Recovery facility for incomplete sync points for distributed application

• US 5,319,774 A
• Filed: 05/16/1990
• Issued: 06/07/1994
• Est. Priority Date: 05/16/1990
• Status: Expired due to Term

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Szczygielski, Thomas J., Tanner, Amos S., Showalter, James M., Bennett, Robert B., Maslak, Barbara A. M., Ainsworth, Michael K.
• Primary Examiner(s): Beausoliel, Jr., Robert W.
• Assistant Examiner(s): HUA, LY
• Application Number:
  US07/525,938
• Time in Patent Office: 1,483 Days
• Field of Search: 395/575, 395/200, 395/600, 395/650, 371/7, 371/8.2, 371/11.3, 371/12, 364/281.3, 364/281.6
• US Class Current: 714/20
• CPC Class Codes: G06F 11/1402 Saving, restoring, recovering or retrying
• CPC Class Codes: G06F 9/466 Transaction processing

Abstract

To recover a failed commit procedure, a first recovery facility collects and compares sync point state information from a variety of different types of resources including a data repository and a protected conversation. The comparison of the sync point state information is made automatically in response to a sync point failure event. A second recovery facility serving another execution environment notifies the first recovery facility after a failure and subsequent recovery procedure of the second execution environment has completed. The first recovery facility makes automatic cyclic attempts to communicate with the second recovery facility. The first recovery facility can receive a manually initiated but automatically simulated response of sync point state information when there is a failure such that actual sync point state information cannot be received. The first recovery facility is responsive to the simulated sync point state information to proceed with a recovery operation substantially as if actual sync point state information were received.

Claims

1. A computer system or network comprising:

a first execution environment for executing a first application that makes one or more work requests including a work request to make a protected conversation with a second application executing in a second execution environment;

first recovery facility means, serving said first execution environment, for recovering a commit procedure for said one or more work requests of said first application;

a second execution enviroment for executing said second application, said second application making one or more work requests; and

means for communicating said protected conversation from said first execution environment to said second execution environment, said protected conversation requesting or requiring said one or more work requests made by said second application, said one or more work requests of said first application and said one or more work requests of said second application being related to one another and having a common commit scope;

second means, serving said second execution environment, for supplying syncpoint state information via the communicating means to said first recovery facility means for said protected conversation communicated to said second execution environment or said one or more work requests of said second application; and

keyboard means, coupled to said first recovery facility means, for entering syncpoint state information for said protected conversation communicated to said second execution environment or said one or more work requests of said second application when said second means cannot supply said syncpoint state information to said first recovery facility means for said protected conversation communicated to said second execution environment or said one or more work requests of said second application; and

wherein said first recovery facility means includes means for receiving and processing said sync point state information from said keyboard means when said second means cannot supply said syncpoint state information to said first recovery facility means for said protected conversation communicated to said second execution environment or said one or more work requests of said second application, said first recovery facility means being responsive to said sync point state information from said keyboard means to proceed with a recovery operation substantially as if said sync point state information was received from said second means.

2. A computer system or network as set forth in claim 1 wherein said commit procedure is two phase and said sync point state information indicates points of completion in said two phase commit procedure.

3. A computer system or network as set forth in claim 1 further comprising:

a first real machine which includes said first execution environment;

first sync point manager means for controlling the commit procedure for said one or more work requests of said first application;

a second real machine including said second execution environment;

second sync point manager means for controlling a commit procedure for said one or more work requests of said second application; and

a second recovery facility which includes said second means.

4. A computer network comprising:

a first real machine comprising

a first execution environment,

means for coupling a manager of a resource to said first execution environment to permit said first execution environment to access said resource, the resource manager maintaining a sync point log for said resource,

first means for coordinating a first commit procedure initiated by said first execution environment, said first commit procedure committing a first work request made by a first application executing within said first execution environment for said resource,

first recovery facility means, coupled to the first coordinating means to receive sync point information from the first coordinating means, for recovering said first commit procedure, said first recovery facility means including a first sync point log to store said sync point information received from said first coordinating means, said recovery of said first commit procedure including a comparison of sync point information in said first sync point log with sync point information in the resource manager'"'"'s sync point log, and

first coupling means for coupling said first recovery facility means to said resource manager, said first recovery facility means and the resource manager making said comparison by exchanging sync point information via said first coupling means; and

a second real machine comprising

a second execution environment,

first means for coupling said resource manager to said second execution environment to permit said second execution environment to access said resource,

second means for coordinating a second commit procedure initiated by said second execution environment, said second commit procedure committing a second work request made by a second application executing within said second execution environment for said resource, said second work request being independent of said first work request,

second recovery facility means, coupled to the second coordinating means to receive sync point information from the second coordinating means, for recovering said second commit procedure, said second recovery facility means including a second sync point log to store said sync point information received from said second coordinating means, said recovery of said second commit procedure comprising a comparison of sync point information in said second sync point log with sync point information in the resource manager'"'"'s sync point log, and

second coupling means for coupling said second recovery facility means to said resource manager, said second recovery facility means and said resource manager making said comparison by exchanging sync point information via said second coupling means.

5. A computer network as set forth in claim 4 wherein

said exchange of sync point information between said first recovery facility means and said resource manager bypasses said first coordinating means, said second coordinating means and said second recovery facility means;

and

said exchange of syncpoint information between said second recovery facility means and said resource manager bypasses said first coordinating means, said second coordinating means and said first recovery facility means.

6. A computer network as set forth in claim 4 wherein

said first coordinating means also coordinates a third commit procedure initiated by said first execution environment for a third work request for said resource and a protected conversation with said second execution environment;

and

said first and second recovery facility means recover said third commit procedure by comparing sync point information in said first sync point log with sync point information in said second sync point log; and

further comprising

third coupling means for coupling said first recovery facility means to said second recovery facility means, said first and second recovery facility means making said recovery of said third commit procedure by exchanging sync point information via said third coupling means, bypassing said first coordinating means and said second coordinating means.