Communication automatic recovery method and device

Abstract

The invention discloses a communication automatic recovery method and device. According to the method, after the whole cabinet management unit monitors that the communication state between the whole cabinet management unit and the substrate management controller is in fault, the substrate management controller can be controlled to be restarted, and therefore communication between the substrate management controller and the middle board and communication between the substrate management controller and the whole cabinet management unit are recovered. Therefore, the whole cabinet management unitcan acquire the real-time operation state of the server node sent by the baseboard management controller in real time; the system risk caused by the fact that the RMC cannot obtain the real-time operation state of the server node, heat dissipation speed regulation abnormity, power consumption monitoring abnormity and the like of the whole cabinet fan are possibly caused is avoided, and therefore monitoring and operation of the server cabinet become more stable.

Description

technical field [0001] The present application relates to the field of computer technology, in particular to a method and device for automatic communication restoration. Background technique [0002] With the development of new technologies such as cloud computing and big data, the demand for cabinet servers is increasing, and subsequently, the demand for heat dissipation stability of the whole cabinet server is getting higher and higher. [0003] In order to meet these requirements, the nodes integrated on the whole cabinet are becoming denser and more and more types of nodes (including computing nodes, storage nodes and GPUBOX nodes, etc.) The communication stability between RMC) and server nodes is getting higher and higher. This is because, when the communication fails, the RMC cannot normally obtain the real-time running information of the nodes monitored by the Baseboard Management Controller (BMC) and sent to the RMC. The speed, power consumption, and power supply u...

AI Technical Summary

Problems solved by technology

This is because, when the communication fails, the RMC cannot normally obtain the real-time running information of the nodes monitored by the Baseboard Management Controller (BMC) and sent to the RMC. The speed, power consumption, and power supply unit (PSU), etc. are adjusted by the RMC by obtaining the real-time temperature, voltage, power consumption, etc. information of each node. If the communication fails, it will cause risks to the operation of the entire cabinet and possibility of instability

Images