Reducing data volume durability state for block-based storage

摘要

A block-based storage system may implement reducing durability state for a data volume. A determination may be made that storage node replicating write requests for a data volume is unavailable. In response, subsequent write requests may be processed according to a reduced durability state for the data volume such that replication for the data volume may be disabled for the storage node. Write requests may then be completed at a fewer number of storage nodes prior to acknowledging the write request as complete. Durability state for the data volume may be increase in various embodiments. A storage node may be identified and replication operations may be performed to synchronize the current data volume at the storage node with a replica of the data volume maintained at the identified storage node.

主权项

1. A system, comprising:
a plurality of storage nodes, wherein each of the plurality of storage nodes maintains respective replicas of a plurality of data volumes in block-based storage devices implemented at the storage node, wherein a write request received from a client for a data volume of the plurality of data volumes with replication enabled is completed at each respective replica of the data volume prior to acknowledging the write request to the client; and each of the plurality of storage nodes, configured to:
determine that a peer storage node of the plurality of storage nodes maintaining a same replica of a particular data volume of the plurality of data volumes as is maintained at the storage node is unavailable for replication;in response to determining that the peer storage node maintaining the same replica of the particular data volume as is maintained at the storage node is unavailable, disable replication for the particular data volume, comprising:
complete subsequently received write requests from the client for the particular data volume at the storage node; andacknowledge the completed subsequently received write requests from the client for the particular data volume at the storage node as complete without completing the subsequently received write requests from the client for the particular data volume at any other storage node of the plurality of storage nodes.