CN112256539B - PCIE link error statistical method, device, terminal and storage medium - Google Patents

Abstract

The invention discloses a PCIE link error statistics method, device, terminal and storage medium, which can monitor the non-fatal error count of the PCIE link in real time; If it exceeds the threshold of the first number of times, an alarm is issued and/or the PCIE link is interrupted; when it is detected that the non-fatal error count changes in N segments within the second preset time period, and N exceeds the threshold of the second number of segments, and each time it changes continuously, If the number of changes of the non-fatal error count does not exceed the first number of times threshold, an alarm is issued and/or the PCIE link is interrupted; the change within the first preset time period is a one-stage change. The present invention performs statistics in two dimensions: the number of errors and the time when errors are generated. When the generated errors meet the statistical conditions, an alarm will be generated or the link will be interrupted, so as to avoid serious system failures caused by excessive errors and greatly improve the stability of system operation. sturdiness and reliability.

Description

Method, device, terminal and storage medium for PCIE link error statistics

technical field

The invention relates to the field of PCIE link monitoring, in particular to a PCIE link error statistics method, device, terminal and storage medium.

Background technique

In recent years, with the continuous improvement of users' requirements for integration, unification, efficiency, space, and energy consumption, PCIE (peripheral component interconnect express, high-speed serial computer expansion bus) equipment has been widely used in the server and storage fields, so it can effectively monitor The health status of the PCIE link and the adoption of security protection strategies based on the monitoring situation are more and more important to improve the stability and reliability of system operation. At present, most of the various PCIE devices provide error data. How to effectively use the data to judge the health status of the link has always been a difficulty in this field, and there is currently no effective error statistics method.

SUMMARY OF THE INVENTION

In order to solve the above problems, the present invention provides a PCIE link error statistics method, device, terminal and storage medium, which can reasonably count the non-fatal errors on the PCIE link, so as to avoid serious system failure caused by excessive errors.

The technical scheme of the present invention is: a PCIE link error statistics method, comprising the following steps:

Real-time monitoring of non-fatal error counts of PCIE links;

When it is detected that the non-fatal error count continuously changes within the first preset time period, and the number of changes exceeds the first number of times threshold, an alarm is issued and/or the PCIE link is interrupted;

When it is detected that the non-fatal error count changes in N segments within the second preset time period, and N exceeds the second threshold of the number of segments, and for each continuous change, the number of non-fatal error count changes does not exceed the first threshold, an alarm and /or interrupt the PCIE link; the change within the first preset time period is a period of change.

Further, the following steps are also included:

Apply for several object pools; the number of object pools is the same as the threshold of the second segment;

When a change in the non-fatal error count is detected, the monitoring information is recorded in the corresponding object pool;

If the non-fatal error count changes continuously within the first preset time period, the monitoring information is continuously updated in the current object pool;

If the time interval between the next non-fatal error count change and the last non-fatal error count change is greater than the first preset duration, move to the next object pool to record monitoring information, so as to cover and use each object pool cyclically according to the order of the object pool. object pool;

If all the object pools are covered and used within the second preset time period, it means that when N segments of non-fatal error counts are detected within the second preset duration, N exceeds the second segment count threshold, and each time it changes continuously, a non-fatal error occurs. If the number of count changes does not exceed the first threshold, an alarm is issued and/or the PCIE link is interrupted.

Further, the recorded monitoring information includes: the time when the non-fatal error count change was last monitored, the latest value of the non-fatal error count, and the number of times the non-fatal error count changed in this segment.

Further, the non-fatal error count includes the data link layer packet error count and the transport layer packet error count.

The technical solution of the present invention also includes a PCIE link error statistics device, including,

Count monitoring module: monitor the non-fatal error count of PCIE link in real time;

The first exception handling module: when it is detected that the non-fatal error count continuously changes within the first preset time period, and the number of changes exceeds the first number of times threshold, an alarm is issued and/or the PCIE link is interrupted;

The second exception processing module: when it is detected that the non-fatal error count changes in N segments within the second preset time period, and N exceeds the second segment number threshold, and each continuous change, the number of non-fatal error count changes does not exceed the first number of times If the threshold is exceeded, an alarm is issued and/or the PCIE link is interrupted; the change within the first preset time period is a change in one segment.

Further, it also includes,

Object pool application module: apply for several object pools; the number of object pools is the same as the threshold for the number of second segments;

Monitoring information recording module: when the non-fatal error count changes are detected, the monitoring information is recorded in the corresponding object pool; if the non-fatal error count changes continuously within the first preset time period, the monitoring information is continuously updated in the current object pool; If the time interval between the next non-fatal error count change and the last non-fatal error count change is greater than the first preset duration, move to the next object pool to record monitoring information, so as to cover and use each object cyclically according to the order of the object pool pool;

The second exception processing module monitors whether all the object pools are covered and used within the second preset time period, and if so, it means that N segments of non-fatal error counts have been changed within the second preset time period, and N exceeds the second segment count threshold, and each If the number of non-fatal error count changes does not exceed the threshold for the first number of consecutive changes, an alarm is issued and/or the PCIE link is interrupted.

Further, the recorded monitoring information includes: the time when the non-fatal error count change was last monitored, the latest value of the non-fatal error count, and the number of times the non-fatal error count changed in this segment.

Further, the non-fatal error count includes the data link layer packet error count and the transport layer packet error count.

The technical solution of the present invention also includes a terminal, including:

processor;

memory for storing instructions for execution of the processor;

Wherein, the processor is configured to perform any of the methods described above.

The technical solution of the present invention also includes a computer-readable storage medium storing a computer program, which implements the method described in any one of the above when the program is executed by a processor.

A method, device, terminal and storage medium for PCIE link error statistics provided by the present invention perform effective and reasonable statistics on non-fatal errors occurring on the link, and conduct statistics in the two dimensions of error quantity and error generation time. When the number of errors meets the statistical conditions, an alarm will be generated or the link will be interrupted to avoid serious system failures caused by excessive errors and greatly improve the stability and reliability of system operation. The method fills the gap in PCIE link error statistics, does not distinguish PCIE devices, and has wider applicability and strong generality.

Description of drawings

1 is a schematic flowchart of a method according to a specific embodiment of the present invention;

2 is a schematic block diagram of a specific embodiment of the present invention—a specific implementation mode object pool architecture;

FIG. 3 is a schematic block diagram of the structure of the second embodiment of the present invention.

Detailed ways

The present invention will be described in detail below with reference to the accompanying drawings and specific examples. The following examples are to explain the present invention, but the present invention is not limited to the following embodiments.

As shown in FIG. 1, this embodiment provides a PCIE link error statistics method, including the following steps:

S1, monitor the non-fatal error count of the PCIE link in real time;

S2, when it is detected that the non-fatal error count continuously changes within the first preset time period, and the number of changes exceeds the first number of times threshold, an alarm is issued and/or the PCIE link is interrupted;

S3, when it is detected that the non-fatal error count changes in N segments within the second preset time period, and N exceeds the second segment number threshold, and each continuous change, the number of non-fatal error count changes does not exceed the first number of thresholds, then send out Alarm and/or interrupt the PCIE link; the change within the first preset time period is a period of change.

This method sets the first preset duration, the first number of times threshold, the second preset duration and the second segment number threshold. When the count (referring to the non-fatal error count) changes, if it changes continuously within the first preset duration, and the continuous If the number of changes exceeds the threshold of the first number of times, it means that an exception occurs and needs to be dealt with; in addition, if there is a segment change that exceeds the second threshold of the number of segments within the second preset time period, it also means that an exception occurs and needs to be handled. It should be noted that each segment of change may have only one change, or may include multiple consecutive changes. If the changes are continuous, the number of non-fatal error count changes does not exceed the first threshold, otherwise it will not reach the second preset duration. alarm and/or interrupt the PCIE link.

Through the above method, reasonable and effective statistics are carried out on the non-fatal errors (Uncorrectable Errors) that occur on the link, and statistics are made in the two dimensions of the number of errors and the time of error generation. When the generated errors meet the statistical conditions, an alarm will be generated or the This link is interrupted to avoid excessive errors leading to serious system failure.

In this embodiment, the non-fatal error count monitoring information is saved and counted through the object pool. First, several object pools are applied for, and the number of object pools is the same as the second segment number threshold, so as to count the number of count change segments within the second preset time period. The object pools are sorted by serial number and used cyclically. For example, the monitoring information of the first stage of monitoring is stored in the first object pool, and the monitoring information of the second stage of monitoring is stored in the second object pool. If there are M object pools, monitoring If no abnormality occurs after the count of the M segment is changed, the monitoring information of the M+1 segment is saved in the first object pool, and the previous information in the pool is overwritten.

When a change in the count of non-fatal errors is detected, the monitoring information is recorded in the corresponding object pool. The monitoring information includes: the time when the non-fatal error count was changed most recently, the latest value of the non-fatal error count, and the number of changes of the non-fatal error count in this segment.

If the non-fatal error count changes continuously within the first preset time period, the monitoring information will be continuously updated in the current object pool; if the time interval between the next non-fatal error count change and the last non-fatal error count change is greater than the first If the preset duration is set, it will move to the next object pool to record monitoring information, so as to cover and use each object pool cyclically according to the order of the object pools.

Based on this, all the object pools are covered and used within the second preset duration, which means that when N segments of non-fatal error counts are detected within the second preset duration, N exceeds the threshold of the second segment count, and changes continuously each time, If none of the non-fatal error count changes exceeds the first threshold, an alarm is issued and/or the PCIE link is interrupted.

A specific implementation manner is provided below to further understand the solution.

As shown in Figure 2, in this specific implementation, there are 51 object pools in total, the first preset duration is set to 20 seconds, the second preset duration is set to 1 hour, the first threshold is 4, and the second threshold is 4 51 times.

In addition, the non-fatal error count includes a data link layer packet error count (Bad Data Link LayerPacket Count, referred to as Bad DLLP Count) and a transport layer packet error count (Bad Transaction LayerPacket Count, referred to as Bad TLP Count).

The system applies for a total of 51 statistical object pools from 0 to 50, and 51 object pools are used cyclically and are used from scratch; each object pool has a description of the error, and the description includes: Fatal error count change time), Bad TLP Count, Bad DLLP Count, number of statistical changes (that is, the number of non-fatal error count changes in this segment).

The statistical process is as follows:

(1) The system cyclically reads the non-fatal error count values (BadTLPCount and Bad DLLPCount) on the PCIE link. When either of the two changes (that is, the count changes), statistics are performed;

(2) If the continuous count changes within 20s, it will be updated in the current object pool;

(3) When the time between the current count change and the last count change is greater than 20s, move to the next object pool for statistics.

When the count changes for 4 consecutive times within 20s, or 51 object pools are used within 1 hour, it means that an abnormality occurs, and an alarm can be issued and/or the PCIE link can be interrupted.

It should be noted that statistics are not performed during the normal link phase, during device insertion and removal, and during link changes. In addition, if all 51 object pools are used up within 1 hour, for the convenience of statistics, the cyclic coverage can be stopped, that is, the monitoring information can be stopped, and an alarm and/or the PCIE link can be interrupted.

Embodiment 2

As shown in FIG. 3 , based on the first embodiment, this embodiment provides a PCIE link error statistics device, which includes the following functional modules.

Count monitoring module 101: monitor the non-fatal error count of the PCIE link in real time;

The first exception processing module 102: when monitoring that the non-fatal error count continuously changes within the first preset time period, and the number of changes exceeds the first number of times threshold, issue an alarm and/or interrupt the PCIE link;

The second exception processing module 103 : when it is detected that the non-fatal error count has changed in N segments within the second preset time period, and N exceeds the second threshold of the number of segments, and each continuous change, the number of non-fatal error count changes does not exceed the first time If the number of thresholds is exceeded, an alarm is issued and/or the PCIE link is interrupted; the change within the first preset time period is a change in one segment;

Object pool application module 104: apply for several object pools; the number of object pools is the same as the threshold for the number of second segments;

Monitoring information recording module 105: when the non-fatal error count changes are detected, the monitoring information is recorded in the corresponding object pool; if the non-fatal error count changes continuously within the first preset time period, the monitoring information is continuously updated in the current object pool; If the time interval between the next non-fatal error count change and the last non-fatal error count change is greater than the first preset duration, move to the next object pool to record monitoring information, so as to cover and use each object pool cyclically according to the order of the object pool. object pool.

The second exception processing module monitors whether all the object pools are covered and used within the second preset time period, and if so, it means that N segments of non-fatal error counts have been changed within the second preset time period, and N exceeds the second segment count threshold, and each If the number of non-fatal error count changes does not exceed the threshold for the first number of consecutive changes, an alarm is issued and/or the PCIE link is interrupted.

The recorded monitoring information includes: the time when the non-fatal error count was changed in the last monitoring, the latest value of the non-fatal error count, and the number of times the non-fatal error count was changed in this segment. Non-fatal error counts include data link layer packet error counts and transport layer packet error counts.

Embodiment 3

This embodiment provides a terminal, where the terminal includes a processor and a memory.

Memory is used to store instructions for execution by the processor. Memory can be implemented by any type of volatile or non-volatile storage terminals or their combination such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Read Only Memory (PROM), Read Only Memory (ROM), Magnetic Memory, Flash Memory, Magnetic Disk or Optical Disk. When the execution instructions in the memory are executed by the processor, the terminal is enabled to execute some or all of the steps in the foregoing method embodiments.

The processor is the control center of the storage terminal, using various interfaces and lines to connect various parts of the entire electronic terminal, by running or executing the software programs and/or modules stored in the memory, and calling the data stored in the memory to execute. Various functions of the electronic terminal and/or processing data. The processor may be composed of an integrated circuit (Integrated Circuit, IC for short), for example, may be composed of a single packaged IC, or may be composed of a plurality of packaged ICs connected with the same function or different functions.

Embodiment 4

This embodiment provides a computer storage medium, wherein the computer storage medium can store a program, and when the program is executed, it can include some or all of the steps in the various embodiments provided by the present invention. The storage medium may be a magnetic disk, an optical disc, a read-only memory (English: read-only memory, ROM for short) or a random access memory (English: random access memory, RAM for short).

The above disclosure is only the preferred embodiment of the present invention, but the present invention is not limited thereto, any non-creative changes that can be conceived by those skilled in the art, and some improvements and modifications made without departing from the principles of the present invention , should fall within the protection scope of the present invention.

Claims (10)

1. a PCIE link error statistics method, is characterized in that, comprises the following steps: Real-time monitoring of non-fatal error counts of PCIE links; When it is detected that the non-fatal error count continuously changes within the first preset time period, and the number of changes exceeds the first number of times threshold, an alarm is issued and/or the PCIE link is interrupted; When it is detected that the non-fatal error count changes in N segments within the second preset time period, and N exceeds the second threshold of the number of segments, and for each continuous change, the number of non-fatal error count changes does not exceed the first threshold, an alarm and /or interrupt the PCIE link; the change within the first preset time period is a period of change. 2. PCIE link error statistics method according to claim 1, is characterized in that, also comprises the following steps: Apply for several object pools; the number of object pools is the same as the threshold of the second segment; When a change in the non-fatal error count is detected, the monitoring information is recorded in the corresponding object pool; If the non-fatal error count changes continuously within the first preset time period, the monitoring information is continuously updated in the current object pool; If the time interval between the next non-fatal error count change and the last non-fatal error count change is greater than the first preset duration, move to the next object pool to record monitoring information, so as to cover and use each object pool cyclically according to the order of the object pool. object pool; If all the object pools are covered and used within the second preset time period, it means that when N segments of non-fatal error counts are detected within the second preset duration, N exceeds the second segment count threshold, and each time it changes continuously, a non-fatal error occurs. If the number of count changes does not exceed the first threshold, an alarm is issued and/or the PCIE link is interrupted. 3. PCIE link error statistics method according to claim 2, it is characterised in that the recorded monitoring information comprises: the most recent monitoring to the time that the non-fatal error count changed, the latest value of the non-fatal error count, the non-fatal error of this segment Count the number of changes. 4. The PCIE link error statistics method according to claim 1, 2 or 3, wherein the non-fatal error count includes a data link layer packet error count and a transport layer packet error count. 5. a PCIE link error statistics device, is characterized in that, comprises, Count monitoring module: monitor the non-fatal error count of PCIE link in real time; The first exception handling module: when it is detected that the non-fatal error count continuously changes within the first preset time period, and the number of changes exceeds the first number of times threshold, an alarm is issued and/or the PCIE link is interrupted; The second exception processing module: when it is detected that the non-fatal error count changes in N segments within the second preset time period, and N exceeds the second segment number threshold, and each continuous change, the number of non-fatal error count changes does not exceed the first number of times If the threshold is exceeded, an alarm is issued and/or the PCIE link is interrupted; the change within the first preset time period is a change in one segment. 6. PCIE link error statistics device according to claim 5, is characterized in that, also comprises, Object pool application module: apply for several object pools; the number of object pools is the same as the threshold for the number of second segments; Monitoring information recording module: when the non-fatal error count changes are detected, the monitoring information is recorded in the corresponding object pool; if the non-fatal error count changes continuously within the first preset time period, the monitoring information is continuously updated in the current object pool; If the time interval between the next non-fatal error count change and the last non-fatal error count change is greater than the first preset duration, move to the next object pool to record monitoring information, so as to cover and use each object cyclically according to the order of the object pool pool; The second exception processing module monitors whether all the object pools are covered and used within the second preset time period, and if so, it means that N segments of non-fatal error counts have been changed within the second preset time period, and N exceeds the second segment count threshold, and each If the number of non-fatal error count changes does not exceed the threshold for the first number of consecutive changes, an alarm is issued and/or the PCIE link is interrupted. 7. The PCIE link error statistics device according to claim 6, wherein the recorded monitoring information comprises: the time when the non-fatal error count is changed in the last monitoring, the latest value of the non-fatal error count, the non-fatal error of this segment Count the number of changes. 8. The PCIE link error statistics device according to claim 5, 6 or 7, wherein the non-fatal error count includes a data link layer packet error count and a transport layer packet error count. 9. A terminal, characterized in that, comprising: processor; memory for storing instructions for execution of the processor; wherein the processor is configured to perform the method of any one of claims 1-4. 10. A computer-readable storage medium storing a computer program, characterized in that, when the program is executed by a processor, the method according to any one of claims 1-4 is implemented.

Images