Browse Prior Art Database

Operational Surveillance Function

IP.com Disclosure Number: IPCOM000113591D
Original Publication Date: 1994-Sep-01
Included in the Prior Art Database: 2005-Mar-27
Document File: 2 page(s) / 76K

Publishing Venue

IBM

Related People

Claffey-Cohen, M: AUTHOR [+7]

Abstract

Disclosed is an Operational Surveillance Function (OSF) providing an automatic method to recover from a catastrophic error in a computing system, when such an error would otherwise require a total system shutdown with power off. When this function is enabled, the OSF can discern when the system processor is no longer servicing critical system functions and can provide a reset mechanism occurring without operator intervention.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 52% of the total text.

Operational Surveillance Function

      Disclosed is an Operational Surveillance Function (OSF)
providing an automatic method to recover from a catastrophic error in
a computing system, when such an error would otherwise require a
total system shutdown with power off.  When this function is enabled,
the OSF can discern when the system processor is no longer servicing
critical system functions and can provide a reset mechanism occurring
without operator intervention.

      The OSF includes a sense input, a timer, and a reset output, as
well as enable and status controls provided so that application
programs can determine whether a catastrophic error has occurred.
The OSF also incoroprates a diagnostic function to allow testing of
the OSF without requiring a real error input.

      A system "hang" problem is usually caused by a catastrophic
error that generates a Non-Maskable Interrupt (NMI), halting the
system.  Some personal computer systems have an NMI service routine,
which attmepts to determine the causes of the NMI, to save the event
in battery-backed memory and to reboot the system (1,2,3).  However,
when the operating system cannot process the NMI service routine and
the system is hung, the system power supply must be powered down.
When this process is required in a system installed at a remote
location, the system may remain out of service for an unacceptably
long time.  Thus, a system processor can receive hardware
non-maskable interrupts, which cannot be ignored, from system
functions that must be serviced to maintain proper system operation.
If the processor cannot service an NMI, a catastrophic error has
occurred, from which the processor has no chance of self-recovery.

      The OSF determines the occurrence of a non-maskable interrupt
by monitoring the level of the -NMI signal of the microprocessor.  If
the system is operating normally, the -N...