Generic error recovery

• US 5,159,597 A
• Filed: 05/21/1990
• Issued: 10/27/1992
• Est. Priority Date: 05/21/1990
• Status: Expired due to Term

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Monahan, Mary L., Monahan, Christopher J., Willson, Dennis L.
• Primary Examiner(s): Beausoliel, Robert W.
• Assistant Examiner(s): HUA, LY
• Application Number:
  US07/525,927
• Time in Patent Office: 890 Days
• Field of Search: 371/7, 371/16.5, 371/16.1, 371/29.1, 371/17, 371/18
• US Class Current: 714/2
• CPC Class Codes: G06F 11/0751 Error or fault detection not based on redundancy power supply failures G06F1/30; network fault management H04L41/06
• CPC Class Codes: G06F 11/0793 Remedial or corrective actions recovery from an exception in an instruction pipeline G06F9/3861; by retry G06F11/1402; for recovering from a failure of a protocol instance or entity H04L69/40
• CPC Class Codes: G06F 11/1415 at system level

Abstract

An error recovery subsystem which can be easily modified for use with any physical hardware which is being monitored is disclosed. The error recovery subsystem employs a user editable file including the rules for defining the system state, the error states, and the sequences of recovery actions to be taken depending upon the comparison between the system state and the error states. The rules for defining the system state, include don'"'"'t care variables, and the sequences of recovery actions are specified using an index into a set of elemental recovery actions. Because the system state, error state, and sequence of recovery actions are defined in a user editable file modifications to the error recovery scheme can be made without recompiling the error recovery subsystem program code. Such modifications to the error recovery subsystem may therefore be made on a real time basis. A method for recovering from an error and a program product therefore are also disclosed.

Claims

1. An error recovery subsystem for use in a data processing system comprising

a user editable file including rules for machine-executed definition of fields in a system state of the data processing system and including possible error states of the data processing system;

and

means for comparing the system state and the error states for invoking an appropriate sequence of recovery actions depending upon which error state matches the system state, the means for comparing and invoking coupled to the user editable file.

2. The error recovery subsystem of claim 1 wherein the user editable file includes a don'"'"'t care variable to enable one of the error states to represent different possible states of the data processing system having the same appropriate sequence of recovering actions.

3. The error recovery subsystem of claim 2 wherein the user editable file includes sequences of recovery action indices, the indices indexing into a list of individual recovery actions for the data processing system.

4. The error recovery subsystem of claim 2 wherein one of the error states includes all don'"'"'t care variables.

5. The error recovery subsystem of claim 1 wherein the user editable file includes sequences of recovery action indices, the indices indexing into a list of individual recovery actions for the data processing system.

6. The error recovery subsystem of claim 1 wherein the rules only define fields in the system state relevant to a comparison of the system state to at least one error state.

7. An error recovery subsystem for use in an automated storage library comprising:

a user editable file including rules for machine-executed definition of fields in a system state of the automated storage library, including possible error states of the automated storage library using at least one don'"'"'t car variable, and including sequences of recovery action indices, the indices indexing into a list of individual recovery actions for the automated storage library; and

means for comparing the system state and the error states and for invoking a sequence of recovery actions depending upon which error state matches the system state, the means for comparing and invoking coupled to the user editable file.

8. The error recovery subsystem of claim 7 wherein the rules only define fields in the system state relevant to a comparison of the system state to at least one error state.

9. The error recovery subsystem of claim 7 wherein one of the error states includes all don'"'"'t care variables.

10. A method for recovering from an error in a data processing system comprising machine-executed steps of:

determining a system state of the data processing system;

comparing the system state to possible error states; and

invoking a sequence of recovery actions depending upon which error state matches the system state using a sequence of recovery action indices, the indices determined according to which error state matches the system state and indexing into a list of individual recovery actions for the data processing system.

11. The method of claim 10 wherein the sequence of recovery actions can be modified by a recovery action.