Method, apparatus and program product for disruptive recovery in a data processing system

• US 5,732,206 A
• Filed: 07/23/1996
• Issued: 03/24/1998
• Est. Priority Date: 07/23/1996
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Mendel, Gili
• Primary Examiner(s): Hua, Ly
• Application Number:
  US08/685,295
• Time in Patent Office: 609 Days
• Field of Search: 395/182.02, 395/182.13, 395/182.14, 395/182.08, 395/682, 395/670, 395/200.1, 370/221
• US Class Current: 714/4.1
• CPC Class Codes: G06F 11/1438 Restarting or rejuvenating

Abstract

A method, apparatus and program product for performing a disruptive recovery in a data processing system including establishing an upper limit of delay that may be taken before starting a recovery process and establishing a set time for stalling. A first event for which a recovery process is normally started is then received and the start of the recovery process is stalled for the set time for stalling after receipt of the first event. Events subsequent to the first event are received until either the time between the receipt of successive events exceeds the set time for stalling, or until the upper limit of delay between the receipt of the first event and the receipt of the last received event is exceeded. After the time limit is exceeded, the recovery process is performed and the operation is started over with the receipt of a new first event.

Claims

1. A method for performing a disruptive recovery in a data processing system comprising the steps of:

establishing an upper limit of delay that may be taken before starting a recovery process;

establishing a set time for stalling;

waiting for the receipt of a first event for which a recovery process is normally started;

stalling the start of the recovery process for said set time for stalling after receipt of said first event;

receiving events subsequent to said first event until either the time between the receipt of successive events exceeds said set time for stalling, or until the upper limit of delay between the receipt of said first event and the receipt of the last received event is exceeded;

conditionally starting said recovery process; and

after completion of said recovery process, returning to said waiting for receipt of said first event step.

2. The method of claim 1 wherein, before the starting said recovery process step, including the following steps:

determining if the recovery process is still needed; and

if said recovery process is not needed, returning to said waiting for the receipt of said first event step.

3. An apparatus for performing a disruptive recovery in a data processing system comprising:

means for waiting for the receipt of a first event for which a recovery process is normally started;

means for stalling the start of the recovery process for a predetermined set time for stalling after receipt of said first event;

means for receiving events subsequent to said first event until either the time between the receipt of successive events exceeds said set time for stalling, or until the delay between the receipt of said first event and the receipt of the last received event exceeds a predetermined upper limit;

means for conditionally starting said recovery process; and

means for returning to said waiting for receipt of said first event after completion of said recovery process.

4. The apparatus of claim 3 including:

means for determining if the recovery process is still needed; and

if said recovery process is not needed, means for returning to said waiting for the receipt of said first event without starting the recovery process.

5. A program product recorded on a computer readable medium, the program product for performing a disruptive recovery in a data processing system, the program product comprising:

means for waiting for the receipt of a first event for which a recovery process is normally started;

means for stalling the start of the recovery process for a predetermined set time for stalling after receipt of said first event;

means for receiving events subsequent to said first event until either the time between the receipt of successive events exceeds said set time for stalling, or until the delay between the receipt of said first event and the receipt of the last received event exceeds a predetermined upper limit;

means for starting said recovery process; and

means for returning to said waiting for receipt of said first event after completion of said recovery process.

6. The program product of claim 5 including:

means for determining if the recovery process is still needed; and

if said recovery process is not needed, means for returning to said waiting for the receipt of said first event without starting the recovery process.

7. A method for merging disruptive recoveries in a data processing system comprising the steps of:

a) calculating a MAXLIMIT to be equal to a predetermined upper limit;

b) waiting for the receipt of a first event for which a recovery process is normally started;

c) setting a timeout period equal to the lesser of a predetermined stall time or the last calculated MAXLIMIT;

d) waiting for said timeout period;

e) if a subsequent event for which a recovery process is normally started is received before the end of the timeout period, recalculating the MAXLIMIT equal to MAXLIMIT less the waiting time; and

f) returning to step c.

8. The method of claim 7 further comprising the steps of:

g) if an event for which recovery is normally started is not received before the end of the last calculated timeout period, performing a diagnosis to determine if a recovery is still needed;

h) if recovery is not needed returning to step a; and

i) if step g determines that recovery is still needed, performing said recovery, and

j) returning to step a, thereby merging disruptive recoveries into a single recovery process.

9. An apparatus for merging disruptive recoveries in a data processing system, said apparatus comprising:

means for calculating a MAXLIMIT to be equal to a predetermined upper limit;

receipt means for waiting for the receipt of a first event for which a recovery process is normally started;

setting means connected to said receipt means for setting a timeout period equal to the lesser of a predetermined stall time or the last calculated MAXLIMIT after receipt of an event for which recovery is normally started;

waiting means connected to said setting means for waiting for said timeout period;

first determining means connected to said waiting means for determining if a subsequent event for which a recovery process is normally started is received before the end of the timeout period;

recalculating means connected to said first determining means for recalculating the MAXLIMIT equal to MAXLIMIT less the waiting time if said first means for determining determines that a subsequent event is received before the end of the time period; and

means connected to said recalculating means for returning to said receipt means to wait for another first event after said MAXLIMIT is recalculated.

10. The apparatus of claim 9 further comprising:

second determining means connected to said first determining means for performing a diagnosis to determine if a recovery is still needed if said first determining means determines that an event for which recovery is normally started was not received before the end of the last calculated timeout period;

means connected to said second determining means for returning to said setting means thereby setting said MAXLIMIT equal to said upper limit if said second determining means determines that recovery is not needed; and

means connected to said second determining means for performing said recovery if said second determining means determines that recovery is still needed, thereby merging disruptive recoveries into a single recovery.

11. A program product recorded on a computer readable medium, said program product for merging disruptive recoveries in a data processing system, said program product comprising:

means for calculating a MAXLIMIT to be equal to a predetermined upper limit;

receipt means for waiting for the receipt of a first event for which a recovery process is normally started;

setting means for setting a timeout period equal to the lesser of a predetermined stall time or the last calculated MAXLIMIT after receipt of an event for which recovery is normally started;

waiting means for waiting for said timeout period;

first determining means for determining if a subsequent event for which a recovery process is normally started is received before the end of the timeout period;

recalculating means for recalculating the MAXLIMIT equal to MAXLIMIT less the waiting time if said first means for determining determines that a subsequent event is received before the end of the time period; and

means for returning to said receipt means to wait for another first event after said MAXLIMIT is recalculated.

12. The program product of claim 11 further comprising:

second determining means for performing a diagnosis to determine if a recovery is still needed if said first determining means determines that an event for which recovery is normally started was not received before the end of the last calculated timeout period;

means for returning to said setting means thereby setting said MAXLIMIT equal to said upper limit if said second determining means determines that recovery is not needed; and

means for performing said recovery if said second determining means determines that recovery is still needed, thereby merging disruptive recoveries into a single recovery.