Resource integrity during partial backout of application updates

Patent metadata

Field	Value
Publication number	US-9921905-B2
Application number	US-201615081245-A
Country	US
Kind code	B2
Filing date	Mar 25, 2016
Priority date	Feb 10, 2009
Publication date	Mar 20, 2018
Grant date	Mar 20, 2018

Abstract

In response to failure of an application that initiated updates to a group of operational system resources without the updates being successfully committed, for each physically inconsistent operational system resource that was left in a non-fully functional data indexing and access state as a result of the failure of the application, a portion of available pending updates are performed to change the respective physically inconsistent operational system resource to a partially backed out operational system resource with a fully functional data indexing and access state. Remaining available pending updates are ignored for the respective partially backed out operational system resource after the respective fully functional data indexing and access state is achieved to expedite system restart.

First claim

What is claimed is:

1. A method, comprising:
by a processor in response to failure of an application that initiated updates to a plurality of operational system resources without the updates being successfully committed:
for each physically inconsistent operational system resource that was left in a non-fully functional data indexing and access state as a result of the failure of the application:
performing a portion of available pending updates to change the respective physically inconsistent operational system resource to a partially backed out operational system resource with a fully functional data indexing and access state; and
ignoring any remaining available pending updates for the respective partially backed out operational system resource after the respective fully functional data indexing and access state is achieved to expedite system restart.

2. The method of claim 1 , further comprising:
identifying a previous state of fully functional data indexing and access for each of the plurality of operational system resources based at least upon stored recovery log records associated with each of the plurality of operational system resources;
determining, for each of the plurality of operational system resources, whether any non-committed pending update operations by the failed application are indicated by the stored recovery log records at a time of the failure of the application; and
identifying each physically inconsistent operational system resource based upon a determination that at least one non-committed pending update operation by the failed application caused the non-fully functional data indexing and access state of the respective physically inconsistent operational system resource.

3. The method of claim 1 , further comprising retrieving, for each physically inconsistent operational system resource, the respective portion of the available pending updates from stored recovery log records referenced as previously initiated updates to the respective physically inconsistent operational system resource by the failed application.

4. The method of claim 1 , where ignoring any remaining available pending updates for the respective partially backed out operational system resource after the respective fully functional data indexing and access state is achieved comprises ignoring available pending updates referenced by stored recovery log records associated with the respective partially backed out operational system resource after the respective fully functional data indexing and access state is achieved.

5. The method of claim 1 , further comprising marking each partially backed out operational system resource for verification in association with the system restart.

6. The method of claim 5 , further comprising verifying functional data indexing and access state integrity of each partially backed out operational system resource prior to restart of each partially backed out operational system resource.

7. The method of claim 1 , further comprising verifying physical integrity of any of the plurality of operational system resources determined to be in the fully functional data indexing and access state after the failure of the application prior to the system restart.

8. A system, comprising:
a memory; and
a processor programmed to:
in response to failure of an application that initiated updates to a plurality of operation system resources without the updates being successfully committed and using resource recovery information stored in the memory:
for each physically inconsistent operational system resource that was left in a non-fully functional data indexing and access state as a result of the failure of the application:
perform a portion of available pending updates to change the respective physically inconsistent operational system resource to a partially backed out operational system resource with a fully functional data indexing and access state; and
ignore any remaining available pending updates for the respective partially backed out operational system resource after the respective fully functional data indexing and access state is achieved to expedite system restart.

9. The system of claim 8 , where the processor is further programmed to:
identify a previous state of fully functional data indexing and access for each of the plurality of operational system resources based at least upon stored recovery log records associated with each of the plurality of operational system resources;
determine, for each of the plurality of operational system resources, whether any non-committed pending update operations by the failed application are indicated by the stored recovery log records at a time of the failure of the application; and
identify each physically inconsistent operational system resource based upon a determination that at least one non-committed pending update operation by the failed application caused the non-fully functional data indexing and access state of the respective physically inconsistent operational system resource.

10. The system of claim 8 , where the processor is further programmed to retrieve, for each physically inconsistent operational system resource, the respective portion of the available pending updates from stored recovery log records referenced as previously initiated updates to the respective physically inconsistent operational system resource by the failed application.

11. The system of claim 8 , where, in being programmed to ignore any remaining available pending updates for the respective partially backed out operational system resource after the respective fully functional data indexing and access state is achieved, the processor is programmed to ignore available pending updates referenced by stored recovery log records associated with the respective partially backed out operational system resource after the respective fully functional data indexing and access state is achieved.

12. The system of claim 8 , where the processor is further programmed to mark each partially backed out operational system resource for verification in association with the system restart.

13. The system of claim 8 , where the processor is further programmed to verify physical integrity of any of the plurality of operational system resources determined to be in the fully functional data indexing and access state after the failure of the application prior to the system restart.

14. A computer program product comprising a computer useable storage medium including a computer readable program code, where the computer readable program code when executed on a computer causes the computer to:
in response to failure of an application that initiated updates to a plurality of operational system resources without the updates being successfully committed:
for each physically inconsistent operational system resource that was left in a non-fully functional data indexing and access state as a result of the failure of the application:
perform a portion of available pending updates to change the respective physically inconsistent operational system resource to a partially backed out operational system resource with a fully functional data indexing and access state; and
ignore any remaining available pending updates for the respective partially backed out operational system resource after the respective fully functional data indexing and access state is achieved to expedite system restart.

15. The computer program product of claim 14 , where the computer readable program code when executed on the computer further causes the computer to:
identify a previous state of fully functional data indexing and

Assignees

• IBM

Inventors

• Kettley Paul
• Mitchell Ian J

Classifications

• G06F11/0706

  the processing taking place on a specific hardware platform or in a specific software environment · CPC title

• G06F2201/84

  Using snapshots, i.e. a logical point-in-time copy of the data · CPC title

• G06F11/0751

  Error or fault detection not based on redundancy (power supply failures G06F1/30; network fault management H04L41/06) · CPC title

• G06F11/0748

  in a remote unit communicating with a single-box computer node experiencing an error/fault (remote testing G06F11/2294) · CPC title

• G06F11/079

  Root cause analysis, i.e. error or fault diagnosis (in a hardware test environment G06F11/22; in a software test environment G06F11/36) · CPC title