System and method for maintaining in a multi-processor system a spare processor that is in lockstep for use in recovering from loss of lockstep for another processor

摘要

According to one embodiment, a method comprises assigning a first processor of a multi-processor system a role of spare processor for at least a second processor, and responsive to detecting loss of lockstep (LOL) for any of the at least a second processor, the first processor replaces the processor for which the LOL is detected. The method further comprises reestablishing lockstep for the processor for which the LOL is detected, and assigning the processor having its lockstep reestablished the role of spare processor for at least the first processor.