US5463733A
Expired / Lapsed
US
⏱️ 0.0 years remaining
Failure recovery apparatus and method for distributed processing shared resource control
Assignee: International Business Machines Corp
•
Filing: 1994-08-08
•
Granted: 1995-10-31
•
Expires: 2014-08-08
Abstract
Abstract Communicating the failure of a master process controlling one or more shared resources to all process sharing the resources. A shared resource control file is established that contains the identities of all sharing processes. Master process failure triggers a race to establish exclusive access over the shared control file. The new master reads shadow address data from the old shared control file, marks it as invalid and establishes a new control file based on renewed registrations from the sharing processes. The master process maintains the sharing process list as processes begin and end sharing.
Inventors
- 👤 Ira R. Forman
- 👤 Hari H. Madduri
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-11-01 14:14
CPC Classifications
G; G06; G06F; G06F11/00; G06F11/07; G06F11/16; G06F11/20; G06F11/202; G06F11/2023; G06F11/2033; G06F11/2028; G06F11/2035
Claims Analysis
Total Claims: 7
Claim 1
Independent Claim
178 words
A method for managing recovery of a distributed processing system in which shared resources are each controlled by a master process, the distributed processing system having a plurality of processors each executing a plurality of processes and each controlled by a separate operating system, each of said processors having memory, and each of said processors interconnected to the other processors by means of a communications network, the method comprising the steps of: detecting failure of a master process for a shared resource; requesting exclusive access to a control file using a network file system management procedure independent of said operating system for said processors, if said detecting step detects a failure; establishing exclusive access to said control file using said network file system management, if said exclusive access request is granted; determining from said control file all other processes accessing said shared resource; invalidating said control file; sending a message to each of said other processes indicating failure of said master process; and creating a new control file and entering data for each process responding to said message.
Claim 2
Independent Claim
178 words
A method for managing recovery of a distributed processing system in which shared resources are each controlled by a master process, the distributed processing system having a plurality of processors each executing a plurality of processes and each controlled by a separate operating system, each of said processors having memory, and each of said processors interconnected to the other processors by means of a communications network, the method comprising the steps of: detecting failure of a master process for a shared resource; requesting exclusive access to a control file using a network file system management procedure independent of said operating system for said processors, if said detecting step detects a failure; establishing exclusive access to said control file using said network file system management, if said exclusive access request is granted; determining from said control file all other processes accessing said shared resource; invalidating said control file; sending a message to each of said other processes indicating failure of said master process; and creating a new control file and entering data for each process responding to said message.
Claim 3
Dependent Claim
37 words
The method of claim 1, wherein the step of detecting failure comprises the steps of: attempting to communicate with a master process for a shared resource; signalling failure of a master process if no response is received.
Claim 4
Independent Claim
177 words
A system for failure recovery in a distributed processing computer system in which shared resources are each controlled by a master process selected among a plurality of peer processes, said master process having exclusive access to a shared resource control file for said resource, each of said processes executing in a computer system having a processor and memory, and each processor controlled by a separate operating system, the system comprising: means for detecting master failure by a shadow process; means for requesting exclusive access to a resource control file in response to said means for detecting using a file system management procedure independent of said operating system controlling said processors; means for establishing exclusive access using said file system management procedure if said request is granted; reading means for reading from said resource control file addresses of other processes accessing said shared resource; communications means for communicating master failure and recovery to said other processes; and writing means for invalidating said resource control file and recreating a resource control file based on responses to said communication means.
Claim 5
Independent Claim
177 words
A system for failure recovery in a distributed processing computer system in which shared resources are each controlled by a master process selected among a plurality of peer processes, said master process having exclusive access to a shared resource control file for said resource, each of said processes executing in a computer system having a processor and memory, and each processor controlled by a separate operating system, the system comprising: means for detecting master failure by a shadow process; means for requesting exclusive access to a resource control file in response to said means for detecting using a file system management procedure independent of said operating system controlling said processors; means for establishing exclusive access using said file system management procedure if said request is granted; reading means for reading from said resource control file addresses of other processes accessing said shared resource; communications means for communicating master failure and recovery to said other processes; and writing means for invalidating said resource control file and recreating a resource control file based on responses to said communication means.
Claim 6
Dependent Claim
27 words
The system of claim 3 wherein said means for detecting failure is a network operating system that informs communicating processes of a failure of the other process.
Claim 7
Dependent Claim
32 words
The system of claim 3 wherein said means for detecting failure comprises: communication means for communicating with said master process; and means for detecting lack of communications response from said master process.
References Cited (Backward Citations)
US4408649A
Cited
US4590554A
Cited
US4634110A
Cited
US4709325A
Cited
US5058056A
Cited
US4827399A
Cited
US5003464A
Cited
US5235700A
Cited
US5303243A
Cited
US5214778A
Cited
US5253359A
Cited
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 1998-10-30 | 1999-05-01 | 1999-10-31 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 2002-10-30 | 2003-05-01 | 2003-10-31 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2006-10-30 | 2007-05-01 | 2007-10-31 | $7700.00 | Window Closed / Lapsed |