PCIe error self-repairing method, device and apparatus and readable storage medium

Abstract

The invention discloses a PCIe error self-repairing method, which comprises the following steps of: when a PCIe link in a system runs, monitoring the running state of the system by acquiring CE reporting times and UCE reporting times in the PCIe link, if the CE reporting times reach a corresponding reporting threshold value, or the UCE reporting times reach the corresponding reporting threshold value, removing error reporting equipment from the system. In this way, the error reporting equipment can be used to avoid the adverse effect of the error reporting equipment on the continuous operation of the system, a SI parameter register of the error reporting equipment is modified according to pre-stored adjustable parameters of all PCIe devices in a server, the SI parameters of the error reporting equipment are automatically optimized, the system is re-accessed after the PCIe error is repaired, and the error self-repairing is realized. Therefore, the implementation cost caused by the participation of operation and maintenance personnel and server customer service personnel in equipment replacement is reduced. The invention further discloses a PCIe error self-repairing device and apparatus and a readable storage medium, which have corresponding technical effects.

Description

technical field [0001] The present invention relates to the technical field of equipment operation and maintenance, in particular to a PCIe fault self-recovery method, device, equipment and readable storage medium. Background technique [0002] PCIe (peripheral component interconnect express, a high-speed serial computer expansion bus standard) device is an indispensable part of the server. The performance, calculation, and functions of the server are all related to the PCIe device, and the PCIe device also involves the calculation of the server. (such as GPU, FPGA), storage (such as SAS HBA, NVME SSD), network (NIC), etc., play an important role. [0003] During use, due to the long-term operation and aging of the device, and the joint influence of multiple devices in the complex PCIe link, PCIe errors may occur. At present, after the system is running, the server system will have an automatic error reporting mechanism. When a PCIe error occurs in the system, some minor pr...

AI Technical Summary

Problems solved by technology

[0003] During use, with the long-term operation of the device and the aging of the device, as well as the joint influence of multiple devices in the complex PCIe link, PCIe errors may occur

At present, after the system is running, the server system will have an automatic error reporting mechanism. When a PCIe error occurs in the system, some minor problems can be automatically repaired, but most problems cannot be automatically repaired by the system, which will cause the system to be blocked, and even cause the machine to shut down or restart. , causing the equipment to fail to operate normally; moreover, most of the PCIe error reports are manually performed by the operation and maintenance personnel for fault judgment and troubleshooting, resulting in the consumption of manpower by the operation and maintenance personnel and the increase in the cost of equipment replacement.

Images