Abstract: Provided are a method, system and program from for managing failures in a mirrored system. A copy relationship between primary and secondary storage locations, wherein updates to the primary storage locations are copied to the secondary storage locations. A failure is detected preventing an application from writing updates to the primary storage locations. A failure message is received for the application in response to detecting the failure, wherein the failure message is not provided to the application. The copying of updates to the primary storage locations to the secondary storage locations is suspended. The failure message is provided to the application in response to suspending the copying.

Abstract: A system creates a secondary stack containing execution information of at least one function operating on the computer system, and receives an attack notification of an attack on the computer system. The system determines a point in the secondary stack at which a recovery from the attack is possible. In one embodiment, the system then generates a signature of the attack based on the execution information contained within the secondary stack.

Abstract: In one embodiment, a method and apparatus for predicting and preventing escalations is disclosed. The method includes generating a predictive model from analysis of historical defect data and corresponding historical escalation data, and applying the predictive model to current defect data to determine an escalation risk of the current defect data.

Abstract: The present invention relates to methods, a system, computer readable mediums and computer program products for recovering an application state of an electronic trading system including a first server and a second server after a fail of both of the servers, wherein the first server functions as a primary server for receiving incoming messages of the system and the second server as a secondary server during operation of the system, wherein each incoming message is processed in both servers. The method comprises the steps of: initiating a recovery procedure after a fail of both servers by restarting the first server and the second server; identifying the last message of respective server; and selecting a server as the primary server on basis of the latest messages.

Abstract: An apparatus, system, and method are disclosed for backing up vital product data for a data processing device. A communication module establishes a communications channel with a maintenance module of a service center. The service center remotely supports the data processing device. A backup module copies vital product data to the maintenance module and the maintenance module maintains the vital product data copy. In one embodiment, a restore module receives a command to restore the vital product data copy. The communication module establishes the communications channel between the restore module and the maintenance module and the maintenance module copies the vital product data copy to the restore module.

Abstract: The invention verifies the validity of instructions transmitted between controllers without complicating the structure of the electrical connection between controllers. The initial setup detection unit detects whether an initial setup of the lower communications network has been completed. When a fault in the path to the storage unit has been detected by the fault detection unit, the switching request output unit outputs a switching request signal to the switching permission unit. When the preparation status detection unit is aware that the lower communications network has completed initial setup, the preparation status detection unit outputs the preparation complete signal to the switching permission unit. When both the switching request signal and the preparation complete signal are input, the switching permission unit outputs the switching signal to the path switching unit. The path switching unit switches the path in accordance with the signal from the switching permission unit.

Abstract: Resynchronization of data between a primary (production) data site and a secondary (recovery) site following a failure is enhanced when the size of a data track at the production site is different from the size of a data track at the recovery site. The recovery site reads an out-of-sync (OOS) bitmap created at the production site and expands or contracts the bitmap to accommodate the size difference. The resulting production site bitmap is merged with a OOS bitmap maintained at the recovery site to indicate those tracks which are to be transferred from the recovery site to the production site. Thus, only those tracks which are required to be transferred are transferred. Buffer space may be allocated in which to expand or contract the production site OOS bitmap. Buffer space may be conserved by sequentially reading portions of the production site OOS bitmap into a small buffer.

Abstract: A method and data processing system for isolating a faulty component in a computer. A first microcontroller detects a fault in a component of a computer. Responsive to detecting the fault, the first microcontroller sets a first fault record for the component to pending fault, sets a second fault record for the first microcontroller to pending fault, and fails over to a second microcontroller. If the second microcontroller detects the fault in the component of the computer, then the first fault record for the component is set to permanent fault, and the second fault record for the first microcontroller is cleared. If the second microcontroller determines the component of the computer does not have the fault, then the first fault record for the component is cleared, and the second fault record for the first microcontroller is set to permanent fault.

Abstract: One aspect of the present invention involves storing the operating time offset of a thread within an area of memory of the thread, and storing a load time and a relative load time for each frame of the thread stored within an area of memory of the frame. The operating time offset tracks the amount of time the thread has actively operated, the frame load time tracks the actual time the frame was loaded, and the relative frame load time tracks the time the frame was loaded after being adjusted for the inactivity of the thread. When a thread dump is performed at a specified time, a frame age and a relative frame age can be calculated for each frame based on the time of the thread dump, the operating time offset of the thread, the load time of the frames within the thread, and the relative load time of the frames within the thread.

Abstract: A serial SCSI (SAS) storage drive system includes a drive enclosure having a first interface card coupled to one storage controller over a single SAS path and a second interface card coupled to another storage controller over a different single SAS path. At least one disk drive within the enclosure is assigned to the first storage controller and interconnected to the storage controller through the first interface card. At least a second disk drive within the enclosure is assigned to the second storage controller and interconnected to the storage controller through the second interface card. The interface cards are selectively interconnected with each other through a crossover port. In the event of a failure in a storage controller or an interface card, the crossover port may be activated, thereby maintaining access to both sets of disk drives.

Abstract: An apparatus, system, and method are disclosed for providing parallel access to a data set configured for automatic recovery. The apparatus includes an access module, a conversion module, a recovery module, and a restore module. The access module establishes parallel access to a data set configured for “pair and a spare” automatic recovery. The conversion module converts access to the data set from parallel access to single point access in response to an application instance detecting an error in the data set. The recovery module recovers the data set using an automatic recovery process initiated by the application instance that detects the error. The restore module restores access to the data set from single point access to parallel access in response to successful recovery of the data set. Providing parallel access to a data set configured for automatic recovery ensures that data transactions are performed swiftly, reliably, and efficiently.

Abstract: A method and system for writing a data image in the main memory of a computer to a secondary storage device minimizes delay in performing a required reboot. Prior to the reboot, a portion of the physical memory is copied either to a reserved memory space or to secondary storage, or one portion of memory may be copied to a reserved memory space and another portion may be copied to secondary storage. After reboot, memory usage is initially limited to the portion of memory that was copied prior to the reboot. The remaining physical memory is incrementally written to the storage device and dynamically added as memory for use by the operating system. If the reboot is due to a crash or other problem, completion of a full memory dump may be cancelled if, upon scanning memory, it is determined that the problem is a known issue.

Abstract: A solution for managing a computer infrastructure is provided. In particular, the solution can generate a set of tentative recovery segments for the computer infrastructure based on a related resource graph and a set of partition rules. The related resource graph can be generated based on a managed resource graph, which includes resource type information for each node in the managed resource graph. The set of tentative recovery segments can be used to manage the computer infrastructure and/or a set of recovery segments, which includes recovery characteristic(s) for each segment can be generated based on the set of tentative recovery segments and a set of recovery segment templates. In any event, the invention provides a solution that fully and/or partially automates the identification of recovery segments within a computer infrastructure, thereby enabling a more effective management of the computer infrastructure.

Abstract: The present invention provides a new multi nodal computer system comprising a number of nodes on which chips of different types reside. The new multi nodal computer system is characterized in that there is one clock chip per node, each clock chip controlling only the chips residing on that node said chips being appropriate for sending a check stop request to the associated clock chip in case of a malfunction. A new check stop handling method is characterized in that depending on the source of the check stop request the clock chip that received the check stop request initiates a system check stop, a node check up, or a chip check stop.

Abstract: A cluster system coupled to a client computer via a network has plural nodes which provide services to the client computer. A first node of the plural nodes has a priority recognizing processing module, which consults a priority of a frame received from the client computer. A second node of the plural nodes has a fail over processing module, which performs fail over on plural software resources managed by the first node according to an order of fail over priority that is based on the priority consulted by the priority recognizing processing module of the first node.

Abstract: A method is provided that rejuvenates a software application to reduce the effects of software aging. An active replica corresponding to a software application is identified. If rejuvenation of the software application is appropriate, a new replica is created and state information is transferred from the active replica to the new replica. In addition, client requests are redirected to the new replica. After the state data has been transferred and requests have been redirected, the active replica is terminated. Once the active replica has been terminated, the new replica becomes the active replica. When rejuvenation is again proper, another new replica is created and the state data is transferred from the new active replica to the new replica and requests are redirected to the new replica. This process repeats whenever rejuvenation of the application is needed.