Abstract: Techniques are disclosed for early detection of fan failure in a computing device, such as a server. A management device located in the computing device or remote from the computing device can receive information about the fan duty and use the fan duty information to determine a safe fan speed below which future or imminent fan failure can be inferred. The management device can receive the current speed of the fan and compare it with the safe fan speed to determine if the fan is likely to fail, in which case the management device can direct the computing device to perform mitigation action and/or a maintenance warning can be sent. The maintenance warning can result in replacement of the failing or soon-to-fail fan.

Abstract: A system and method for tracking the components of a computing device is disclosed. Each of the components of the computing devices includes identification data. At least one bus is coupled to the components. A memory device stores identification information corresponding with each of the components. A management controller is coupled to the bus and the memory device. The management controller periodically requests identification data from each of the components over the bus. The management controller receives the requested identification data from each of the components. The management controller compares the received identification data with the stored identification data to determine if any of the components has been replaced.

Abstract: Techniques are disclosed for early detection of fan failure in a computing device, such as a server. A management device located in the computing device or remote from the computing device can receive information about the fan duty and use the fan duty information to determine a safe fan speed below which future or imminent fan failure can be inferred. The management device can receive the current speed of the fan and compare it with the safe fan speed to determine if the fan is likely to fail, in which case the management device can direct the computing device to perform mitigation action and/or a maintenance warning can be sent. The maintenance warning can result in replacement of the failing or soon-to-fail fan.

Abstract: A system and method for tracking the components of a computing device is disclosed. Each of the components of the computing devices includes identification data. At least one bus is coupled to the components. A memory device stores identification information corresponding with each of the components. A management controller is coupled to the bus and the memory device. The management controller periodically requests identification data from each of the components over the bus. The management controller receives the requested identification data from each of the components. The management controller compares the received identification data with the stored identification data to determine if any of the components has been replaced.

Abstract: A method and system writing a server BIOS image to a BIOS flash by a BMC is provided. The method includes receiving a Golden BIOS image within a BMC flash. The BMC flash is associated with the BMC. The method also includes determining that an active BIOS image within a BIOS flash is either a damaged BIOS image or a maliciously altered BIOS image. In response to a damaged or a maliciously altered BIOS image, the method includes overwriting the active BIOS image in the BIOS flash with the Golden BIOS.

Abstract: The present disclosure describes devices and methods that include a heartbeat signal for detecting an abnormality in the control of a fan for an electronic system. A fan controller can assume control of a fan over a fan speed control circuit in response to the abnormality. Control of the fan by the fan controller can eliminate or reduce overheating within the system caused by improper fan speed, in response to the abnormality.

Abstract: A system to record equipment location in an equipment rack is disclosed. The system includes a support frame having at least one slot. An identification label is located in proximity to the support frame. The identification label corresponds to the location of the slot. A device is installed in the at least one slot. An imaging sensor is coupled to the device and captures an image of the identification label.

Abstract: Technologies for remote keyboard-video-mouse sessions can include failover mechanisms. In some embodiments, a system can establish a first remote keyboard-video-mouse session between a console application on the system and a first server application executed by a controller on a server. The first remote keyboard-video-mouse session can be established via a first network connection between the system and the controller. Next, the system can detect an error associated with the first remote keyboard-video-mouse session. In response to the error, the system can establish a second remote keyboard-video-mouse session between the console application and a second server application executed by an operating or a basic input/output program on the server. The second remote keyboard-video-mouse session can be established via a second network connection between the system and the second server application.

Abstract: A system to record equipment location in an equipment rack is disclosed. The system includes a support frame having at least one slot. An identification label is located in proximity to the support frame. The identification label corresponds to the location of the slot. A device is installed in the at least one slot. An imaging sensor is coupled to the device and captures an image of the identification label.

Abstract: Various embodiments of the present technology provide methods for monitoring storage needs of one or more server systems, determining a specific server system that does not need storage capacity from corresponding storage subsystems, determining one or more storage subsystems corresponding to the specific server system, and causing at least one component of the one or more storage subsystems to be switched into a power saving mode.

Abstract: A method for ensuring management controller firmware security, by a security manager of a computing device, includes storing a public key and raw identity data, and obtaining, from a management firmware for a management controller of the computing device, encrypted identity data. The security manager decrypts the encrypted identity data with the public key into decrypted identity data, and compares the decrypted identity data with the raw identity data to determine whether the management firmware is authentic. The security manager protects the computing device from harm by the management firmware, in response to determining that the management firmware is not authentic.

Abstract: A method for controlling air flow in a server rack includes storing, by a switch, a connection table, the connection table comprising information for rack positions corresponding to switch ports of the switch. The switch obtains address information of at least one management controller corresponding to at least one device connected to the switch ports of the switch. The switch determines a rack position for each of the at least one device, based on the table and the address information. The switch receives from a rack management controller (RMC), a request for the rack position for each of the at least one device. The switch sends to the RMC, the rack position for each of the at least one device. The RMC controls fan speed for the at least one device, based on the rack position for each of the at least one device.

Abstract: In some implementations, network interface controller (NIC) configuration information can be obtained from a NIC prior to booting up an operating system. For example, a Basic Input Output System (BIOS) can obtain the NIC configuration information from the NIC during the execution of a system check (e.g., Power-On Self-Test). A system controller can receive the NIC configuration information from the BIOS. The system controller can store the NIC configuration information in memory associated with the system controller. A management system can request the NIC configuration information from the system controller using an out-of-band communication channel. For example, the management system can send the request for NIC configuration information to the system controller prior to powering on a server using a dedicated network interface of the system controller.

Abstract: A computer system and a method for setting basic input/output system (BIOS) are disclosed. The computer system comprises a remote computer and servers. The remote computer transmits a setting command. Each of the servers comprises a first management unit and a motherboard. The motherboard comprises a storage device and a processor. The storage device stores the BIOS. The processor executes the BIOS. The processor communicates with the first management unit to determine whether the BIOS configuration needs to be updated after the server rebooted. The processor updates the BIOS according to the setting command when the BIOS configuration must to be updated.

Abstract: A system and method of updating device firmware on a rack server computer system using a network switch. The network switch receives a request to update the firmware of a device in the rack server computer system. The network switch determines the location of the device in the rack server system and transmits the firmware update to the device and commands the device to update its firmware using the firmware update transmitted by the switch. Alternatively, the network switch automatically detects a new device connected to the network switch in the rack server computer system. The network switch determines the identification of the new device, including its firmware, and determines if a newer version of the device firmware is available. The network switch obtains the newer version of the device firmware, transmits the firmware to the device, and commands the device to update its firmware with the newer version.

Abstract: Technologies for remote keyboard-video-mouse sessions can include failover mechanisms. In some embodiments, a system can establish a first remote keyboard-video-mouse session between a console application on the system and a first server application executed by a controller on a server. The first remote keyboard-video-mouse session can be established via a first network connection between the system and the controller. Next, the system can detect an error associated with the first remote keyboard-video-mouse session. In response to the error, the system can establish a second remote keyboard-video-mouse session between the console application and a second server application executed by an operating or a basic input/output program on the server. The second remote keyboard-video-mouse session can be established via a second network connection between the system and the second server application.

Abstract: In some implementations, a system controller of a computing device can be configured to obtain power supply unit (PSU) fan airflow direction information. The system controller is configured to retrieve t a pre-defined airflow direction of PSU fan, and to determine whether the current airflow direction of PSU fan is consistent with the retrieved pre-defined fan airflow direction. In some implementations, upon determining that the current airflow direction of PSU fan is inconsistent with the pre-defined fan airflow direction, the system controller can cause an indication device associated with the PSU to be toggled. The indication device indicates that the PSU fan has been incorrectly installed and notifies the technician that over-temperature fault can occur due to incorrect installment.

Abstract: A method for managing network wake-up commands by a controller of a server includes receiving a system health data. The method includes receiving a network wake-up command from a network interface controller of the server. The method further includes the determining whether to wake-up the server based on the system health data to yield a determination, and commanding the server to wake-up, when the determination is to wake-up the server.

Abstract: A method for controlling air flow in a server rack includes storing, by a switch, a connection table, the connection table comprising information for rack positions corresponding to switch ports of the switch. The switch obtains address information of at least one management controller corresponding to at least one device connected to the switch ports of the switch. The switch determines a rack position for each of the at least one device, based on the table and the address information. The switch receives from a rack management controller (RMC), a request for the rack position for each of the at least one device. The switch sends to the RMC, the rack position for each of the at least one device. The RMC controls fan speed for the at least one device, based on the rack position for each of the at least one device.

Abstract: Embodiments generally relate to data storage in a computing system. The present technology discloses techniques that that can enable an optimized mechanism to change spinning speed of data storage disk drives. The present technology can use a service controller, e.g. a Baseboard Management Device (BMC), to communicate with a disk controller to change the spinning speed of disk drives. The present technology can improve energy efficiency by efficiently controlling the spinning speed of disk drives. It can also reduce data access latency by promptly spinning up a disk from a spun-down state.