File-based cluster-to-cluster replication recovery
View Patent ↗Resiliency in file replication scenarios via maintenance of multiple replication queues with differing susceptibility to various sources of data loss. A plurality of requests are received at a first file system location to write to one or more file system objects at the first file system location. Details about the plurality of requests are recorded in at least two replication queues which differ, at least in part, as to types of failures which will cause data loss. Responsive to detection of data loss in one of the replication queues, details about the plurality of requests are retrieved from another of the replication queues.
1. A computer-implemented method comprising:
receiving, at a first file system location, a plurality of requests to write to one or more file system objects at the first file system location;
recording details about the plurality of requests in both a primary and a secondary replication queue;
responsive to detection of data loss in the primary replication queue, retrieving the details about the plurality of requests from the secondary replication queue;
responsive to receiving the plurality of requests, writing to a first copy of the one or more file system objects at the first file system location;
responsive to the retrieval of the details, replicating results of the plurality of write requests to a second file system location by writing to a second copy of the one or more file system objects at the second file system location;
maintaining a bitmap of inodes that catalogs: (i) changes in changed files, and (ii) a plurality of byte ranges respectively associated with a last consistent state of the changed file; and
ensuring recovery in data replication, failure recovery and filesystem restore scenarios;
wherein:
both the primary and the secondary replication queues are at the first file system location; and
the primary and secondary queues differ, at least in part, as to types of failures which will cause data loss.