This counts how many times UCS Manager has retried this specific stage of the workflow after a prior attempt at that same stage failed. A retry count climbing across polls while stuck on the same stage name is a stronger, more localized signal than the workflow-level fsmTry field, since it pinpoints exactly which step keeps failing. For example, a retry count of 4 on a chassis-discovery stage specifically, while other stages show 0 retries, tells an admin discovery itself, not some later step, is the recurring failure point.