Node fault recording method of high-end fault tolerance server

Abstract

The invention discloses a node fault recording method of a high-end fault tolerance server. The specific design process comprises the steps of: firstly, defining a fault management LOG level of an RMC aiming at internal nodes of a rack, wherein the RMC is a Racks Management Controller, and LOG is a log file; secondly, defining fault management LOG content of the RMC aiming at the internal nodes of the rack; and finally defining fault management LOG triggering conditions of the RMC aiming at the internal nodes of the rack. Compared with the prior art, node fault LOG recording and fault management of the RMC aiming at the rack level can effectively finished, a user is enabled to conveniently manage fault information of the high-end fault tolerance server as managing a single server, the practicality is high, and the popularization is easy.

Description

technical field [0001] The invention relates to the technical field of computer servers, in particular to a highly practical and high-end fault-tolerant server node fault recording method. Background technique [0002] As users' computing demands on computers increase, users have higher and higher requirements on computing performance of a single computer. The high-end fault-tolerant server is a multi-socket server. Compared with traditional servers, it has great advantages in terms of computing performance and reliability, and is more and more widely used in fields that require real-time performance, reliability and availability. At the same time, the high-end fault-tolerant server integrates multiple computing nodes in the cabinet. The entire cabinet performs node fault management and fault logging through RMC. RMC manages a large number of nodes. Compared with traditional servers, high-end fault-tolerant servers pose new challenges to node fault management and fault log...

AI Technical Summary

Problems solved by technology

The entire cabinet performs node fault management and fault logging through RMC. There are a large number of nodes managed by RMC. Compared with traditional servers, high-end fault-tolerant servers pose new challenges to node fault management and fault logging.

[0003] At present, in the fault management and fault record content of high-end fault-tolerant servers, the specific content is not clear. If it is carried out under the traditional server, it is difficult to meet the management requirements due to the large number of nodes