Dismiss
InnovationQ will be updated on Sunday, Oct. 22, from 10am ET - noon. You may experience brief service interruptions during that time.
Browse Prior Art Database

Early Warning of Pending Microcode/Hardware Failures

IP.com Disclosure Number: IPCOM000103441D
Original Publication Date: 1990-Nov-01
Included in the Prior Art Database: 2005-Mar-17
Document File: 1 page(s) / 38K

Publishing Venue

IBM

Related People

Brey, TM: AUTHOR [+3]

Abstract

In a system provided with a Service Processor, a failure in the Service Processor can cause the main system to fail if the hardware of the Service Processor fails. In this design, the Service Processor has added detection within the Service Processor to recognize that is microcode is not healthy, although the Service Processor is still functional, and for the Service Processor to signal the Primary System, so that the Primary System may continue to run. This is done before the Primary System detects the Service Processor is broken and it prevents the Primary System from calling the Service Processor and thus causing the Primary System from becoming broken during a transaction request to the Service Processor. The alternative to this design, heretofore, has been to system checkstop the Primary System.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 100% of the total text.

Early Warning of Pending Microcode/Hardware Failures

      In a system provided with a Service Processor, a failure in the
Service Processor can cause the main system to fail if the hardware
of the Service Processor fails.  In this design, the Service
Processor has added detection within the Service Processor to
recognize that is microcode is not healthy, although the Service
Processor is still functional, and for the Service Processor to
signal the Primary System, so that the Primary System may continue to
run.  This is done before the Primary System detects the Service
Processor is broken and it prevents the Primary System from calling
the Service Processor and thus causing the Primary System from
becoming broken during a transaction request to the Service
Processor.  The alternative to this design, heretofore, has been to
system checkstop the Primary System.

      The implementation is for the Service Processor to remove
itself from the Primary System, to appear to the Primary System as a
hardware failure upon the detection, by itself, of a selected number
of microcode detected errors.

      This same mechanism can be used by Primary System hardware to
signal the Service Processor that it has degraded itself, although
still functional, to have the Service Processor signal the Primary
System through the same path to shut down before a permanent error
occurs.

      Disclosed anonymously.