Method and system for processing system temperature sensor fault of server

Abstract

The invention discloses a method and system for processing a system temperature sensor fault of a server. The method comprises the following steps that firstly, a monitoring management unit is used for polling a temperature sensor through an I2C bus to obtain instant system temperature information, and controlling the output duty ratio of a fan controller according to the current system temperature conditions; secondly, when the temperature sensor works normally, the monitoring management unit outputs high-level GATE signals, a system fan is controlled by PWM signals, and when the temperature sensor breaks down, the monitoring management unit immediately reduces GATE signals, revolving speed control signals of the fan are directly increased into the direct voltage VCC, and the fan rotates at a full speed; thirdly, the monitoring management unit obtains correct system temperature information again in a polling mode and releases the GATE signals, the system fan is controlled by the PWM signals, and normal work is restored. By means of the method and system, the reliability and stability of the server can be guaranteed, and the stabile operation of a calculation system can be guaranteed.

Description

technical field [0001] The invention belongs to the technical field of computer servers, and relates to a method and system for dealing with failures of system temperature sensors of servers. Background technique [0002] In the field of computer servers, since the motherboard size is usually larger than that of a general computer, and its structure is also more complex than that of a general computer; the monitoring and management unit of the motherboard may be far away from each temperature sensor, which will increase The loss of the I2C bus, or too many sensors mounted on the same I2C bus, reduces the driving capability of the I2C bus. [0003] In the actual PCB layout design, the same I2C bus will form a bifurcation, resulting in increased signal delay. Based on the above, it can be seen that in the server, the I2C signal quality between the monitoring management unit and the temperature sensor is not easy to be guaranteed, and there will be a certain probability of err...

AI Technical Summary

Problems solved by technology

[0002] In the field of computer servers, since the motherboard size is usually larger than that of a general computer, and its structure is also more complex than that of a general computer; the monitoring and management unit of the motherboard may be far away from each temperature sensor, which will increase The loss of the I2C bus, or too many sensors mounted on the same I2C bus, reduces the driving capability of the I2C bus

[0003] In the actual PCB layout design, the same I2C bus will form a bifurcation, resulting in increased signal delay

Based on the above, it can be seen that in the server, the I2C signal quality between the monitoring management unit and the temperature sensor is not easy to be guaranteed, and there will be a certain probability of error codes, so that the temperature sensor is configured in an abnormal state, and the monitoring management unit cannot be correct. Obtaining system temperature information, the fan system is out of control, seriously affecting the reliability and stability of the server

Images