| CPC G06F 11/1662 (2013.01) [G06F 3/0622 (2013.01); G06F 3/064 (2013.01); G06F 3/0679 (2013.01); G06F 11/1088 (2013.01); G06F 11/3034 (2013.01); G06F 16/27 (2019.01)] | 19 Claims |

|
1. A non-transitory machine readable medium storing instructions, which when executed by one or more processing resources of a distributed storage system comprising a cluster of a plurality of nodes, cause the distributed storage system to:
implement a redundancy scheme for data blocks stored by the distributed storage system by maintaining at least one redundant copy of a given data block of a first node of the plurality of nodes by at least one other node of the plurality of nodes;
based on encountering a number of block read errors associated with a particular Redundant Array of Independent Disks (RAID) stripe of a RAID group associated with the first node in which the number of block read errors meets or exceeds a predetermined or configurable threshold, identify the particular RAID stripe as a failed RAID stripe; and
restore a plurality of data blocks that were stored within the failed RAID stripe without performing RAID recovery/reconstruction by making use of the redundancy scheme.
|