Browse Prior Art Database

Error Handling and Recovery for Storage Control Units

IP.com Disclosure Number: IPCOM000040394D
Original Publication Date: 1987-Nov-01
Included in the Prior Art Database: 2005-Feb-02
Document File: 2 page(s) / 34K

Publishing Venue

IBM

Related People

Krygowski, MA: AUTHOR

Abstract

The error-handling procedure shown in Fig. 3 allows the storage control elements (SCEs) 10 and 12 of main storage elements (MSEs) 14 and 16 to recover from error conditions without the multiprocessing system shown in Fig. 1 going through a check stop state. Error checkers for the SCEs 10 and 12 detect two types of failures. An error detected by either the type A checker 18 or the type B checker 20 results in the failing SCE notifying the operational SCE and the other functional elements of the system, i.e., CPUs 22, the I/O channels 24 and the processor controllers 26, that it is going into an error-handling and -recovery procedure and not to send anymore requests. If a type A error is detected, the SCE will enter a phase A mode of operation and process all requests outstanding at the time it sends the notice.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 55% of the total text.

Page 1 of 2

Error Handling and Recovery for Storage Control Units

The error-handling procedure shown in Fig. 3 allows the storage control elements (SCEs) 10 and 12 of main storage elements (MSEs) 14 and 16 to recover from error conditions without the multiprocessing system shown in Fig. 1 going through a check stop state. Error checkers for the SCEs 10 and 12 detect two types of failures. An error detected by either the type A checker 18 or the type B checker 20 results in the failing SCE notifying the operational SCE and the other functional elements of the system, i.e., CPUs 22, the I/O channels 24 and the processor controllers 26, that it is going into an error-handling and - recovery procedure and not to send anymore requests. If a type A error is detected, the SCE will enter a phase A mode of operation and process all requests outstanding at the time it sends the notice. It will then go into a phase B mode of operation and accept, but not process, all requests in progress at the time the notice is transmitted to the other units. If the error is detected by the type B checker, it will go directly into the phase B mode of opera

(Image Omitted)

tion. That is, it will not process pending requests, but will log requests in progress at the time of the notice and will completely buffer any in-transit data destined for main storage. When it receives notice that the other SCE is going into error- handling and -detection mode, the operational SCE also goes into the phase A mode o...