Abstract: A system, method, and computer program product for reporting and recovering from an internal processor error in a multiprocessor system supporting system management mode. In accordance with the method of the present invention one or more replacement agents are allocated such as during system startup within the multiprocessor system. Machine specific error-reporting registers are monitored for one or more active processor agents during system operation. In response to detecting a faulty agent via the monitoring, a system management interrupt (SMI) request is issued to one or both the faulty agent and a selected replacement agent. In response to receiving the SMI request, the operating state of the faulty agent is copied to the selected replacement agent in system management mode. Operating system processing is then resumed using the replacement agent.

Abstract: A system, method, and computer program product for reporting and recovering from an internal processor error in a multiprocessor system supporting system management mode. In accordance with the method of the present invention one or more replacement agents are allocated such as during system startup within the multiprocessor system. Machine specific error-reporting registers are monitored for one or more active processor agents during system operation. In response to detecting a faulty agent via the monitoring, a system management interrupt (SMI) request is issued to one or both the faulty agent and a selected replacement agent. In response to receiving the SMI request, the operating state of the faulty agent is copied to the selected replacement agent in system management mode. Operating system processing is then resumed using the replacement agent.

Abstract: A system, method, and computer program product for reporting and recovering from an internal processor error in a multiprocessor system supporting system management mode. In accordance with the method of the present invention one or more replacement agents are allocated such as during system startup within the multiprocessor system. Machine specific error-reporting registers are monitored for one or more active processor agents during system operation. In response to detecting a faulty agent via the monitoring, a system management interrupt (SMI) request is issued to one or both the faulty agent and a selected replacement agent. In response to receiving the SMI request, the operating state of the faulty agent is copied to the selected replacement agent in system management mode. Operating system processing is then resumed using the replacement agent.

Abstract: A method and system for dynamically replacing a failing processor in a server system configured with IA-32 architecture without requiring hardware changes to the IA-32 architecture or administrative effort. At least one processor of the multiprocessor system (MP) is initially provided as a reserve (or hot-spare) processor that remains in an idle, off, or low-power mode. While in that mode, the OS is prevented from initially utilizing the hot-spare processor. When a processor failure is detected, SMI code running on a good processor instructs the OS to hold off allocating processes to the failing processor. Contemporaneously, the SMI (and OS) activates and completes an initialization of the hot-spare processor to prepare it to begin receiving the held-off processes. Control is then returned to the OS, which updates the “active” processor list and allocates the threads that were running on the failing processor to the hot-spare processor.

Abstract: A data processing network includes a set of servers, at least one switch module to interconnect the servers, and a management module. The management module consults power state information stored in the network following a power transition and restores power to at least some of the servers and switch modules based on the power state information. The power state information prevents the management module from restoring power to servers and switch modules having incompatible communication protocols. In one embodiment, the plurality of servers and the switch modules are hot-swappable modules that are all inserted into a single chassis. In this embodiment, the server modules and at least one switch module share selected resources of the network including system power. The switch modules and servers may employ Ethernet, fiber channel, optical, and serial communication protocols.

Abstract: When a blade and/or interconnect device is inserted into the chassis of a powered or live server the procedure is known as hot-plugging. Before power is applied to the hot-plugged blade and/or interconnect device the fabric type of already installed blades and/or interconnect devices is correlated with fabric types of newly hot-plugged blade and/or interconnect device. Depending upon results of the correlation, power to the hot-plugged blade and/or interconnect device is allowed or denied.

Abstract: A method for co-operative thermal management of a number of independent electronic devices housed within a common enclosure that includes designating a priority number and measuring a temperature for each of the independent electronic devices. Next, for each of the independent electronic devices, the measured temperature is evaluated to determine if it exceeds a threshold value for the independent electronic device. If the measured temperature exceeds a threshold value, a count-down value is initialize to the designated priority number of the independent electronic device. Following which, the count-down value is evaluated to determine if it is equal to zero. If the count-down value is equal to zero, the independent electronic device is powered down. However, if the count-down value is not equal to zero, a specified time interval is allowed to expire before the count-down value is decremented and re-evaluated to determine if it is equal to zero.

Abstract: A data processing network includes a set of servers, at least one switch module to interconnect the servers, and a management module. The management module consults power state information stored in the network following a power transition and restores power to at least some of the servers and switch modules based on the power state information. The power state information prevents the management module from restoring power to servers and switch modules having incompatible communication protocols. In one embodiment, the plurality of servers and the switch modules are hot-swappable modules that are all inserted into a single chassis. In this embodiment, the server modules and at least one switch module share selected resources of the network including system power. The switch modules and servers may employ Ethernet, fibre channel, optical, and serial communication protocols.

Abstract: When a blade and/or interconnect device is inserted into the chassis of a powered or live server the procedure is known as hot-plugging. Before power is applied to the hot-plugged blade and/or interconnect device the fabric type of already installed blades and/or interconnect devices is correlated with fabric types of newly hot-plugged blade and/or interconnect device. Depending upon results of the correlation, power to the hot-plugged blade and/or interconnect device is allowed or denied.

Abstract: Signals to a management module are generated on an occurrence of an event affecting one of a plurality of server blades housed in a common chassis with the management module and aggregated in the management module. Signaling in accordance with this invention may be originated at a number of levels of operation of information handling systems, and distinction can be drawn between an occurrence requiring prompt attention from an operator—an alert—and occurrences where such prompt action is unnecessary. Occurrences signaled are logged for possible later review, and such a log will, in the contemplation of this invention, contain events related to a number of server blades.

Abstract: A method for co-operative thermal management of a number of independent electronic devices housed within a common enclosure that includes designating a priority number and measuring a temperature for each of the independent electronic devices. Next, for each of the independent electronic devices, the measured temperature is evaluated to determine if it exceeds a threshold value for the independent electronic device. If the measured temperature exceeds a threshold value, a count-down value is initialize to the designated priority number of the independent electronic device. Following which, the count-down value is evaluated to determine if it is equal to zero. If the count-down value is equal to zero, the independent electronic device is powered down. However, if the count-down value is not equal to zero, a specified time interval is allowed to expire before the count-down value is decremented and re-evaluated to determine if it is equal to zero.