Abstract: A method and apparatus for handling system service requests such as disk change requests in a system containing a primary processor and at least one secondary processor, each of which has one or more processes running thereon. Requests originating from processes running on the primary processor are added directly to a primary queue and the corresponding processes suspended pending the completion of the requested system service, at which time the processes are resumed and the requests are purged from the primary queue. Each secondary processor manages a secondary queue of system service requests originating from processes running on that processor, which are suspended while the requests remain on the secondary queue. When a request is added to an empty secondary queue, a process on the secondary processor issues its own request to the primary processor, which is added to the primary queue, and suspends execution while awaiting a response from the primary processor.