JP5995265B2 - Information processing system, maintenance method and program - Google Patents

Description

  The present invention relates to an information processing system, a maintenance method, and a program, and more particularly, to an information processing system, a maintenance method, and a program that output instructions related to maintenance work.

  Patent Document 1 discloses a failure management system that provides information including an accurate suspicion rate using an initial FRU table. According to this document, when a failure event is detected by the service processor, the failure management system searches failure history information using information for identifying the failure event as a key, and the failure event in which the failure event has occurred in the past. And the failure history portion having the trigger history that caused the failure event is extracted from the failure history information. Then, when a failure factor part having the incentive history is extracted, the initial suspect ratio corresponding to the failure factor part of the FRU table is corrected according to the cause frequency of the failure factor part. Calculate the corrected suspicion rate. It is described that, when the relationship between the failure event and the failure factor site matches between the FRU table and the failure history information, the incentive frequency corresponding to the failure history information is incremented.

  Further, in Patent Document 2, it is possible to notify an error to an appropriate destination by calculating a recovery difficulty level of an abnormal state that has occurred and transmitting a notification mail to a destination corresponding to the recovery difficulty level. An electronic device is disclosed. According to the document, the electronic device is detected by a state detection unit that detects an abnormal state, a mail generation unit that generates a notification mail having contents according to the abnormal state detected by the state detection unit, and a state detection unit. A destination extracting unit that extracts a destination according to an abnormal state, and a communication control unit that controls communication and retransmission of e-mail between the destination extracted by the destination extracting unit. Yes. Then, the state detection unit calculates the recovery difficulty level of the detected abnormal state, the destination extraction unit extracts a destination to which the notification mail is transmitted based on the calculated recovery difficulty level, and the mail generation unit It is described that a notification mail for the destination extracted by the destination extraction unit is generated, and the communication control unit transmits the notification mail generated by the mail generation unit to the destination extracted by the destination extraction unit. Yes.

JP 2011-175513 A JP 2007-30444 A

  The following analysis is given by the present invention. In recent years, with downsizing of electronic equipment products, distributed and parallel processing technologies such as a high-density mounting server (blade, cluster) technology that connects many commodity processors through a high-speed network are progressing. Structural design for reducing maintenance costs is also an important factor due to the advancement of these technologies, but on the contrary, depending on the system configuration or configuration, higher-density mounting is taken, so that regular maintenance or failure occurs In some cases, the maintenance itself requires advanced training and appropriate experience.

  In addition, even if it does not depend on the above system configuration or form, there may be design defects, initial defects or lot defects (quality variation) in the manufacturing process, deterioration due to the operating environment at the customer site, aging deterioration of parts, etc. If the maintenance response does not follow according to the factors, appropriate maintenance cannot be performed, which may cause prolonged operation stop time and side effects on a soundly operated system, and may cause serious damage to the user. Sometimes.

  As described above, in order to perform appropriate maintenance, training and experience corresponding to the difficulty level of maintenance work are required. However, the failure management system of Patent Document 1 is accurate about the failure factor site when a failure occurs. The main goal is to provide a high suspicion rate, and it is not possible to provide a difficulty level of maintenance work.

  The electronic device disclosed in Patent Document 2 is described to calculate a recovery difficulty level of an abnormal state when an abnormal state is detected. The recovery difficulty level is used to determine a destination of a notification mail that notifies the abnormal state. It is only used for. Further, the restoration difficulty level calculation method itself is also read out from the table (abnormal status area) in which the abnormal status and the restoration difficulty level are associated as shown in FIG. 4, and as described above, the system configuration and configuration In addition, it is impossible to cope with a system in which various elements are intertwined and appropriate maintenance is required.

  The present invention has been made in view of the above-described circumstances, and an object of the present invention is to provide an information processing system, a maintenance method, and a program for supporting correspondence to the above-described various maintenance tasks having different difficulty levels. It is to provide.

According to a first aspect of the present invention , there is provided an information processing system including a replaceable part including at least a processor and a memory, and when an error code is output from the replaceable part, the replaceable part Access to an information source storing information for identifying a failure factor portion of the error history information and failure history information of the failure factor portion, failure factor portion information with a suspect ratio corresponding to the error code , Based on the failure factor analysis unit that acquires the failure history information of the failure factor part, the failure factor part with the suspect ratio, and the failure history information, a maintenance level that indicates a difficulty level of the maintenance work is calculated. And an information processing system that outputs maintenance work instruction information including the maintenance level.

According to a second aspect of the present invention, there is provided a maintenance method by an information processing system including a replaceable part including at least a processor and a memory, and when an error code is output from the replaceable part, the replacement and information for specifying the cause of the error portion of the possible sites, to access the information source of the history information stored in the failure of the fault-site disorder with suspect rate corresponding to the error code and cause part information, acquiring the history information of the failure of the disorder factors site, based on the history information of the suspected ratio with a fault-site with the disorder, a maintenance level indicating a degree of difficulty of maintenance There is provided a maintenance method including a step of calculating and a step of outputting maintenance work instruction information including the maintenance level. This method is associated with a specific machine called an information processing system that outputs maintenance work instruction information including a maintenance level to maintenance personnel.

According to a third aspect of the present invention, when an error code is output from the replaceable part to a computer included in an information processing system including a replaceable part including at least a processor and a memory , the suspect ratio The degree of difficulty of maintenance work is indicated based on the process of obtaining the failure factor part information with the fault, the history information of the fault of the fault factor part, and the fault factor part with the suspect ratio and the fault history information. There is provided a program for executing a process for calculating a maintenance level and a process for outputting maintenance work instruction information including the maintenance level. This program can be recorded on a computer-readable storage medium. That is, the present invention can be embodied as a computer program product.

  ADVANTAGE OF THE INVENTION According to this invention, it becomes possible to support suitably the response | compatibility to the various maintenance work from which the difficulty level differs.

It is a figure which shows the structure of one Embodiment of this invention. It is a figure which shows the structure of the information processing system of the 1st Embodiment of this invention. It is a figure which shows the information hold | maintained at the FRU table of the information processing system of the 1st Embodiment of this invention. It is a flowchart showing operation | movement of the information processing system of the 1st Embodiment of this invention. It is a figure showing the flow of maintenance work. This is an example of a server in which a plurality of systems using logical partitions are operated. It is a figure which shows the structural example of a logical partition. It is a figure which shows the influence range at the time of the failure generation in the structure of FIG. It is a figure which shows the circuit structure which performs lighting control of the status display lamp (error / maintenance state) of the information processing system of the 1st Embodiment of this invention. It is a figure which shows an example of the maintenance report means and fault report means which can be mounted. It is a figure for demonstrating the error threshold value change process in the structure of FIG. It is a flowchart showing the determination flow at the time of error detection.

  First, an outline of an embodiment of the present invention will be described with reference to the drawings. Note that the reference numerals of the drawings attached to this summary are attached to the respective elements for convenience as an example for facilitating understanding, and are not intended to limit the present invention to the illustrated embodiment.

  As shown in FIG. 1, the present invention includes a failure factor analysis unit 110, a maintenance level calculation unit 120, and a display unit 130 that outputs maintenance work instruction information including the maintenance level. This can be realized with the configuration provided.

  More specifically, the failure factor analysis unit 110 stores information for identifying a failure factor portion of replaceable portions included in the managed system, and failure history information of the failure factor portion. The information source 100 is accessed, and from the error code generated from the replaceable part included in the managed system, the fault factor part information with the suspect ratio and the fault history information of the fault factor part are generated. .

The maintenance level calculation unit 120 calculates a maintenance level indicating the difficulty level of the maintenance work based on the failure factor site and the failure history information. For example, the maintenance level calculation unit 120 increases the maintenance level when a failure frequently occurs in a certain failure factor part. As a result, maintenance by maintenance personnel with appropriate advanced training and experience, and maintenance work offline rather than online are performed.
[First Embodiment]

  Next, a first embodiment of the present invention will be described in detail with reference to the drawings. FIG. 2 is a block diagram showing the configuration of the first exemplary embodiment of the present invention.

  Referring to FIG. 2, an information processing system 11 connected to the failure information management server 12 is shown. The information processing system 11 includes a main storage device (hereinafter, “MEM”) 21, a plurality of processors (hereinafter, “PROC”) 22, a plurality of node controllers (hereinafter, “NC”) 23, and a plurality of crossbar switches. (Hereinafter referred to as “Xbar”) 24, a plurality of input / output devices (hereinafter referred to as “IO”) 25, a service processor (hereinafter referred to as “SVP”) 26, a data collection unit 40, a failure factor analysis unit 41, A maintenance level calculation unit 42, a console 43, and a configuration information analysis unit 44 are provided.

  In addition, the information processing system 11 includes a field replaceable unit (FRU) table 30, a failure history storage unit A31, and a failure history storage unit B32 as information sources.

  If a failure is detected at any one or more of MEM21, PROC22, NC23, Xbar24, and IO25, an error is reported to SVP 26 via signal line e001.

  Upon receiving the error report, the SVP 26 collects the failure information of the MEM 21, PROC 22, NC 23, Xbar 24, and IO 25 from the service log. Further, the SVP 26 extracts the failure factor part (NAME), its suspicious rate (RATE), etc. from the FRU table 30 using the error indicator flag (EIF) included in the failure information as a key, and registers it in the failure history storage unit A31. To do.

  The FRU table 30 registers a failure factor part (NAME) corresponding to an error indicator flag (EIF) indicating an error code, a suspected ratio (RATE), a manufacturing lot number, a revision number (REV), a vendor ID (VID), and the like. It is a table.

  FIG. 3 is a diagram illustrating an example of the FRU table 30. For example, the information “NAME = N0_NCC, RATE = 100, REV = A0001, VID = 000” is associated as FRU [0] with the EIF “N0_EIF_0”. This indicates that the card “N0_NCC” of the N0 node controller is specified at a rate of 100% as the failure factor portion from the error indicator flag (EIF) with “N0_EIF_0”. Similarly, the port 0 “N0_P0” of the N0 node controller, the port 0 “N1_P0” of the N1 node controller, and the cable A (CABLE_A) are identified from the error indicator flag (EIF) with “N0_EIF_2”. . Further, information on the suspect ratios of 49%, 50%, and 1% is obtained.

  In addition, the FRU table of FIG. 3 can hold the number of errors (Error Count) and the maintenance status flag (MN1) for each EIF. Such information is appropriately updated from the SVP 26.

  The failure history storage unit A31 stores history information of failures detected by the information processing system 11. In addition, by including manufacturing lots, vendor IDs, and the like in the failure history information, it is possible to analyze parts with high failure occurrence frequency based on functions and production lots that have no margin in design margin from the failure occurrence frequency. . The failure history storage unit A31 is provided with a table for managing the error count field for each EIF so that the error occurrence count of each EIF can be grasped.

  The failure history storage unit B32 stores history information of failures detected by other information processing systems provided from the failure information server 12 via the signal line n001. Also, the failure history storage unit B32 is provided with a table for managing the error frequency field for each EIF, so that the error occurrence frequency of each EIF can be grasped.

  When the update of the FRU table 30 and the failure history storage unit A31 by the SVP 26 is completed, the data collection unit 40 collects data of the FRU table 30, the failure history storage unit A31, and the failure history storage unit B32, and the failure factor analysis unit 41 and The data is output to the configuration information analysis unit 44.

  Based on the data output from the data collection unit 40, the failure factor analysis unit 41 analyzes the reported failure history, failure history of other information processing systems, manufacturing lots, etc. Fault factor part information and fault history information of the fault factor part are generated. Further, the failure factor analysis unit 41 may make an inquiry to an external server or the like for information on a part specified as a failure factor part, if necessary.

  The maintenance level calculation unit 42 is based on the failure factor part information with the suspect ratio output from the failure factor analysis part 41, the failure history information of the failure factor part, and the maintenance information obtained via the signal line n002. Then, the maintenance difficulty level is calculated, and the maintenance work instruction information including the maintenance difficulty level information (maintenance level) is output together with the failure factor part information with the suspect ratio. For example, when the maintenance support information includes the number of failures due to maintenance errors (procedure errors) or failures due to maintenance (occurrence of excessive push-in, pull-in, etc.), the maintenance level calculation unit 42 The maintenance level of a part having a lot of failure frequency information due to mistakes during maintenance is calculated as a high maintenance level (high difficulty level). The maintenance level is calculated, for example, by converting a factor of failure, the number of failures, the number of times of mistakes during maintenance, etc. into a score using a predetermined mathematical formula, a threshold for each predetermined level, and this score. Can be obtained by comparing with The threshold value is preferably determined for each failure factor site from the FIT (Failure In Time) value of the component.

  The console 43 outputs maintenance work instruction information including failure factor information with a suspect ratio and maintenance difficulty information (maintenance level) output from the maintenance level calculator 42.

  The configuration information analysis unit 44 analyzes the configuration information from the operation mode based on the data output from the data collection unit 40 and transmits the configuration information to the SVP 26. This analysis result includes, for example, operational availability information due to secondary effects associated with maintenance operations (cover removal, cable movement / removal, module replacement) at the failure factor site. If it is determined that the operation is possible as a result of the analysis, the SVP 26 shifts the MEM 21, PROC 22, NC 23, Xbar 24, and IO 25 to the maintenance mode, and changes the error count during the maintenance mode.

  The failure information management server 12 is connected to other information processing systems including the information processing system 11 via signal lines n003 and n004, and is a server that collects failure information. More specifically, the failure information management server 12 is operated by a failure information database (failure information DB) 35 that stores the collected failure information and other information processing systems including the information processing system 11 as maintenance support information. And a maintenance information database (maintenance information DB) 36 for storing information on the maintenance work. Information stored in the failure information DB 35 is transferred to the failure history storage unit B32 of the information processing system 11 through the signal line n001 at a predetermined timing.

  Note that the data collection unit 40, failure factor analysis unit 41, maintenance level calculation unit 42, and configuration information analysis unit 44 of the information processing system 11 shown in FIG. It can also be realized by a computer program that executes the above-described processes using the hardware.

Next, the operation of this embodiment will be described in detail with reference to the drawings. In the information processing system 11, the following failures may occur.
・ Between MEM21 and PROC22 ・ Between PROC22 and NC23 ・ Between NC23 and Xbar24 ・ Between Xbar24 and IO25 ・ Between SVP26 and MEM21, between PROC22, NC23, Xbar24 and IO25 In the description, an operation when a failure occurs between the Xbar 24 and the IO 25 connecting the plurality of nodes of the information processing system 11 will be described.

  FIG. 4 is a flowchart showing an operation when an error is detected in either Xbar 24 or IO 25 in the information processing system according to the first embodiment of this invention. Referring to FIG. 4, first, when an error is detected in either Xbar 24 or IO25 due to a failure between Xbar24 and IO25 (step S001), an error is reported to SVP 26 (step S002).

  Upon receiving the error report, the SVP 26 collects the failure information of the Xbar 24 and IO 25 from the service log (SVP log) (step S003).

  Next, the SVP 26 searches for corresponding data from the FRU table 30 using the error indicator (EIF) included in the collected failure information as a key (step S004). Next, the SVP 26 stores the data retrieved from the FRU table 30 in the failure history storage unit A31 and activates the data collection unit 40. Here, the SVP 26 adds 1 to the value of the error count field of the corresponding entry in the FRU table 30.

  The data collection unit 40 first confirms whether or not the data corresponding to the error indicator (EIF) specified in step S003 has been registered in the failure history storage unit A31 before the registration in step S004 (step S004). S005).

  Here, when the data corresponding to the error indicator (EIF) identified in step S003 is registered in the failure history storage unit A31 (Yes in step S005), the data collection unit 40 determines that the error identified in step S003. It is confirmed whether or not the data corresponding to the indicator (EIF) has been registered in the failure history storage unit B32 (step S006-1).

  Similarly, when the data corresponding to the error indicator (EIF) specified in step S003 is not registered in the failure history storage unit A31 (No in step S005), the data collection unit 40 similarly specifies in step S003. It is confirmed whether or not the data corresponding to the error indicator (EIF) is registered in the failure history storage unit B32 (step S006-2).

  Depending on the results of steps S005, S006-1, and S006-2, any one of steps S007 to S010 is performed. If the same failure history exists in either or both of the failure history storage unit A31 and the failure history storage unit B32, the data collection unit 40 determines the number of error occurrences in the table managing each error number field. Add 1 to the field value.

  First, when the same failure history exists in both the failure history storage unit A31 and the failure history storage unit B32 (both steps S005 and S006-1 are Yes), the data collection unit 40 determines that both of the failure histories Is output to the failure factor analysis unit 41. The failure factor analysis unit 41 compares and analyzes conditions such as the production lot and the vendor ID for both of the failure history information, and determines the necessity of correcting the failure factor part and the suspect ratio of the failure factor part (step S007). .

  On the other hand, if the same failure history exists in the failure history storage unit A31, but the same failure history does not exist in the failure history storage unit B32 (Yes in step S005, No in S006-1), The data collection unit 40 outputs the failure history stored in the failure history storage unit A31 to the failure factor analysis unit 41. The failure factor analysis unit 41 compares and analyzes conditions such as the manufacturing lot and the vendor ID for the failure history information in the failure history storage unit A31, and determines the necessity of correcting the failure factor part and the suspect ratio of the failure factor part ( Step S008).

  On the other hand, when the same failure history does not exist in the failure history storage unit A31, but the same failure history exists in the failure history storage unit B32 (No in step S005, Yes in S006-2), The data collection unit 40 outputs the failure history in the failure history storage unit B32 to the failure factor analysis unit 41. The failure factor analysis unit 41 compares and analyzes the conditions such as the manufacturing lot and the vendor ID for the failure history information in the failure history storage unit B32, and determines the necessity of correcting the failure factor part and the suspect ratio of the failure factor part ( Step S009).

  On the other hand, when the same failure history does not exist in both the failure history storage unit A31 and the failure history storage unit B32 (both No in steps S005 and S006-1), the data collection unit 40 stores the data in the FRU table. It transmits as it is (step S010). The failure factor analysis unit 41 outputs the failure factor site and the suspected ratio of the failure factor site using the data in the FRU table (step S010).

  In addition, the correction method of the suspicious ratio in the above-described steps S007 to S009 is described in detail in Patent Document 1.

  Next, the maintenance level calculation unit 42 calculates the maintenance difficulty level based on the information obtained in steps S007 to S010 and the maintenance support information, and performs maintenance together with the failure factor part information with the suspect ratio and the maintenance support information. Maintenance work instruction information including the difficulty level information (maintenance level) is generated and output (step S011). Here, if the maintenance level is high (difficulty level) as a result of the calculation of the maintenance level, or if a secondary effect of maintenance on other devices is predicted, the maintenance level calculation unit 42 does not perform online maintenance. An instruction to do so is generated.

  Finally, the maintenance instruction output from the maintenance level calculation unit 42 is displayed on the console 43 (step S012). In accordance with the maintenance instruction, for example, the maintenance staff creates a maintenance schedule for switching to maintenance after operation stop (offline maintenance), and starts maintenance work.

  Next, a series of maintenance work will be described. FIG. 5 is a diagram showing the flow of maintenance work. Hereinafter, it is assumed that the information processing system 11 of this embodiment is a server in which a plurality of systems using logical partitions as shown in FIG. 6 are operated. Also, the NC 23, Xbar 24, and IO 25 are connected as shown in FIG. 7, and partition 0 (PAR0) and partition 1 (PAR1) are configured, and the first operating system (OS_0) and second In the following description, it is assumed that the operating system (OS_1) is assigned and operated.

  Here, by detecting a failure between Xbar1 and IO1 in FIG. 8, the information processing system 11 instructs a maintenance operation lock temporarily prohibiting the maintenance operation, and detects a failure by a blinking operation of a failure state display lamp or the like. Display. FIG. 9 shows an error status display lamp (EF display) 48 and a maintenance status display lamp (MF display) 47 provided in the information processing system 11 (in FIG. 11, only the devices A11A and B11B in the information processing system 11 are shown). It is a figure which shows the circuit structure which performs lighting control of. Here, when a failure (EF) is detected in any of the devices 11A and 11B included in the information processing system 11, the EF control unit 46 causes the error state display lamp (EF display) 48 to blink. Thereby, it is possible to make a maintenance person or the like recognize the occurrence of a failure other than the console.

  Next, the information processing system 11 determines whether or not the failure is a failure that can be corrected by the automatic correction function of the system (step S101). If it is determined that the fault can be corrected (YES in step S101), the information processing system 11 performs an automatic correction process and turns off the error status display lamp (EF display) 48 (step S102).

  If it is determined that the failure is not correctable (NO in step S101), it is next determined whether or not the data can be retransmitted. Here, if it is determined that the corresponding data is a failure that cannot be retransmitted (NO in step S103), it is determined that the failure is a non-recoverable failure, and a process of closing between Xbar1 and IO1 is performed.

  If it is determined that the data can be retransmitted (YES in step S103), it is determined whether or not to shift to the maintenance mode (step S104). Here, if there is no other system in operation and there is no influence of a secondary system failure due to a maintenance operation error, etc., it is determined that the transition to the maintenance mode is unnecessary, and maintenance is performed based on the maintenance instruction. Is performed (NO in step S104). Specifically, if the number of failures (number of errors) detected this time is less than a predetermined threshold (NO in step S105), the operation is continued (step S106). Judgment is performed and processing between Xbar1 and IO1 is closed. In the example of FIG. 5, an error determination is made as to whether or not a preventive maintenance notification (step S108) is to be output (step S107) before closing between Xbar1 and IO1.

  On the other hand, as shown in FIG. 8, when the first operating system (OS_0) and the second operating system (OS_1) are in operation, the process transits to the maintenance mode in step S104. In this case, the information processing system 11 lights the maintenance status display lamp (MF display) 47 shown in FIG.

  The maintenance staff who has recognized that the maintenance mode has been switched to by the lighting of the maintenance status display lamp 47 or the display of the console 43 is comprehensively determined from the failure history, maintenance information, maintenance level (difficulty level), etc. displayed on the console 43. It is determined whether or not maintenance during operation (online maintenance) is to be performed (step S109).

  Here, for example, when the maintenance level is high (difficulty level is high), the side effects on the other devices due to maintenance are foreseen, so the maintenance staff does not perform online maintenance but leads to system down by consulting with the customer. It is possible to change to a maintenance schedule that eliminates such secondary effects (interruption of the flowchart in FIG. 5).

  On the other hand, if the maintenance level is not high (difficulty level: medium to medium) and it is determined that online maintenance is possible, the maintenance staff releases the maintenance lock instruction and starts maintenance work.

  First, the maintenance staff performs a threshold value changing process prior to replacement of the failure factor site IO1 (step S110). FIG. 10 is a diagram showing a location where the error threshold is changed when the failure factor site IO1 is replaced. In the example of FIG. 10, the error count threshold of Xbar0-I0, Xbar0-I1, Xbar1-I0, and Xbar1-I1 is changed from the SVP 26 in order to minimize the influence on the logical partition due to the replacement of IO1. ing. Specifically, even if an error occurs during replacement, the error count threshold is temporarily raised or the error count is disabled so that system disconnection and preventive maintenance notification output are suppressed. Measures such as

  As a result, as shown in FIG. 11, parameters such as the threshold value are adjusted in the maintenance mode (S202, S204, S206), so even if a failure is detected at a location related to the replacement site (FIG. 11). S203, S205, S207), and error determinations 1-3 (S208-S210), a negative determination is made. In the error analysis (S211) that follows, the analysis is performed based on the fact that it is the maintenance mode, the configuration information, and these error determination results, and whether or not the maintenance is interrupted or whether or not an error is displayed. Is determined. Further, these results are accumulated in the maintenance information DB 36.

  After the threshold value change, if the number of faults detected this time (number of errors) is less than the threshold value after the change (NO in step S111), the operation is continued with the maintenance mode maintained (step S112). As a result, the side effects of maintenance on other systems are minimized.

  On the other hand, if the number of faults detected this time (number of errors) exceeds the threshold after the change, error determination is performed in the flow shown in FIG. 11 (step S113). As a result of the error determination, if it is determined that NG (maintenance is interrupted), an error determination notification is output to the console 43 or the like (step S116), it is determined that the failure is unrecoverable, and a process of closing Xbar1-IO1 is performed. In this case, the maintenance status display lamp (MF display) 47 and the error status display lamp (EF display) 48 shown in FIG. At that time, the number of lighting of the error state display lamp (EF display) 48 may be controlled in accordance with the number of error locations, the number of errors, and the like.

  On the other hand, if it is determined that the result of the error determination is OK (maintenance can be continued), the operation is continued with the maintenance mode maintained (step S114). As a result, the side effects of maintenance on other systems are minimized.

  After the above process, when the maintenance is completed, the error count threshold shown in FIG. 10 is returned again, the maintenance mode is canceled, and the normal operation is resumed.

  As described above, according to the present embodiment, maintenance work instruction information including a maintenance level is output, so that maintenance personnel can be appropriately maintained. In addition, even when maintenance is not performed at a high level, it is possible to reduce the influence on other operating systems by appropriately raising the threshold value of the number of errors as in the above-described maintenance work flow. . As a result, the mean recovery time (MTTR), which has become a problem in recent years, can be shortened to reduce the burden on maintenance personnel, and the influence on customers can be minimized.

  By reflecting the failure history and maintenance history information when the configuration is changed due to system degeneration or expansion, it is possible to avoid the site where the failure is as high as possible and support the configuration and maintenance of the system. As a result, the influence of the mean time between failures (MTBF) can be minimized, and as a result, the operating rate of the system can be improved and the reliability of the entire system can be improved.

  Although the embodiments of the present invention have been described above, the present invention is not limited to the above-described embodiments, and further modifications, substitutions, and adjustments may be made without departing from the basic technical idea of the present invention. Can be added. For example, in the above-described embodiment, an example of maintenance at the time of detecting a failure has been described, but the present invention can be similarly applied to parts replacement by periodic inspection and work by predictive maintenance.

  Further, in the above-described embodiment, it has been described that the secondary influence by the maintenance work is reduced by changing the threshold value of the number of errors, but the operation mode is temporarily changed to lower the data transfer rate, etc. Thus, the side effects of maintenance may be minimized.

In the above-described embodiment, the maintenance information DB 36 is described as storing information on maintenance work performed in other information processing systems including the information processing system 11, but the following information is recorded. It is also desirable to keep it.
-Video information of maintenance and inspection operations by the maintenance recorder These video information are tagged with tags such as faulty parts and error codes so that related video information can be indexed and referenced from faulty parts when a fault occurs. preferable. Further, the video information may be collected from a fixed-point WEB camera, a small camera worn by an experienced maintenance worker (for example, a small camera attached to glasses), or the like. In addition, these WEB cameras and small cameras are preferably provided with means for transmitting and storing information from the start to completion of maintenance or inspection to the failure information management server 12. Further, a remote maintenance staff may be able to view the video information in real time. Of course, viewing control based on the security level and the access policy of maintenance personnel is performed.

  In the above-described embodiment, the information processing system 11 has been described as operating alone. However, each time a failure report is received from a plurality of information processing systems and a failure report is received from another information processing system, a maintenance level is provided. May be recalculated and presented to maintenance personnel. For example, when a failure report for a certain part is received from another information processing system, the information processing system 11 can determine the presence or absence of the same part or a related part with reference to past fault history information.

  In the above-described embodiment, the parameters for calculating the maintenance level are described as being registered in advance. However, these parameters may be appropriately inspected and corrected. For example, more precise maintenance instructions can be output by using information on the type of maintenance actually performed (online, offline) and server type (rack mount, blade) information. In addition, although maintenance is possible online, it may be determined whether or not there is a possibility of affecting a device that is not subject to maintenance (not subject to management) when the maintenance is performed. Depending on the result, for example, if there is a risk of a fatal failure that causes the system to go down due to the influence of being out of maintenance (out of management), it is possible to perform operations such as inquiring customers and establishing a maintenance schedule Become.

  It should be noted that the disclosures of the above patent documents are incorporated herein by reference. Within the scope of the entire disclosure (including claims) of the present invention, the embodiments and examples can be changed and adjusted based on the basic technical concept. Further, various combinations or selections of various disclosed elements (including each element of each claim, each element of each embodiment or example, each element of each drawing, etc.) within the scope of the claims of the present invention. Is possible. That is, the present invention of course includes various variations and modifications that could be made by those skilled in the art according to the entire disclosure including the claims and the technical idea.

DESCRIPTION OF SYMBOLS 11 Information processing system 11A, 11B apparatus 12 Fault information management server 21 Main memory (MEM)
22 Processor (PROC)
23 Node controller (NC)
24 Crossbar switch (Xbar)
25 Input / output unit (IO)
26 Service Processor (SVP)
30 FRU (Field Replaceable Unit) table 31 Failure history storage unit A
32 Fault history storage B
35 Failure information database (failure information DB)
36 Maintenance Information Database (Maintenance Information DB)
40 Data Collection Unit 41 Failure Factor Analysis Unit 42 Maintenance Level Calculation Unit 43 Console 44 Configuration Information Analysis Unit 46 EF Control Unit 47 Maintenance Status Display Lamp (MF Display)
48 Error status indicator lamp (EF display)
DESCRIPTION OF SYMBOLS 100 Information source 110 Failure factor analysis part 120 Maintenance level calculation part 130 Display part e001, n001, n002, n003, n004 Signal line

Claims (9)

An information processing system having a replaceable part including at least a processor and a memory,
When an error code is output from the replaceable part , access to an information source storing information for identifying a fault factor part of the replaceable part and fault history information of the fault factor part Then, a failure factor analysis unit that acquires failure factor part information with a suspicion ratio corresponding to the error code , and failure history information of the failure factor part,
A maintenance level calculation unit that calculates a maintenance level indicating a difficulty level of maintenance work based on the failure factor part with the suspect ratio and the history information of the failure, and
An information processing system for outputting maintenance work instruction information including the maintenance level. The information source includes a maintenance information database storing maintenance support information related to each failure factor site,
The information processing system according to claim 1, wherein the maintenance work instruction information including the maintenance support information is output. The maintenance information database further includes failure information due to a maintenance error that occurred in each failure factor site,
The information processing system according to claim 2, wherein the maintenance level calculation unit calculates a maintenance level by using failure information due to a mistake at the time of maintenance in addition to the failure factor part and the failure history information. If the maintenance level is below a predetermined level, instruct online maintenance work;
The information processing system according to any one of claims 1 to 3, wherein when the maintenance level exceeds the predetermined level, an offline maintenance work is instructed.   5. The information processing system according to claim 1, further comprising a maintenance operation lock function that prohibits online maintenance operation until a predetermined condition is satisfied after the maintenance work instruction information including the maintenance level is output. In addition, based on the configuration information of the system to be managed, a configuration information analysis unit that determines the presence or absence of another operating system affected by the maintenance work or another logical partition in operation, and notifies the service processor,
6. The information processing system according to claim 1, wherein the service processor controls another operation system affected by the maintenance work or another logical partition in operation to suppress failure detection by the maintenance work. . The information source includes moving image data recording a video of each maintenance operation,
The information processing system according to claim 1, wherein the moving image data is reproduced in response to a request from a maintenance staff. A maintenance method by an information processing system having at least a replaceable part including a processor and a memory ,
If an error code from the substitutable positions is output, information source which stores information for specifying the cause of the error portion of said substitutable positions, and history information of the failure of the fault-site And acquiring failure factor part information with a suspicion rate corresponding to the error code , and failure history information of the failure factor part, and
Calculating a maintenance level indicating a difficulty level of maintenance work based on the failure factor portion with the suspect ratio and the history information of the failure ;
Outputting maintenance instruction information including the maintenance level. In a computer included in an information processing system including at least a processor and a replaceable part including a memory ,
When an error code is output from the replaceable part, a process of acquiring the failure factor part information with the suspect ratio and the failure history information of the failure factor part;
A process of calculating a maintenance level indicating a difficulty level of maintenance work based on the failure factor part with the suspect ratio and the history information of the failure ;
A program for executing a process of outputting maintenance work instruction information including the maintenance level.

Images