METHOD FOR CONTROLLING CORRECTABLE ERROR REPORTING FUNCTION FOR SERVER DEVICE

• US 20190310908A1
• Filed: 04/04/2019
• Published: 10/10/2019
• Est. Priority Date: 04/10/2018
• Status: Active Grant

• Current Assignee: MiTAC Computing Technology Corporation (MiTAC Holdings Corp.)
• Original Assignee: MiTAC Computing Technology Corporation (MiTAC Holdings Corp.)
• Inventors: HUANG, Chia-Jen
• Granted Patent: US 10,970,151 B2
• Time in Patent Office: Days
• CPC Class Codes: G06F 11/0709 in a distributed system consisting of a plurality of standalone computer nodes, e.g. clusters, client-server systems
• CPC Class Codes: G06F 11/0751 Error or fault detection not based on redundancy power supply failures G06F1/30; network fault management H04L41/06
• CPC Class Codes: G06F 11/0772 Means for error signaling, e.g. using interrupts, exception flags, dedicated error registers
• CPC Class Codes: G06F 11/0775 Content or structure details of the error report, e.g. specific table structure, specific error fields
• CPC Class Codes: G06F 11/0793 Remedial or corrective actions recovery from an exception in an instruction pipeline G06F9/3861; by retry G06F11/1402; for recovering from a failure of a protocol instance or entity H04L69/40
• CPC Class Codes: G06F 11/2284 by power-on test, e.g. power-on self test [POST]

Abstract

A method for controlling a correctable error reporting function and applicable to a server device is provided, including: receiving, by control unit, a plurality of first error messages sent by a first hardware component in which a plurality of correctable errors occurs in a plurality of hardware components; determining, by the control unit, according to the first error messages, error types of the errors occurring in the first hardware component; determining, by the control unit, whether the number of occurrences of the errors of the error types that occur in the first hardware component within first preset duration reaches a preset number of times; and if the determining result is yes, controlling, by the control unit, the first hardware component to stop performing an error reporting function corresponding to the first error type.

Claims

1. A method for controlling a correctable error reporting function and applicable to a server device, comprises:

receiving, by a control unit, a plurality of first error messages sent by a first hardware component in which a plurality of correctable errors occurs in a plurality of hardware components;

determining, by the control unit, according to the first error messages, error types of the errors occurring in the first hardware component;

determining, by the control unit, whether the number of occurrences of the errors of the error types that occur in the first hardware component within first preset duration reaches a preset number of times; and

when the control unit determines that the number of occurrences of the errors of a first error type that occur in the first hardware component within the first preset duration reaches the preset number of times, controlling, by the control unit, the first hardware component to stop performing an error reporting function corresponding to the first error type after the first hardware component sends the first error messages.

2. The method for controlling a correctable error reporting function for a server device according to claim 1, wherein the hardware components are PCIe interface cards, and the control unit determines the number of occurrences of the errors of the first error type that occur in the first hardware component within the first preset duration reaches the preset number of times in the hardware components by scanning a plurality of root ports coupled to the hardware components.

3. The method for controlling a correctable error reporting function and applicable to a server device according to claim 2, wherein the control unit controls the first hardware component to stop performing the error reporting function corresponding to the first error type, in the hardware components through the root ports coupled to the hardware components.

4. The method for controlling a correctable error reporting function and applicable to a server device according to claim 1, wherein the hardware components are memory units, each of the hardware components comprises a plurality of memory channels, each of the memory channels is provided with at least one dual in-line memory module, and the control unit further scans each of the dual in-line memory modules of each of the memory channels, to determine, in the first hardware component, a dual in-line memory module whose number of occurrences of errors of the first error type within the first preset duration reaches the preset number of times.

5. The method for controlling a correctable error reporting function and applicable to a server device according to claim 4, wherein the control unit further controls a dual in-line memory module comprised in the first hardware component to stop performing the error reporting function corresponding to the first error type.

6. The method for controlling a correctable error reporting function and applicable to a server device according to claim 1, wherein after the step of controlling, by the control unit, the first hardware component to stop performing an error reporting function corresponding to the first error type, the method further comprises:

calculating, by the control unit, a first disabled duration that the first hardware component stops performing the error reporting function corresponding to the first error type;

determining, by the control unit, whether the first disabled duration reaches a second preset duration; and

when the first disabled duration reaches the second preset duration, determining, by the control unit, whether the first hardware component needs to be controlled to re-enable the error reporting function corresponding to the first error type.

7. The method for controlling a correctable error reporting function and applicable to a server device according to claim 6, wherein the step of determining, by the control unit, whether the first hardware component needs to be controlled to re-enable the error reporting function corresponding to the first error type comprises:

determining, by the control unit, whether another error corresponding to the first error type occurs in the first hardware component within the first disabled duration of the error reporting function.

8. The method for controlling a correctable error reporting function and applicable to a server device according to claim 7, wherein, when the control unit determines that another error corresponding to the first error type does not occur within the first disabled duration, the control unit controls the first hardware component to re-enable the error reporting function corresponding to the first error type.

9. The method for controlling a correctable error reporting function and applicable to a server device according to claim 7, further comprising:

when the control unit determines that another error corresponding to the first error type occurs in the first hardware component within the first disabled duration of the error reporting function, recalculating, by the control unit, a second disabled duration for which the error reporting function is disabled once the first disabled duration reaches the second preset duration; and

determining, by the control unit, whether the second disabled duration reaches the second preset duration, to determine whether the first hardware component needs to be controlled to re-enable the error reporting function corresponding to the first error type.

10. The method for controlling a correctable error reporting function and applicable to a server device according to claim 1, further comprising:

after a first power-on self-test (POST) procedure being executed by the server device, when the control unit runs an operating system and determines, that the number of occurrences of the errors of the first error type that occur in the first hardware component within the first preset duration reaches the preset number of times, prolonging, by the control unit, the first preset duration when the server device executes a second POST procedure.

11. The method for controlling a correctable error reporting function and applicable to a server device according to claim 1, wherein in the step of determining, by the control unit, whether the number of occurrences of the errors of the error types that occur in the first hardware component within first preset duration reaches a preset number of times, the control unit performs, within the first preset duration, debug operation on the errors occurring in the first hardware component.

12. The method for controlling a correctable error reporting function and applicable to a server device according to claim 10, wherein in the step of prolonging, by the control unit, the first preset duration, the control unit prolongs the first preset duration in the second POST procedure based on a preset multiple.

13. The method for controlling a correctable error reporting function and applicable to a server device according to claim 1, wherein the control unit receives the first error messages from the first hardware component when running an operating system between a first POST procedure and a second POST procedure executed sequentially by the control unit at different time points, and the control unit determines, when running the operating system between the first POST procedure and the second POST procedure, whether the number of occurrences of the errors of the error types that occur in the first hardware component within the first preset duration reaches the preset number of times.

14. The method for controlling a correctable error reporting function and applicable to a server device according to claim 13, further comprising:

when the errors occur in the first hardware component, setting, by the control unit, a flag signal to a first logic level according to the first error messages when running the operating system; and

when the control unit determines, when running the operating system, that the number of occurrences of the errors occurring in the first hardware component within the first preset duration reaches the preset number of times, corresponding handling the correctable errors by the control unit when running the operating system, and resetting the flag signal to a second logic level other than the first logic level after the correctable errors are handled by the control unit.

15. The method for controlling a correctable error reporting function and applicable to a server device according to claim 14, further comprising:

determining, by the control unit when the second POST procedure is executed, whether the flag signal is the first logic level, to determine whether the correctable errors have been handled correspondingly when running the operating system; and

when the control unit determines, in the second POST procedure, that the flag signal is the first logic level, prolonging, by the control unit, the first preset duration based on a preset multiple in the second POST procedure, and resetting the flag signal to the second logic level.

16. The method for controlling a correctable error reporting function and applicable to a server device according to claim 15, wherein the flag signal is stored in a non-volatile memory.

17. The method for controlling a correctable error reporting function and applicable to a server device according to claim 1, wherein the preset number of times is three times.

18. The method for controlling a correctable error reporting function and applicable to a server device according to claim 1, wherein the first preset duration is one hour.

19. The method for controlling a correctable error reporting function and applicable to a server device according to claim 12, wherein the preset multiple is twofold.