VMware, Inc.
Facilitating the Recovery of Full HCI Clusters

Abstract:

Techniques for facilitating the recovery of a full hyper-converged infrastructure (HCI) cluster are provided. In one set of embodiments, a hypervisor of a host system in the HCI cluster can create a recovery reserve on a disk group of the host system, where the disk group stores a namespace object component of a virtual machine (VM), and where the recovery reserve corresponds to an amount of storage space reserved on the disk group for the namespace object component. Then, at a time of receiving a write request directed to the namespace object component, the hypervisor can (1) determine that the disk group is out of space, (2) determine that the write request is a recovery I/O (input/output) that is useful for freeing space on the disk group, and upon determining (1) and (2), allow the write request to succeed by consuming space from the recovery reserve.