cannot achieve this. This is why RedisShake adopts a "stateless synchronization" and "static topology" design.

Therefore, issues such as network interruption, process crash, server restart, cluster scaling, and failover may cause synchronization to terminate, requiring a full resync from the beginning after restart.

Suitable Scenarios: RedisShake is suitable for one-time data migration, not for long-term synchronization scenarios that require high availability. For the latter, consider using commercial products like Alibaba Cloud DTS or Tair Global Active-Active, which provide complete resumable transfer and topology awareness capabilities.