Automatic BMC (baseboard management controller) fault solution method based on RMC (rack server management center) management

Abstract

The invention discloses an automatic BMC (baseboard management controller) fault solution method based on RMC (rack server management center) management, and belongs to the field of computer performance management. The method comprises the following specific steps: (1) monitoring a running state of a BMC by an RMC, if monitoring a BMC fault, re-authenticating whether the fault occurs in the BMC state or not, and if confirming that the fault occurs in the BMC, actively finishing a BMC restarting action by the RMC, wherein a server system comprises the RMC, an IPMB (intelligent platform management bus), an intermediate node plate and the BMC; (2) after actively finishing BMC restarting by the RMC, if monitoring that the BMC runs normally by the RMC, continuously keeping a monitoring state, or otherwise, notifying the BMC fault to users and prompting that the BMC fault is solved through a hardware restarting mode by a notification system. The method is used for an RACK server, automatic judgment of a BMC fault situation and fault removal and recovery by the system are realized through the RMC, the manual operation is not required, and the demands of BMC fault judgment, removal and recovery required to be performed by the server in time are effectively met.

Description

technical field [0001] The invention discloses a method for solving BMC faults, which belongs to the field of computer performance management, and in particular relates to an automatic BMC fault solving method based on RMC management. Background technique [0002] The management board RMC is the management center of the large-scale server Smart Rack, responsible for node management, power management, and fan management in the system. The management system consists of RMC, which implements system monitoring, management, and alarm; node middle board, which implements out-of-band monitoring, secondary management, alarm reporting, and fan monitoring of all nodes; node BMC, which is integrated on each node and is responsible for sub- System-level status real-time monitoring, fault diagnosis, power consumption detection, etc.; and management network and other components. Among them, BMC is responsible for node asset management, real-time monitoring, and fault diagnosis. In view ...

AI Technical Summary

Problems solved by technology

[0004] In the present invention, when BMC breaks down, it is necessary to carry out some manual measures to troubleshoot and restore in time to ensure the continuous operation of the server, but this consumes time and manpower, and the single-particle BIOS and BMC of the server are designed when the BIOS and BMC are upgraded and maintained. , there is also the problem that the service needs to be stopped before upgrading. An automatic BMC fault solution method based on RMC management is provided, which is suitable for RACK servers. Through RMC, the system can automatically determine the BMC fault situation and troubleshoot and restore the method without manual work. Effectively meet the needs of the server to perform BMC fault judgment, troubleshooting and recovery in a timely manner

Images