Method and system for upgrading CPE firmware

1 . A fault diagnosis method, performed by a monitoring platform, comprising:

obtaining log files of a redundant array of independent disks (RAID) card in a server to be monitored;

comparing each log file of the RAID card with a fault diagnosis rule base, and determining whether an object file matching a current log file of the RAID card exists in the fault diagnosis rule base, wherein the fault diagnosis rule base is established by analyzing historical faults of the RAID card and the hard disk managed by the RAID card in advance; matching each diagnosis keyword corresponding to the object file with each line of content in the current log file under the condition that the object file exists; determining that the RAID card and/or the hard disk managed by the RAID card has a fault under the condition that at least one line of content matching each diagnosis keyword exists in the current log file; and

determining whether a fault diagnosis method corresponding to the object file exists in the fault diagnosis rule base, in a case that the fault diagnosis method corresponding to the object file exists, extracting a last keyword corresponding to the object file from the fault diagnosis rule base, and extracting a numerical value corresponding to the last keyword from the current log file;

filtering the content of a line where the numerical value is located from the current log file under the condition that the numerical value is equal to 0;

extracting a serial number from the current log file under the condition that the numerical value is great than 0, circularly traversing the current log file according to a fourth preset regular expression, searching, under the condition that a line of content matching the fourth preset regular expression exists in the current log file, forwards from the line of content matching the fourth preset regular expression for a content matching a fifth preset regular expression, and extracting a third slot number from the content matching the fifth preset regular expression;

providing the fault processing solution that is to replace the hard disk corresponding to the third slot number and the serial number under the condition that the diagnosis keyword in the fault diagnosis rule base comprises any one of hwErrors, mediumErrors, and smartWarning;

obtaining the number of numerical values greater than 0 and corresponding to the last keyword under the condition that the diagnosis keyword in the fault diagnosis rule base does not comprise any one of hwErrors, mediumErrors, and smartWarning;

providing the fault processing solution that is to replace the hard disk corresponding to the third slot number and the serial number under the condition that the number is equal to 1; and

providing the fault processing solution that is to take a link problem into account, and suggest replacing the RAID card, a backplane and an SAS cable one by one in sequence under the condition that the number is greater than 1.

2 . The fault diagnosis method according to claim 1 , further comprising:

extracting state information from the current log file under the condition that no object file matching the current log file of the RAID card exists in the fault diagnosis rule base, or under the condition that an object file matching the current log file of the RAID card exists in the fault diagnosis rule base and no at least one line of content matching each diagnosis keyword in the current log file;

constructing a log training set according to the state information, selecting a sample from the log training set by means of a Relief filter selection algorithm, searching for a near-hit sample from samples of the same class as the sample, randomly selecting a near-miss sample from samples of different classes from the sample, increasing, under the condition that a distance between the sample and the near-hit sample in a feature is less than a distance between the sample and the near-miss sample in the same feature, a weight of the feature, reducing, under the condition that the distance between the sample and the near-hit sample in the feature is not less than the distance between the sample and the near-miss sample in the same feature, the weight of the feature, performing training on the feature preset times, and obtaining an average weight of the feature after the preset times of training;

adding a feature with an average weight greater than a preset value into a feature set, and outputting the feature set; and

receiving an analysis result of the feature set, and adding the analysis result of the feature set to the fault diagnosis rule base.

3 . The fault diagnosis method according to claim 1 , further comprising:

outputting the log file that determines that the RAID card and/or the hard disk managed by the RAID card has a fault, and the fault processing solution to a mobile terminal by means of an email and/or a short message.

4 . A fault diagnosis apparatus, comprising:

a memory, configured to store a computer program; and

a processor, configured to execute the computer program:

obtain log files of a redundant array of independent disks (RAID) card in a server to be monitored;

compare each log file of the RAID card with a fault diagnosis rule base, and determine whether an object file matching a current log file of the RAID card exists in the fault diagnosis rule base, wherein the fault diagnosis rule base is established by analyzing historical faults of the RAID card and the hard disk managed by the RAID card in advance; match each diagnosis keyword corresponding to the object file with each line of content in the current log file under the condition that the object file exists; determine that the RAID card and/or the hard disk managed by the RAID card has a fault under the condition that at least one line of content matching each diagnosis keyword exists in the current log file; and

determine whether a fault diagnosis method corresponding to the object file exists in the fault diagnosis rule base, in a case that the fault diagnosis method corresponding to the object file exists, extract a last keyword corresponding to the object file from the fault diagnosis rule base, and extract a numerical value corresponding to the last keyword from the current log file;

filter the content of a line where the numerical value is located from the current log file under the condition that the numerical value is equal to 0;

extract a serial number from the current log file under the condition that the numerical value is great than 0, circularly traversing the current log file according to a fourth preset regular expression, search, under the condition that a line of content matching the fourth preset regular expression exists in the current log file, forwards from the line of content matching the fourth preset regular expression for a content matching a fifth preset regular expression, and extract a third slot number from the content matching the fifth preset regular expression;

provide the fault processing solution that is to replace the hard disk corresponding to the third slot number and the serial number under the condition that the diagnosis keyword in the fault diagnosis rule base comprises any one of hwErrors, mediumErrors, and smartWarning;

obtain the number of numerical values greater than 0 and corresponding to the last keyword under the condition that the diagnosis keyword in the fault diagnosis rule base does not comprise any one of hwErrors, mediumErrors, and smartWarning;

provide the fault processing solution that is to replace the hard disk corresponding to the third slot number and the serial number under the condition that the number is equal to 1; and

provide the fault processing solution that is to take a link problem into account, and suggest replacing the RAID card, a backplane and an SAS cable one by one in sequence under the condition that the number is greater than 1.

5 . A non-transitory readable storage medium, storing a computer program, wherein the computer program is configured to, when executed by a processor, cause the processor to:

obtain log files of a redundant array of independent disks (RAID) card in a server to be monitored;

compare each log file of the RAID card with a fault diagnosis rule base, and determine whether an object file matching a current log file of the RAID card exists in the fault diagnosis rule base, wherein the fault diagnosis rule base is established by analyzing historical faults of the RAID card and the hard disk managed by the RAID card in advance; match each diagnosis keyword corresponding to the object file with each line of content in the current log file under the condition that the object file exists;

determine that the RAID card and/or the hard disk managed by the RAID card has a fault under the condition that at least one line of content matching each diagnosis keyword exists in the current log file; and

determine whether a fault diagnosis method corresponding to the object file exists in the fault diagnosis rule base, in a case that the fault diagnosis method corresponding to the object file exists, extract a last keyword corresponding to the object file from the fault diagnosis rule base, and extract a numerical value corresponding to the last keyword from the current log file;

filter the content of a line where the numerical value is located from the current log file under the condition that the numerical value is equal to 0;

extract a serial number from the current log file under the condition that the numerical value is great than 0, circularly traversing the current log file according to a fourth preset regular expression, search, under the condition that a line of content matching the fourth preset regular expression exists in the current log file, forwards from the line of content matching the fourth preset regular expression for a content matching a fifth preset regular expression, and extract a third slot number from the content matching the fifth preset regular expression;

provide the fault processing solution that is to replace the hard disk corresponding to the third slot number and the serial number under the condition that the diagnosis keyword in the fault diagnosis rule base comprises any one of hwErrors, mediumErrors, and smartWarning;

obtain the number of numerical values greater than 0 and corresponding to the last keyword under the condition that the diagnosis keyword in the fault diagnosis rule base does not comprise any one of hwErrors, mediumErrors, and smartWarning:

provide the fault processing solution that is to replace the hard disk corresponding to the third slot number and the serial number under the condition that the number is equal to 1; and

provide the fault processing solution that is to take a link problem into account, and suggest replacing the RAID card, a backplane and an SAS cable one by one in sequence under the condition that the number is greater than 1.

6 . The fault diagnosis apparatus according to claim 4 , wherein the processor is further configured to execute the computer program to:

extract state information from the current log file under the condition that no object file matching the current log file of the RAID card exists in the fault diagnosis rule base, or under the condition that an object file matching the current log file of the RAID card exists in the fault diagnosis rule base and no at least one line of content matching each diagnosis keyword in the current log file;

construct a log training set according to the state information, select a sample from the log training set by means of a Relief filter selection algorithm, search for a near-hit sample from samples of the same class as the sample, randomly selecting a near-miss sample from samples of different classes from the sample, increase, under the condition that a distance between the sample and the near-hit sample in a feature is less than a distance between the sample and the near-miss sample in the same feature, a weight of the feature, reduce, under the condition that the distance between the sample and the near-hit sample in the feature is not less than the distance between the sample and the near-miss sample in the same feature, the weight of the feature, perform training on the feature preset times, and obtain an average weight of the feature after the preset times of training;

add a feature with an average weight greater than a preset value into a feature set, and output the feature set; and

receive an analysis result of the feature set, and add the analysis result of the feature set to the fault diagnosis rule base.

7 . The fault diagnosis method according to claim 1 , wherein the log files comprise at least one of the following:

hard disk state information file, logical disk list information file, controller configuration information file, controller device log information.

8 . The fault diagnosis method according to claim 1 , the method further comprises:

registering the information about the server to be monitored to the monitoring platform.

9 . The fault diagnosis method according to claim 1 , wherein the fault diagnosis rule base comprises pieces of rule information, where each piece of rule information has the following format:

file name;

diagnosis keyword;

fault diagnosis method.

10 . The fault diagnosis method according to claim 1 , the method further comprises:

using the log file including no fault information to update the fault diagnosis rule base continuously.